Poštnina placana pri pošti 4102 Kranj Journal of Management, Informatics and Human Resources ISSN 1318-5454 Volume 54, Issue 2, May 2021 Revija za management, informatiko in kadre Organizacija (Journal of Management, Informatics and Human Resources) is an interdisciplinary peer-reviewed journal which is open to contributions of high quality, from any perspective relevant to the organizational phenomena. The journal is designed to encourage interest in all matters relating to organizational sciences and is intended to ap­peal to both the academic and professional community. In particular, journal publishes original articles that advance the empirical, theoretical, and methodological understand­ing of the theories and concepts of management and or­ganization. The journal welcomes contributions from other scientific disciplines that encourage new conceptualiza­tions in organizational theory and management practice. We welcome different perspectives of analysis, including the organizations of various sizes and from various branch­es, units that constitute organizations, and the networks in which organizations are embedded. Topics are drawn, but not limited to the following areas: • organizational theory, management, development, and organizational behaviour; • human resources management (such as organization & employee development, leadership, value creation through HRM, workplace phenomena etc.); • managerial and entrepreneurial aspects of education; • business information systems (such as digital business, decision support systems, business analytics etc.); • enterprise engineering (e.g., organizational design, business process management, enterprise transformation paradigms etc.); • papers that analyse and seek to improve organizational performance. Organizacija (Revija za management, informatiko in cloveške vire) je interdisciplinarna recenzirana revija, ki objavlja visoko kakovostne prispevke z vseh vidikov, ki so pomembni za organizacijske procese in strukture. Revija je zasnovana tako, da spodbuja zanimanje za razlicne vidike v zvezi z organizacijskimi vedami in je namenjena tako akademski kot strokovni skupnosti. Revija objavlja izvirne clanke, ki spodbujajo empiricno, teoreticno in metodološko razumevanje teorij in konceptov managementa in organizacije. Pozdravljamo tudi prispevke iz drugih znanstvenih disciplin, ki spodbujajo nove koncepte v organizacijski teoriji in praksi. Objavljamo clanke, ki analizirajo organiziranost z razlicnih vidikov, so usmerjeni na organizacije razlicnih velikosti in iz razlicnih sektorjev, na enote, ki sestavljajo organizacije, in na mreže, v katere so organizacije vpete. Teme so pokrivajo predvsem naslednja podrocja: • organizacijska teorija, upravljanje, razvoj in organizacijsko vedenje; • management cloveških virov (kot so organizacija in razvoj zaposlenih, vodenje, ustvarjanje vrednosti s pomocjo cloveških virov, organizacijski pojavi na delovnem mestu itd.); • vodstveni in podjetniški vidiki izobraževanja; • poslovni informacijski sistemi (kot so digitalno poslovanje, sistemi za podporo odlocanju, poslovna analitika itd.); • podjetniški inženiring (npr. organizacijsko oblikovanje, upravljanje poslovnih procesov, paradigme preoblikovanja podjetij itd.); • clanki, ki analizirajo organizacijsko uspešnost in prizadevanja za izboljšanje le-te. Jin-Nan WU, Mengmeng SONG, Joseph C. UGRIN, Lin LIU, Tingting ZHU Mehmet ÇETIN, Gulmira SAMENOVA, Filiz TÜRKKAN, Ceylan KARATAS Phuong TRAN HUY, Hong Chuong PHAM Erik RUŽIC, Dragan BENAZIC Anton IVASCHENKO, Alfiya R. DIYAZITDINOVA, Tatiyana NIKIFOROVA Cyberloafing Research 1997-2019: A Citation-based Literature Review The Role of Daily Affect in Leader-Member Exchange: A Multilevel Investigation in Public Health Administration Effect of Management Commitment to Service Quality on Car-hailing Drivers’ Service Behaviors: The Case of GrabCar in Vietnam The Impact of Internal Knowledge Sharing on Sales Department’s Innovativeness and New Product Commercialization Optimisation of the Rational Proportion of Intelligent Technologies Application in Service Organisations RESEARCH PAPERS 98 112 131 147 162 EDITOR / UREDNIK Jože Zupancic University of Maribor, Faculty of Organizational Sciencies, Slovenia CO-EDITORS / SOUREDNIKI Petr Doucek Prague University of Economics, Faculty of Informatics and Statistics, Czech Republic Matjaž Maletic University of Maribor, Faculty of Organizational Sciencies, Slovenia Wlodzimierz Sroka WSB University, Department of Management, Dabrowa Górnicza, Poland EDITORIAL BOARD / UREDNIŠKI ODBOR REVIJE Hossein Arsham, University of Baltimore, USA Franc Cuš, University of Maribor, Slovenia Sasha M. Dekleva DePaul University, School of Accountancy and MIS, Chichago, USA Vlado Dimovski, University of Ljubljana, Slovenia Daniel C. Ganster, Colorado State University, USA Jože Gricar, University of Maribor, Slovenia Werner Jammernegg Viena University of Economics and Business Administration, Austria Marius Alexander Janson, University of Missouri-St. Louis, USA Stefan Klein, University of Münster, Germany Aleksandar Markovic, University of Belgrade, Serbia Hermann Maurer, Technical University Graz, Austria Matjaž Mulej, University of Maribor, Slovenia Valentinas Navickas, Kaunas University of Technology, Lithuania Ota Novotny, University of Economics, Prague, Czech Republic Milan Pagon, Independent University, Bangladesh (IUB), Dhaka, Bangladesh Björn Paape, RWTH-Technical University Aachen, Germany Matjaž Perc University of Maribor, Slovenia Dušan Petrac, NASA, Jet Propulsion Laboratory, California Institute of Technology, USA Nataša Petrovic University of Belgrade, Serbia Tetyana Pimonenko, Sumy State University, Balatsky Academic and Scientific Institute of Finance, Economics and Management, Ukraine Hans Puxbaum, Vienna University of Technology, Austria Vladislav Rajkovic, University of Maribor, Slovenia Gábor Rekettye, University of Pécs, Hungary Henk G. Sol, Faculy of Economics and Business, University of Groningen, Netherlands Eugene Semenkin Reshetnev Siberian State University of Science and Technology, Krasnoyarsk, Russian Federation Velimir Srica, University of Zagreb, Croatia Paula Swatman, University of Tasmania, Australia Brian Timney, The University of Western Ontario, Canada Maurice Yolles, Liverpool John Moores University, UK Douglas R. Vogel, Harbin Institute of Technology-HIT, School of Management, China Gerhard Wilhelm Weber, Middle East Technical University, Turkey Stanislaw Wrycza, University of Gdansk, Poland Anna Lucyna Wziatek-Stasko, Jagiellonian University in Kraków, Poland Yvonne Ziegler, Frankfurt University of Applied Sciences, Germany Hans-Dieter Zimmermann, FSH St. Gallen University of Applied Sciences, Switzerland DOI: 10.2478/orga-2021-0007 Cyberloafing Research 1997-2019: A Citation-based Literature Review Jin-Nan WU1., Mengmeng SONG1., Joseph C. UGRIN2, Lin LIU3, Tingting ZHU1 1 Anhui University of Technology, School of Business, Ma’anshan, Anhui, China; wujinnanseu@aliyun.com (JW), 1500253842@qq.com (MS), 772142722@qq.com (TZ). 2 University of Northern Iowa, College of Business Administration, Cedar Falls; USA, joseph.ugrin@uni.edu (corresponding author) 3 Anhui University of Technology, School of Management Science and Engineering, Ma’anshan Anhui, China; liulinahut@aliyun.com Background and purpose: The purpose of this study is to illustrate the evolution of cyberloafing research and iden­tify the important papers in the development of cyberloafing knowledge. Methodology: We identify a total of 116 research papers from the Web of Science (WOS) database that were pub­lished in top journals during 1997-2019. We performed citation analysis, co-citation analysis, and main path analysis to identify the most influential authors, countries, and institutions in the development of cyberloafing research, and graphically map the relationships between them. Results: Cyberloafing research has progressed through three stages which we coin the emergent stage, the explo­ration stage, and the burgeoning stage. We also identify the relationships between articles and identify the distinct and significant knowledge streams, 16 influential papers that are seminal or highly sighted, and we identify the au­thors, institutions, and countries from which the research emerged. Researchers in the United States have produced the most research, followed by China, the United Kingdom, and Australia. The National University of Singapore produced the most research in our sample, and Vivian K. Lim of the National University of Singapore was the most influential author, anchored by her seminal piece on cyberloafing and organizational justice published in 2002. Conclusion: This study is, to our best knowledge, the first to examine the development of cyberloafing research using evidence-based analysis methods. It depicts the development of the cyberloafing literature and identifies signif­icant streams that researchers can explore in the future. We conclude that future research should incorporate more direct measures of cyberloafing, examine individual consequences caused by too much cyberloafing, such as mental health. Keywords: Cyberloafing, Citation analysis, Co-citation analysis, Main path analysis, Visualization 1 Introduction The Internet has profoundly influenced and changed the way people live, work, and learn. It has brought about con­venience and a new form of communication. In the work­place, it has also created new opportunities for employees to engage in non-work-related activities and sometimes non-productive activities. The concept of “cyberloafing” was first proposed in the mid-1990s (Kamins, 1995) and is regarded as a type of workplace deviance and asset mis­use. However, at that time, the digitization of enterprises in countries around the world, especially in developing countries, was not widespread, and employees’ cyberloaf­ing behaviors and the potential harm had not yet attracted the attention of researchers. Companies around the world gradually increased their investment in Internet resources in the early 2000s, and as the technology became more widespread in business enterprises, so did the potential for misuse and abuse. The problems caused by employees’ misuse and abuse of the Internet started to attract the atten­tion of academia soon after. Lim (2002) defined cyberloaf­ing. She pointed out that cyberloafing involves employees who engage in non-work-related online activities during working hours, such as receiving or sending private emails or browsing the news. Lim (2002) developed a cyberloaf­ing scale, which laid the foundation for future research in the field. Scholars have subsequently refined the definition and measurement of cyberloafing behavior (Anandarajan 2002; Blanchard and Henle 2008), identified antecedents at the individual (Blanchard and Henle 2008; Jia et al. 2013; O’Neill et al. 2014; Vitak et al. 2011 ) and organiza­tional levels (Andreassen et al. 2014; Garrett and Danziger 2008; Henle and Blanchard 2008; Lim 2002), identified and explored potential outcomes (Coker 2013; Heflin et al. 2017; Meijman and Mulder 1998; Phillips 2006; Stoddart 2016; Syrek et al. 2017; Wu et al. 2018), and explored the effects at mechanisms aimed at controlling and deterring cyberloafing behavior (Ugrin and Pearson, 2013). Research on cyberloafing has been conducted by scholars from all over the world, including the United States, China, Europe, and the Middle East, suggesting it is a global workplace phenomenon. There is still much to learn about cyberloafing, but researchers have made sub­stantial progress over the last twenty years. Recently, Mer­cado et al. (2017) performed a meta-analysis to investigate the nomological network of cyberloafing. The results show the nature and valence of these non-work related cyber­loafing behaviors. Their findings provide a synopsis of what we know about the antecedents of cyberloafing and provide a starting place for empirically grounded research in the future. However, to the best of our knowledge, no study has been conducted to summarize the influential au­thors, countries, and institutions engaged in cyberloafing research, or to identify the evolutionary path and important papers that have defined this stream of research over the past two decades. One reason for examining the evolution of the literature is the fact that the number of scientific pa­pers on cyberloafing has become large and is continuously growing. That makes it an optimal topic for bibliometric analysis. Bibliometric analysis is a quantitative technique that can provide a macroscopic overview of large bodies of academic literature. It can be used to map the characteris­tics and the development of scientific output within a spe­cific field of research and reveal the significant knowledge flows in the citation network (Jia et al., 2018). This study aims to narrow the gaps in what we already know and define the pathway for future researchers to ex­tend the body of knowledge by using bibliometric analy­sis. Specifically, social network analysis and visualization software such as CiteSpace, HistCite, and Pajek, were applied to perform social network and visualization anal­yses on the cooperation among authors, institutions, and countries in the field, and to quantitatively analyze the lit­erature citation network and the main path between author connections in the field. As a result, this study identifies the influential authors, institutions, and countries of ori­gin, and their degree of collaboration. It also reveals the path through which or knowledge about cyberloafing has evolved, which is useful from a historical perspective and from the perspective of researchers looking to extend or branch off from the existing streams. In the end, this study provides scholars with an understanding of the research progress on cyberloafing, and it identifies directions for future research. 2 Article Selection and Methodology We conducted three literature searches on cyberloafing using information from the Web of Science (WOS). The SCI-EXPANDED (SCI) and SSCI databases served as the main data source to identify relevant articles for analysis. We used a number of subjects and titles in the three lit­erature searches to avoid missing important articles due to the different names used for cyberloafing in the jour­nals. Names for cyberloafing include cyberslacking, work­place internet deviance, and employee internet addiction, amongst other things. As a result, searches included sub­jects such as “cyberloaf*”, “cyberslack*”, “personal social media us*”, and the titles were set included “personal in­ternet us*”, “personal web us*”, “internet deviance”, “Use in the workplace”, “abuse in the workplace”, “employee internet abuse”, “employee internet addiction” et cetera. Considering that the concept of cyberloafing in the WOS database first appeared in 1997, we set the period for the three literature searches to 1997-2019 (The date of data collection is March 28, 2019). 76, 132, and 42 articles were collected from each literature search, respectively. We removed duplicate articles through manual screening and cross-validation, retaining 116 high-quality articles from the SCI and SSCI for analysis. These articles are written by 255 authors, include 333 keywords, and 4487 references. They covered a total of 18 disciplines. Among them, psychology, business and economics, computer science, information science, library and information science, and telecommunication are the most common disciplines. Top journals such as Computers in Human Behavior, Cyberpsychology Behavior, Informa­tion & Management, Behaviour & Information Technolo­gy, Communications of the ACM, Cyberpsychology Be­havior and Social Networking, Internet Research, Journal of Management Information Systems, Journal of Applied Psychology, and the Journal of Business Ethics published the most articles on cyberloafing. We applied CiteSpace software to conduct visual and social network analysis on the authors, institutions, and countries associated with the 116 articles we identified. We used the HistCite citation analysis software to identify local citation scores (LCS), the number of times an article is cited in the current database. Also, we prepared a chro­nology of citations using a citation network analysis on the top 30 LCS articles. Finally, we used Pajek software, a sophisticated network analysis tool, to analyze the signifi­cant articles in the past 20 years and identify the main path in the relationships between them. We discuss this further below. 3 Visualization Analysis 3.1 Publication distribution Through analyzing the chronological distribution of cyberloafing articles published from 1997 to 2019, and drawing a trend line on the volume of publications per given year, we can visualize the growth and development of the literature stream. The chronological distribution and the trend line showing the growth in the volume of publi­cations is shown in Figure 1. According to the quadratic polynomial trend line, it can be seen that the trend line has an upward trend with the growth of the year, which shows that there has been rapid growth in the volume of research, and the growth continues to accelerate. The year 2002 was a turning point where the cyberloafing research left the emergent stage (average of one paper per year) and moved into the exploration stage. From 2002 to 2013, the annual number of publications grew and held steady at approxi­mately four articles per year. After 2014, the average an­nual number of publications grew to nearly 12 significant articles per year, and we call this the burgeoning stage. 3.2 Author visualization A total of 255 authors produced the 116 publications in our analysis, of which 226 (88.63%) authors published just one paper. We used Price Law to determine the most pro­ductive authors (Price 1963). Price Law is a mathematical formula developed to analyze scientific growth and com­monly used to determine the amount of contribution indi­viduals make to a body of work. According to Price Law, the findings suggest that authors publishing two or more papers are productive in this field (Mp=0.749*v6˜1.83). There are 29 productive authors in the cyberloafing field, and the top five most productive authors are Pablo Zogh­bi-Manrique-de-Lara (6 articles), Anandarajan (5 articles), Lim (4 articles), and Simmers (3 articles). To further demonstrate the characteristics of the au­thor’s cooperation network, CiteSpace was used to per­form a co-citation analysis according to the name of the author. The thresholds of C, CC, CCV in the first, middle and last three time periods are set to (1, 2, 20), (2, 3, 20), (2, 3, 20) to extract the data, that is, the beginning, middle and end of the data are assigned according to C, CC and CCV, where C represents the lowest citation or occurrence frequency, CC represents the co-occurrence or co-citation frequency of the time slice, and CCV represents the collin­ear rate or co-citation rate (Li and Chen, 2017). The time slice was set to 1 year, and a map of the knowledge domain of the scientific cooperation network was drawn (Figure 2). The node in the figure 2 is the author’s name, the size of the node represents the number of papers published by the author, the line reflects the strength of the cooperation relationship, and the color of the line represents the time of the first cooperation. The scientific cooperation network in this field has 57 nodes and 40 linkages. The network density is 0.0251, indicating a relatively scattered aca­demic cooperation network. The results show that Lim, Pablo Zoghbi-Manrique-de-Lara, Anandarajan, O’Neill, and Bock have the most significant impact on the citation network in this field. 3.3 Country visualization Betweenness centrality is an indicator of the status of individuals in a network (Freeman 1977). It can be used as a scientific measurement index to measure the importance of multiple indicators of scientific measurement units. The betweenness centrality of a node (=0.1) indicates that the point is a critical node (Li and Chen, 2017). According to an analysis of authors’ countries in CiteSpace, this study finds that American scholars published the most (54, 0.73)1 , accounting for 46.55% of the total publications; the sec­ond is China (10, 0.28) and the United Kingdom (8, 0.12), accounting for 8.62% and 6.90% respectively. Also, the betweenness centrality of Australia (7, 0.2) and Germany (4, 0.11) is higher than or equal to 0.1, indicating those countries play important roles in the scientific cooperation network, albeit to a lesser degree than the United States, China, and United Kingdom respectively. To analyze the cooperative relationship between the countries in this field further, the CiteSpace software was used to draw the map of the knowledge domain of the na­tional cooperation network (Figure 3). The threshold is set to top 100 per slice, that is, the top 100 countries in each time slice are extracted. The node ring shows the distribu­tion of the number of papers published by the country and the corresponding time. The color of the ring is consist­ent with the published time, the light color indicates the latest time and the dark color indicates the longer time. The thicker the ring, the more papers will be published at the corresponding time. The line between the two coun­tries represents the cooperative relationship, while the single node indicates that this country has no cooperative relationship with other countries in the field of research. The color of the line represents the time of the first co­operation between the two countries. The national coop­eration network in this field consists of 27 nodes and 27 connections, and the network density is 0.0769. Although the network density is not high, the cooperation between countries is significantly higher than that between authors. Among them, the United States is at the center of the inter­national cooperation network in this field. Scholars from other major countries mainly cooperate with American re­searchers. China, Australia, and the United Kingdom are significant contributors. Chinese scholars, who are the sec­ond most productive, mainly conduct cooperative studies with scholars from the United States, Australia, the United Kingdom, and Taiwan. 3.4 Institution visualization We used CiteSpace software to draw a map of the insti­tutional relationships in the knowledge domain and the (c, cc, ccv) thresholds were set to (1,2,20), (1,3,20), (1,3,20). The institutional cooperation network shown in Figure 4 is composed of 163 nodes and 151 connections, and the net­work density is 0.0114. The most active research institu­tions in this field are the National University of Singapore (6 articles), Drexel University (5 articles), Universidad Las Palmas (4 articles), Illinois State University (3 arti­cles), Southern Illinois University Carbondale (3 articles). Research amongst Chinese institutions was widely spread and includes the University of Science and Technology of China, Hefei University of Technology, Renmin Univer­sity of China, Zhejiang Gongshang University and Anhui University of Technology. There are only 31 nodes in the author’s cooperation network that have cooperative rela­tionships, forming seven cooperative research groups, and one group has 14 nodes. These differences in nodes indi­cate that the research institutions in this field are scattered and there is a lack of highly cohesive academic coopera­tive groups. 4 Main paths 4.1 Citation Network Analysis Citation network analysis refers to studying the con­text of knowledge dissemination and the distribution of the characteristics of the citations by analyzing the relation­ship between article citations. This study selects the top 30 cited articles and uses the HistCite software to generate the chronology of citations2 . The citations shown in Figure 5 are arranged in chronological order from top to bottom. The size of the ellipse indicates how many times the arti­cle has been cited. The number shown in each node indi­cates the serial number of the article in HistCite. The line with the arrow indicates the citation relationship between nodes. When looking at the direction of the arrow, the cit­ing article points to the cited article. There are a total of 30 nodes and 110 connections in the chronology. The maxi­mum LCS is 56 and the minimum is 7. Among them, nodes 7, 9, 16, 28, 41, 46, and 57 are the core nodes. The articles represented by these nodes are the most cited articles with a high citation frequency. These articles are important for the transfer of knowledge on cyberloafing. They play crit­ical roles in studying cyberloafing and have been widely cited by other scholars in later research. Important papers started with LAVOIE and PYCHYL (node 4) and BLOCK (node 5). Lavoie and Pychyl (2001) surveyed North American employees and they found that cyberloafing is related to procrastination and negative emotions. Block (2001) proposed that cyberloafing is a new and unique management challenge caused by new technologies. He analyzed cyberloafing from the perspec­tive of ethics, supervision, loss of productivity, and public policy. Lim (2002) (node 9) formally proposed the concept of cyberloafing, and developed a widely used measure­ment tool to test the propensity to cyberloaf, and laid a theoretical foundation for subsequent empirical research. Most of the subsequent studies (including Key nodes such as 16, 28, 41, and 43) stem from node 9. 4.2 Main path analysis Main path analysis is a quantitative method for iden­tifying critical paths and key nodes. It calculates all paths from the starting point to the terminal point. That is, start­ing from the earliest known reference or literature that does not cite other research, the method calculates all the paths leading to the latest research. The traversal weight of each edge is calculated after the paths are identified. The traversal weight reflects the importance of each citation in the overall development of the research. This research uses Pajek3 , a complex social network analysis tool, to map the main path of knowledge evolution. The visual presentation of the main path makes it easier to highlight the key arti­cles and citation relationships in the research development process. The path counting method (SPC) proposed by Bata­gelj (2003) was used to calculate the traversal weight of the edges. The global main path graph and regional main path graph are presented in Figure 6 and Figure 7, respec­tively. Among them, the label of each node is composed of the first author’s last name and publication year. The arrow reflects the direction of the knowledge flow, and the thickness of the line reflects the size of the SPC value. The thicker the line, the more significant it is in the main path. Figures 6 and 7 show that, except for the three nodes of Lavoie and Pychyl (2001), Davis et al. (2002) and Oravec (2002), the regional main path is exactly the same as the global main path, indicating that these 16 papers play key roles in the evolution of knowledge in this field. Table 1 summarizes the 16 articles appearing on both the regional main path and the global main path. These articles constitute the central structure of the network and play essential roles in the flow of knowledge in this field. Among them, Akbulut (No. 79, 93 first author), Anandara­jan (No. 6 first author, No. 12 second author), Ugrin (No. 50 first author, No. 104 third author), and Donmez (No. 79 third author, No. 93 second author) appeared twice in the main path. These authors have played important roles in the diffusion of knowledge. As aforementioned, Lim (2002) is a particularly important paper that has played a role in setting the stage for future research, namely through the development of a tool for measuring peoples’ propen­sity to cyberloaf. 5 Literature Summary We split the literature into three stages to depict the lifecycle. The three stages are the emergent stage (1997-2001), the exploration stage (2002 – 2013), and the bur­geoning stage (2014 – present). The earliest literature tends to be exploratory and examines a wide range of fac­tors. We refer to this as the Emergent stage (1997-2001). Cyberloafing is defined during this stage. Research dur­ing the emergent stage takes a high-level view of the con­tent and characteristics of cyberloafing and explores the symptoms and intervention measures. There are four im­portant articles in the stage. The first article appearing in our literature review, Polito (1997), defined cyberloafing, proposed causes and consequences, suggested that such behavior can be controlled. Lavoie and Pychyl (2001) ex­plored the relationship between cyberloafing and need for entertainment, stress relief, and feelings of negative emo­tions. Block (2001) has studied the relationship between cyberloafing and perceptions of the legality and morality of engaging in it. Greengard (2000) outlined hidden costs caused by cyberloafing. The Exploration stage (2002-2013), analyzes cyber­loafing as a workplace behavior in more detail, including its causes, benefits and consequences, and control mecha­nisms. The research is more theoretical and robust. Sem­inal articles in this stage include Lim (2002), who creat­ed a tool for measuring peoples’ propensity to cyberloaf, along with Blanchard and Henle (2008) who proposed a structural model of cyberloafing which laid a theoretical foundation for subsequent empirical research. Blanchard and Henle (2008) and Vitak et al. (2011) identified the an­tecedents of cyberloafing from the perspective of individu­al characteristics. They found that males, individuals with controlled personalities, young people, ethnic minorities, and individuals who are proficient at using computers are more likely to engage in cyberloafing behavior than oth­ers. Jia (2013) suggested that emotional state affects in­dividual cyberloafing behaviors as individuals who have stable emotions are less likely to cyberloaf than neurotic individuals. Other research finds that employees with neg­ative emotions are more likely to reduce emotional stress through cyberloafing (Moody and Siponen 2013). Vitak et al. (2011) also looked at the effects of job characteristics and found that the creativity of peoples’ work affects their propensity to cyberloaf. They found that the simpler peo­ples’ work and the higher the repetition, the more they are to engage in cyberloafing. Studies by Henle and Blanchard (2008), Garrett and Danziger (2010), and Reinecke (2009) have also shown that work stress and work fatigue can af­fect cyberloafing behavior. That is, employees engage in cyberloafing as a recovery experience when they encoun­ter job stress or fatigue. Blanchard and Henle (2008) also found that the more frequently employees use the Inter­net while at work, the more likely they are to cyberloaf. Lim(2002), Garrett and Danziger (2010), Reinecke (2009) and Ugrin and Pearson (2013) investigated cyberloafing from the context of organizational justice, social support, and the degree of control, finding that external conse­quences, social stigmas, and other neutralizing efforts in­fluence peoples’ propensity to cyberloaf. During this exploratory period, most scholars assumed and investigated the adverse effects of cyberloafing. For example, cyberloafing has been found to reduce engage­ment, waste time and reduce productivity (Griffiths 2011; Liberman et al. 2011). However, Mahatanankoon et al. (2004) believed that employees’ cyberloafing does not de­crease their work efficiency. A possibility that is explored in later research. The volume of research on cyberloafing has progres­sively increased since 2014, and we refer to this most re­cent state as the Burgeoning stage (2014 to present). The recent research continues to focus on measurement and antecedents (e.g. Akbulut et al. 2016; Sheikh et al. 2015), but two prominent new features have emerged. One is the extension of the research from the organizational context to the educational context, and the other is the positive ef­fects of cyberloafing. Akbulut et al. (2017) compared the differences in the level of cyberloafing between students and employees. Akbulut et al. (2016) developed a meas­urement tool for cyberloafing among students and com­pared the differences in the level of cyberloafing between male and female students. Gökçearslan et al. (2018) exam­ined the impact of social support and stressors on the un­dergraduate student’s cyberloafing behavior. Yilmaz and Yurdugül (2018) studied the influence of college students’ attitudes towards computers, psychosocial perceptions, and learning strategies on cyberloafing in the classroom. König and Guardia (2014) explored the positive effects of cyberloafing and found that cyberloafing helps employees manage the blurring of the lines between work and non-work time. Wu et al. (2018) analyzed both the positive and negative effects of cyberloafing on student performance, finding that cyberloafing causes harmful interference to students in the classroom. However, a moderate amount of cyberloafing out of the classroom can have a positive impact on student performance by offering a means for re­laxation (Wu et al. 2018). 6 Future research directions The research on cyberloafing has progressed from ear­ly literature that defined the phenomenon, to articulating how people cyberloaf, to looking at why they do it, the benefits and consequences of engaging in it, mechanisms aimed to control it, and cyberloafing in different domains. The most recent branches in figures 6 and 7 look at the effects of cyberloafing on student performance (Wu et al., 2018), smartphone addiction and cyberloafing amongst students (Gokcearslan et al., 2018), and student attitudes and cyberloafing in classrooms (Yilmaz and Yurdugül, 2018). Future research should extend these streams. Wu et al. (2018) recommend that their study can be extend­ed by longitudinal studies that track student performance over time. Gokcearslan et al. (2018) recommend that their study be extended by collecting data from a wide range of students with varying demographics. Yilmaz and Yurdugül (2018) also recommend that their study can be extended by exploring individual differences, including differences in cognitive characteristics, self-efficacy, introversion, et ce­tera. All three of these studies suggest that future research use more direct measures of cyberloafing. These studies, and the majority of cyberloafing research, tend to use in­direct measures such as questionnaires to assess cyberloaf­ing behavior. The literature would also benefit from more research into the mental health consequences of cyberloafing. Re­cent research has found that cyberloafing can help employ­ees cope with stress (Stoddart, 2016), but it also consumes employees’ cognitive resources (Windeler, Chudoba, and Sundrup, 2017). Research should look at the point in which cyberloafing ceases to become a positive or negative activ­ity and starts to have paradoxical mental health effects on employees or students. A recent study by Wu et al. (2020) demonstrates that social cyberloafing can have both posi­tive and negative effects on employee mental health. Much of the early research on cyberloafing was per­formed before mobile Internet devices became as prevalent as they are today. It would be interesting to see people’s reactions to the changing domain. For example, literature has shown that individuals are more likely to cyberloaf when they have access and privacy (Ugrin, Pearson, and Odom, 2007); mobile devices create nearly unlimited ac­cess and privacy. Cyberloafing may be affected by differences across cultures as individuals from different countries may be more or less inclined to use Internet resources for personal purposes when at work or school. Researchers may find models like the Hofstede Model of Cultural Dimensions (Hofstede, Hofstede, and Minkov, 2010) useful for identi­fying facets of various cultures that might lead to more or less cyberloafing in a particular country. Knowledge about the relationship between culture and cyberloafing could be useful for managers at multi-national corporations with of­fices and employees all over the world. Cyberloafing literature would benefit from a better grounding in psychology theory and the integration of various theories. Various aspects of cyberloafing could be explained by theories like Prospect Theory, the Theory of Planned Behavior, Rational Choice Theory, Low Self-Con­trol Theory, Organizational Justice Theory, General Deter­rence Theory, and more. An integrated theory that brings the relevant aspects of these theories into one model would be groundbreaking. 7 Limitations and Conclusions This study examines 116 high-quality papers on cyber­loafing behavior indexed by SSCI and SCIE in the WOS database. We used bibliometric analysis to quantitatively analyze the cooperation network, citation network, and main path for more than 20 years of research on cyberloaf­ing. We map the cooperation network, citation chronology and main path of cyberloafing research from 1997 to 2019, and reached the following conclusions. First, research in­terest in the field of cyberloafing is continuing to rise and is currently in a burgeoning stage. However, a complete theoretical model of cyberloafing has not been formed, and research interest in this field will continue to rise in the future. There are many scholars currently engaged in the research of cyberloafing, but the most influential authors are Lim, PabloZoghbi-Manrique-de-Lara, Anandarajan, and O’Neill. Although there is some cooperation among productive authors, in general, the degree of cooperation is low, and the level of cooperation between scholars across countries could be improved. Second, the findings show that the United States, Chi­na, the United Kingdom, and Australia have a large num­ber of publications and are at the core of the field of cyber­loafing. The United States has the most publications, and China ranks second. However, there is still a significant gap between China and the United States. Third, up to now, the most influential institutions engaged in cyberloafing research include the National University of Singapore, Drexel University, Las Palmas University, Illinois State University, and Southern Illinois University Carbondale. There is a limited amount of con­nection between these active institutions, but overall, the institutions’ cooperation network density is low, and the cooperative relationship is not close-knit. The existing co­operative research is limited to a few core networks, such as the cooperation network centered on Drexel University, St. Joseph University, and the University of Illinois. Fourth, by analyzing the citation chronology and the main path of the cyberloafing research network, we con­clude that cyberloafing research has evolved through three stages, the emergent stage, the exploration stage, and the burgeoning stage. The research topic has developed from concept definition and measurement, to antecedents and outcomes. Further, the research object has shifted from organizational employees to both employees and students in educational settings. Recent research has begun to fo­cus on the positive effects of cyberloafing on individuals and organizations. However, research in this direction has been controversial, shifting from focusing on the negative effects and the assumption the cyberloafing is inherently bad. This study makes several contributions to cyberloafing literature. First, this is the first study to rely on bibliometric and visualization techniques to map the development of cyberloafing research. This quantitative methodology can reduce potential biases due to human subjectivity. Second, this study determines influential authors, countries, and in­stitutions in the cyberloafing research and reveals the den­sity and degree of cooperation among authors, countries, and institutions. Third, by conducting a main path analy­sis, we identify 16 important papers playing critical roles in the development of cyberloafing research. We find that cyberloafing research has evolved through three stages, the emergent, exploration, and burgeoning stages. This finding extends past literature reviews on cyberloafing (Mercado et al., 2017) by shifting to an evolutionary perspective. This study has a few limitations that are important to mention. First, our analysis covers a selected set of high-quality SCIE and SSCI journals in the WOS data­base. We assume that these publication sources can reason­ably represent cyberloafing research. However, the WOS database cannot guarantee the coverage of all possible cy­berloafing publications, which means that we might have missed some papers, particularly those published at con­ferences. Second, this study divides the development pro­cess of cyberloafing research into three stages according to the number of papers published each year and the results of the citation network and main path analysis. That is, the bibliometric approach is not entirely objective, and some amount of human judgment is needed to make the results meaningful. We also notice that it is impossible and some­times undesirable to completely eliminate human subjec­tivity from scientific research. Future research is needed to develop meaningful procedures to keep a suitable balance between objectivity and subjectivity in conducting litera­ture reviews. Acknowledgement This research was funded by the Humanity and Social Science Major Foundation of Education Committee of Anhui province (Grant No. SK2019ZD07) and the Nat­ural Science Foundation of Anhui Province (Grant No. 1908085MG238 and 1908085QG301). Literature 1Received: 14th December 2020; revised: 15th May 2021; accepted: 20th May 2021 . This author shares the first authorship. Note: The date of data collection is March 28, 2019. The curve is quadratic polynomial trend line of the number of paper published each year. Figure 1: Annual distribution of cyberloafing papers (1997-2019) Figure 2: Author visualization 11 The front number is the frequency of occurrence, and the back number is the betweenness centrality of the country. Figure 3: Countries visualization Figure 4: Institution Visualization 12 We chose 30 articles because if the number of articles involved in the chronology of citations is too large, the graphics would be too complicated, and the overall visualization will be poor. However, if the number of articles involved is too small, it would be difficult to find the knowledge evolution path. Figure 5: Citation Chronology 13 Batagelj V. & Mrvar A. (2004) discuss Pajek in more detail. Pajek software can be downloaded at: http://mrvar.fdv.uni-lj.si/pajek/. Figure 6: Global main path Figure 7: Local main path Table 1: Important articles appearing on both global (figure 6) and local (figure 7) main paths No. Title First author Year LCS GCS 6 Internet abuse in the workplace Anandarajan M 2002 6 19 9 The IT way of loafing on the job: cyberloafing, neutralizing and organiza­tional justice Lim VKG 2002 56 213 12 Development of a measure of personal web usage in the workplace Mahatanankoon P 2004 9 26 23 Psychology of communication technology use in the workplace Eastin MS 2007 13 21 30 Disaffection or expected outcomes: Understanding personal Internet use during work Garrett RK 2008 13 33 34 Games at Work: The Recreational Use of Computer Games During Working Hours Reinecke L 2009 5 32 41 Personal Internet use at work: Understanding cyberslacking Vitak J 2011 25 78 43 Employee job attitudes and organizational characteristics as predictors of cyberloafing Liberman B 2011 20 42 50 The effects of sanctions and stigmas on cyberloafing Ugrin JC 2013 17 35 54 Exploring the positive side of personal internet use at work: Does it help in managing the border between work and nonwork? König CJ 2014 7 11 73 The antecedents of cyberloafing: A case study in an Iranian copper industry Sheikh A 2015 5 14 79 In search of a measure to investigate cyberloafing in educational set­tings Akbulut Y 2016 5 11 93 Cyberloafing and social desirability bias among students and employees Akbulut Y 2017 3 6 104 Student Cyberloafing In and Out of the Classroom in China and the Relationship with Student Performance Wu J 2018 0 0 106 Smartphone addiction, cyberloafing, stress and social support among university students: A path analysis Gokcearslan S 2018 0 0 112 Cyberloafing in IT classrooms: exploring the role of the psycho-social environment in the classroom, attitude to computers and computing courses, motivation and learning strategies Yilmaz R 2018 0 0 Akbulut, Y., Dönmez, O., & Dursun, Ö.Ö. (2017). Cyber­loafing and social desirability bias among students and employees. Computers in Human Behavior, 72, 87-95. https://doi.org/10.1016/j.chb.2017.02.043 Akbulut, Y., Dursun, Ö.Ö., Dönmez, O., & Sahin, Y.L. (2016). In search of a measure to investigate cyber­loafing in educational settings. Computers in Human Behavior, 55, 616-625. https://doi.org/10.1016/j.chb.2015.11.002 Anandarajan, M. (2002). Internet Abuse in the Workplace. Communications of the ACM, 45(1), 53-54. Andreassen, C.S., Torsheim, T., & Pallesen, S. (2014). Predictors of Use of Social Network Sites at Work - A Specific Type of Cyberloafing. Journal of Computer-mediated Communication, 19(4), 906-921. https://doi.org/10.1111/jcc4.12085 Batagelj, V. (2003). Efficient algorithms for citation net­work analysis. In: http://arxiv.org/abs/cs.DL/0309023 Batagelj, V., & Mrvar, A. (2004). Pajek — Analysis and Visualization of Large Networks. In: Jünger M., Mutzel P. (eds) Graph Drawing Software. Mathe­matics and Visualization. Springer, Berlin, Heidel­berg. https://doi.org/10.1007/978-3-642-18638-7_4 Blanchard, A.L., & Henle, C.A. (2008). Correlates of dif­ferent forms of cyberloafing: The role of norms and external locus of control. Computers in Human Be­havior, 24(3), 1067-1084. https://doi.org/10.1016/j.chb.2007.03.008 Block, W. (2001). Cyberslacking, Business Ethics and Man­agerial Economics. Journal of Business Ethics, 33(3), 225-231. https://doi.org/10.1023/A:1012002902693 Coker, B.L.S. (2013). Workplace Internet Leisure Brows­ing. Human Performance, 26(2), 114-125. https://doi.org/10.1080/08959285.2013.765878 Davis, R.A., Flett, G.L., & Besser, A. (2002). Validation of a new scale for measuring problematic internet use: implications for pre-employment screening. Cy­berPsychology & Behavior, 5(4), 331-345. https://doi.org/10.1089/109493102760275581 Freeman, L.C. (1977). A Set of Measures of Centrali­ty Based on Betweenness. Sociometry, 40(1), 35-41. https://doi.org/10.2307/3033543 Gökçearslan, S., Uluyol, Ç. & Sahin, S. (2018). Smart­phone addiction, cyberloafing, stress and social sup­port among university students: A path analysis. Chil­dren and Youth Services Review, 91, 47-54. https://doi.org/10.1016/j.childyouth.2018.05.036 Garrett, R.K., & Danziger, J.N. (2008). On cyberslacking: workplace status and personal internet use at work. CyberPsychology & Behavior, 11(3), 287-292. https://doi.org/10.1089/cpb.2007.0146 Garrett, R.K., & Danziger, J.N. (2010). Disaffection or expected outcomes: Understanding personal Internet use during work. Journal of Computer-Mediated Com­munication, 13(4), 937-958. https://doi.org/10.1111/j.1083-6101.2008.00425.x Greengard, S. (2000). The high cost of cyberslacking. Workforce, 79(12), 22-24. https://www.workforce.com/2000/12/01/the-highcost-of-cyberslacking/ Griffiths, M.D. (2011). Internet abuse in the workplace – Issues and concerns for employers and employment counselors. Journal of Employment Counseling, 40(2), 87-96. https://doi.org/10.1002/j.2161-1920.2003.tb00859.x Heflin, H., Shewmaker, J., & Nguyen, J. (2017). Impact of mobile technology on student attitudes, engagement, and learning. Computers & Education, 107, 91-99. https://doi.org/10.1016/j.compedu.2017.01.006 Henle, C.A., & Blanchard, A.L. (2008). The interaction of work stressors and organizational sanctions on cyber­loafing. Journal of Managerial Issues, 20(3), 383-400. https://www.jstor.org/stable/40604617 Hofstede, G., Hofstede, G. J. & Minkov, M. (2010). Cul­tures and Organizations: Software of the Mind (Rev. 3rd ed.). New York: McGraw-Hill. Jia H , Jai, .R., & Karau S. (2013). Cyberloafing and Personality: The Impact of the Big Five Traits and Workplace Situational Factors. Journal of Leadership & Organizational Studies, 20, 358-365. https://doi.org/10.1177/1548051813488208 Jia, R., & Jia, H. (2015). An Individual Trait-based Investi­gation of Employee Cyberloafing. Journal of Informa­tion Technology Management, 26(1), 58-71. Jia, Y., Wang, N., & Ge, S. (2018). Business-IT Alignment Literature Review: A Bibliometric Analysis. Informa­tion Resources Management Journal, 31(3), 34-53. https://doi.org/10.4018/IRMJ.2018070103 Kamins, A. (1995). Cyber-Loafing: does employee time online add up to net losses? New York Daily News, July. König, C.J., & Guardia, M.E.C.D.L. (2014). Exploring the positive side of personal internet use at work: Does it help in managing the border between work and non­work? Computers in Human Behavior, 30(30), 355-360. https://doi.org/10.1016/j.chb.2013.09.021 Lavoie, J.A.A., & Pychyl, T.A. (2001). Cyberslacking and the Procrastination Superhighway: A Web-Based Survey of Online Procrastination, Attitudes, and Emo­tion. Social Science Computer Review, 19(4), 431-444. https://doi.org/10.1177/089443930101900403 Li, J., & Chen, C. (2017). CiteSpace: Text Mining and Visualization in Scientific Literature. Beijing, Capital University of Economics and Business Press. Liberman, B., Seidman, G., Mckenna, K.Y.A., & Buffardi, L.E. (2011). Employee job attitudes and organizational characteristics as predictors of cyberloafing. Comput­ers in Human Behavior, 27(6), 2192-2199. https://doi.org/10.1016/j.chb.2011.06.015 Lim, V.K.G. (2002). The IT Way of Loafing on the Job: Cyberloafing, Neutralizing and Organizational Justice. Journal of Organizational Behavior, 23(5), 675-694. https://doi.org/10.1002/job.161 Mahatanankoon, P., Anandarajan, M., & Igbaria, M. (2004). Development of a measure of personal web usage in the workplace. CyberPsychology & Behavior, 7(1), 93-104. https://doi.org/10.1089/109493104322820165 Meijman, T.F., & Mulder, G. (1998). Psychological As­pects of Workload. In P.J.D. Drenth, H. Thierry and C.J. de Wolff (Eds.), New Handbook of Work & Or­ganizational Psychology (2nd ed.) (pp. 5-33). Hove, England: Psychology Press/Erlbaum (UK) Taylor & Francis. Mercado, B.K., Giordano, C., & Dilchert, S. (2017). A me­ta-analytic investigation of cyberloafing. Career De­velopment International, 22(1), 546-564. https://doi.org/10.1108/CDI-08-2017-0142 Moody, G.D., & Siponen, M. (2013). Using the theory of interpersonal behavior to explain non-work- re­lated personal use of the Internet at work. Informa­tion & Management, 50(6), 322-335. https://doi.org/10.1016/j.im.2013.04.005 O’Neill, T.A., Hambley, L.A., & Bercovich, A. (2014). Prediction of cyberslacking when employees are working away from the office. Computers in Human Behavior, 34(4), 291-298. https://doi.org/10.1016/j.chb.2014.02.015 Oravec, J.A. (2002). Constructive approaches to in­ternet recreation in the workplace. Communi­cations of the ACM, 45(1), 60-63. https://doi.org/10.1145/502269.502298 Phillips, J.G. (2006). The psychology of Internet use and misuse. M. Anandarajan, T. Thompson, C. Sim­mers (Eds.), The Internet and workplace transforma­tion, Sharpe, Armonk (2006), pp. 41-62. Polito, A. (1997). Cyberloafing can be curbed. Workforce, 76, 18. Price, D.J. de S. (1963). Little Science, Big Science. New York: Columbia University Press. Reinecke, L. (2009). Games at work: the recreational use of computer games during working hours. CyberPsy­chology & Behavior, 12(4), 461-465. https://doi.org/10.1089/cpb.2009.0010 Sheikh, A., Atashgah, M.S., & Adibzadegan, M. (2015). The antecedents of cyberloafing: A case study in an Iranian copper industry. Computers in Human Behavior, 51, 172-179. https://doi.org/10.1016/j.chb.2015.04.042 Stoddart, S.R. (2016). The Impact Of Cyberloafing And Mindfulness On Employee Burnout. Wayne State University. Retrieved from: https://digitalcommons.wayne.edu/cgi/viewcontent.cgi?referer=https://schol­ ar.google.com/&httpsredir=1&article=2486&contex­ t=oa_dissertations Syrek, C.J., Kühnel, J., Vahlehinz, T., & Bloom, J.D.J.W. (2017). Share, like, twitter, and connect: Ecological momentary assessment to examine the relationship be­tween non-work social media use at work and work engagement. Work & Stress, 32(3), 209-227. https://doi.org/10.1080/02678373.2017.1367736 Ugrin, J.C., & Pearson, J.M. 2013. The effects of sanctions and stigmas on cyberloafing. Computers in Human Behavior, 29(3), 812-820. https://doi.org/10.1016/j.chb.2012.11.005 Ugrin, J. C., Pearson, J. M., & Odom, M. D. 2007. Pro­filing cyber-slackers in the workplace: Demograph­ic, cultural, and workplace factors. Journal of Inter­net Commerce, 6, 75–89. https://doi.org/10.1300/J179v06n03_04 Vitak, J., Crouse, J., & Larose, R. (2011). Personal Internet use at work: Understanding cyberslacking. Computers in Human Behavior, 27(5), 1751-1759. https://doi.org/10.1016/j.chb.2011.03.002 Windeler, J. B., Chudoba, K. M., & Sundrup, R. Z. (2017). Getting away from them all: Managing exhaustion from social interaction with telework. Journal of Or­ganizational Behavior, 38(7), 977-995. https://doi.org/10.1002/job.2176 Wu, J., Mei, W., & Ugrin, J.C. (2018). Student Cyber­loafing In and Out of the Classroom in China and the Relationship with Student Performance. Cyberpsy­chology,Behavior,and Social Networking, 21(3), 199-204. https://doi.org/10.1089/cyber.2017.0397 Wu, J., Mei, W., Liu, L., & Ugrin, J.C. (2020). The bright and dark sides of social cyberloafing: Effects on em­ployee mental health in China, Journal of Business Research, 112, 56-64. https://doi.org/10.1016/j.jbus­ res.2020.02.043 Yilmaz, R., & Yurdugül, H. (2018). Cyberloafing in IT classrooms: exploring the role of the psycho-social en­vironment in the classroom, attitude to computers and computing courses, motivation and learning strategies. Journal of Computing in Higher Education (4), 1-23. https://doi.org/10.1007/s12528-018-9184-2 Zhang, H., Zhao, H., Liu, J., Xu, Y., & Lu, H. (2015). The dampening effect of employees’ future orientation on cyberloafing behaviors: the mediating role of self-con­trol. Frontiers in Psychology, 6, Article 1482. https://doi.org/10.3389/fpsyg.2015.01482 Jin-Nan Wu is a professor of MIS and the Dean of the School of Business at Anhui University of Tech­nology, China. His research focuses on cyberloafing, e-commerce, online consumer behavior, and social and managerial issues of IT. Dr. Wu has published more than 30 articles in Cyberpsychology Behavior and Social Net­working, Information Technology & People, Industrial Management & Data Systems, Electronic Commerce Research, Journal of Business Research, and others. Mengmeng Song is a postgraduate of Cyberpsychology at Anhui University of Technology, China. Her research focuses on social contagion of cyberloafing. Joseph C. Ugrin is the Deloitte Professor of Accounting and Head of the Department of Accounting at the University of Northern Iowa, USA. He primarily researches behavioral issues in accounting and information systems and has published research in the area of information systems and technology in journals such as Computers in Human Behavior, the Journal of Business Research, Information Technology & People, Communications of the Association for Information Systems, the Journal of Information Systems Education and others. Lin Liu is a lecturer of Human-computer interaction at Anhui University of Technology, China. Her research focuses on human-computer interaction, online consumer behavior, and social media. She has published more than 10 articles in Information Technology & People, Industrial Management & Data Systems, Journal of Business Research, Frontiers in Psychology, Journal of Electronic Commerce in Organizations and some MIS conferences such as PACIS and WHICEB. Tingting Zhu is a lecturer of organizational behavior at Anhui University of Technology, China. Her research focuses on ethics and information technology, and social media use in workplace. Raziskovanje spletnega postopanja 1997-2019: Analiza literature z vidika citiranja Ozadje in namen: Namen te študije je ponazoriti razvoj raziskav o spletnem postopanju in opredeliti pomembne prispevke pri razvoju znanja o omenjenem pojavu. Metodologija: Iz baze podatkov Web of Science (WOS) smo izbrali skupno 116 raziskovalnih clankov, ki so bili objavljeni v najboljših revijah v obdobju 1997–2019. Izvedli smo analizo citiranja, analizo ko-citiranja in analizo glav­nih poti, da smo identificirali najvplivnejše avtorje, države in institucije pri razvoju raziskav spletnega postopanja in graficno predstavili odnose med njimi. Rezultati: Raziskave spletnega postopanja so napredovale skozi tri faze, ki sestavljajo novo nastajajoco, razisko­valno in narašcajoco fazo. Ugotavljali smo tudi razmerja med clanki in prepoznavamo locene in pomembne tokove znanja, 16 vplivnih prispevkov, ki so temeljni ali dobro vidni, ter avtorje, institucije in države, kjer je raziskava nastala. Najvec raziskav so opravili raziskovalci v ZDA, sledile so jim Kitajska, Velika Britanija in Avstralija. Nacionalna uni­verza v Singapurju je opravila najvec raziskav v našem vzorcu, Vivian K. Lim iz Singapurske nacionalne univerze pa je bila najvplivnejša avtorica, njen temeljni prispevek o spletnem postopanju in organizacijski pravicnosti, objavljen leta 2002. Zakljucek: Po našem najboljšem vedenju je študija je prva, ki je preucila razvoj raziskav o spletnem postopanju z uporabo metod, ki temeljijo na dokazih. Prikazuje razvoj literature o spletnem postopanju in identificira pomembne tokove, ki bi jih raziskovalci lahko proucevali v prihodnosti. Sklepamo, da bi morale prihodnje raziskave ustvariti bolj neposredna merila za spletno postopanje, preuciti posamezne posledice, ki jih povzroci prevec spletnega postopa­nja, na primer njegov vpliv na mentalno zdravje. Kljucne besede: Spletno postopanje, Analiza citatov, Analiza citiranja, Analiza glavne poti, Vizualizacija DOI: 10.2478/orga-2021-0008 The Role of Daily Affect in Leader-Member Exchange: A Multilevel Investigation in Public Health Administration Mehmet ÇETIN1, Gulmira SAMENOVA2, Filiz TÜRKKAN2, Ceylan KARATAS2 1 Istanbul Sabahattin Zaim University, Faculty of Business and Management Sciences, Istanbul, Turkey, mehmet.cetin@izu.edu.tr (corresponding author) 2 Istanbul Sabahattin Zaim University, Graduate Education Institute, Istanbul, Turkey, samenova.gulmira@std.izu.edu.tr, turkkan.filiz@std.izu.edu.tr, ceylankaratas88@gmail.com Background and purpose: Although the critical role of affect in the leader-member relationship has been widely accepted, few studies investigated the impact of within-person affect variations in daily leader-member exchange (LMX) or addressed potential cross-level and intra-individual moderators of this relationship. This study examines the effects of followers’ positive and negative affect on their daily LMX in public health care organizations. The moderator roles of emotional labor and trait emotional intelligence were also investigated. Methodology: A multilevel research design was conducted where daily measures were nested in individuals. Sev­enty participants working in a government health organization operating in Istanbul responded to daily surveys for five consecutive workdays (350 day-level responses) and a general survey one week after the daily data collection period (70 person-level responses). Hypotheses were tested using Hierarchical Linear Modeling (HLM). Results: Both positive affect and negative affect were positively related with LMX (day-level), but negative affect had a negative association with LMX on the inter-personal level (when daily scores were averaged across days). Although trait emotional intelligence showed a positive cross-level effect, none of the proposed moderations was significant. Conclusion: The role of affect in LMX development is critical and has a complex structure. Findings emphasize the importance of multilevel research for understanding the affect-LMX relationship as they demonstrate different pic­tures in day-level and person-level analysis. Keywords: Positive affect, Negative affect, Emotional labor, Leader-member exchange, Trait emotional intelligence, Day-level interaction 1 Introduction The importance of public healthcare organizations in the service sector and the economy has been growing in Turkey since the introduction of the health transformation program in 2003, which brought sound quality and capac­ity improvements in the public health care system (Bener et al., 2019). Public hospitals constitute the leading service provider role (having the majority of bed capacity) in Tur­key’s total health care system (Dundar et al., 2010). All over the World, public health organizations have been fac­ing complex and substantial leadership challenges (Shick­le et al., 2014). In the public health care context, the extent of the responsibility and the severity of the outcomes of any possible mistakes are highly critical. In addition, re­quirements for effective management during intense crises and the demanding nature of the job regarding the regula­tion of affects in the workplace make affective structures and leader-member relationships crucial for public health administration. The significance of affect in social exchange and role-making processes has been widely recognized by organizational behavior scientists (Cropanzano et al., 2017a). The term “affect” in organizational behavior stud­ies refers to emotions people experience and transform into work environments while interacting with others (Brief & Weiss, 2002). Favorable emotions like enthusi­asm, joy, happiness, etc., comprise positive affect, whereas sadness, fear, shame, and similar feelings form negative affect. Affective events theory (AET) underscores the role of feelings in the formation of employee attitudes and be­haviors (Cropanzano et al., 2017a). Like daily affective experiences, affective trait constructs, such as trait EI, can shape organizational outcomes (Li et al., 2018). Thus, a substantial number of organizational studies have scruti­nized how affect shapes the interpersonal relationship and the leader-member exchange (LMX) in particular (e.g., Liu et al., 2020). Among the leadership models investigated in the pre­vious millennium, the LMX theory is undoubtedly one of the most fascinating. In contrast to the idea that a leader behaves equally toward all members, the LMX concept propagates a different nature of the link between a super­visor and all his/her subordinates. It is a complex process determined by long-term as well as momentary intrinsic and extrinsic elements like the personality of the leaders and the followers, their affect, job specification, cultural differences, gender, age, organizational environment, etc. (Nahrgang & Seo, 2016; Hofmans et al., 2019). In previ­ous decades, LMX research has been predominantly fo­cused on leadership constructs such as leader personality and leadership style and their impact on organizational outcomes (Uhl-Bien et al., 2014). Fewer studies have in­vestigated the follower side of the construct. In particular, the scarcity of the research is more noticeable for attempts to explain how state affective phenomena (e.g., positive and negative affect) shape followers’ perception of LMX quality. Therefore, in this paper, various affect phenomena of the followers are tested to reveal their impact on LMX. Moreover, the LMX quality is not stable and may vary in time (Ellis et al., 2019). Despite the dominance of stud­ies addressing the construct as a stable concept, the central promise and the roots of the LMX are founded on the sug­gestion that the exchange and relationship between lead­er and followers are dynamic in nature (Hofmans et al., 2019). One of the primary triggers in LMX quality fluc­tuations can be the individual affective state or emotions (Liu et al., 2020). How individuals feel may change across the days, and moods and emotions are strongly tied to in­terpersonal exchange (Cropanzano et al., 2017b). Studies indicating intra-individual variance in affective states and LMX (Nezlek and Allen, 2006; Hofmans et al., 2019); and cross-level effects of affective constructs on interpersonal exchange in organizations (e.g., Tse et al., 2008) are sup­portive for the need for addressing day level variations and cross-level effects regarding study variables. The majority of studies on the affect-LMX relationship have considered it as having a static and between-person nature rather than investigating how momentary within-person variance in­fluence the affect-LMX development. Most of the early research on affect and LMX was carried out via single-lev­el models, and few studies have addressed the momentary within-person affect variations in leader-member exchange processes during their daily contacts (e.g., Tse et al., 2018). Thus, the current study may shed some light on the re­search into this matter. A more significant gap remains in investigating possible daily and person-level interactions among affect constructs and daily LMX. Addressing the aforementioned gaps, the present study analyzes the role and influence mechanisms of various day and person-level affect constructs (positive affect, negative affect, emotional labor, and trait emotional intelligence) on the perception of LMX in the work environment. In particular, the research focuses on investigating how subordinates’ affective states influence their daily LMX and which individual emotional characteristics may regulate this interaction. The outcomes of daily emotional experiences can vary according to extent and type of the emotion regula­tion strategy used for managing the internal state or the external expression of emotions (Costakis et al., 2021). The current study adopts Lee and Brotheridge (2011)’s conceptualization of emotional labor (EL) which defines EL as the reconstruction of the emotion (deep acting) or modification of its presentation (surface acting) in three dimensions (dividing the latter into two), not showing the genuine emotions (hiding feelings), changing the display of the sentiment (faking emotions) and modifying the in­ternal emotion (deep acting). How employees regulate the NA and PA they experience during the day can affect their impact on LMX outcomes (Tse et al., 2018). Along with the day level affect constructs (EL, NA, and PA) that are addressed at the within-person level, the current study also investigated cross-level effects of EI as a person-level affective trait variable. EI describes the ca­pacity, tendencies, and competence to detect, internalize, realize, and finally manage emotions (Mayer & Salovey, 1997). Thus, EI is about knowing or understanding emo­tions and includes using them in a functional way (reg­ulation and management of emotions), which makes this construct essential for the present study. Among differ­ent conceptualizations that have been provided for EI in the literature, ‘ability EI’ (assessing EI as a competence through performance tests) and ‘trait EI’ (assessing EI as a personality structure through self-reports) approaches are prevalent (see, O’Connor & Little, 2003). Petrides & Furnham (2001) underscore the significance of the meas­urement method for the operational definition of the con­cept and suggest a self-report, dispositional measurement approach instead of assessing the performance (ability). In line with the present study aims, following Petrides and Furnham (2001), EI is conceptualized as a trait that de­scribes the behavioral tendencies and self-perceived per­sonal characteristics. This study aims to contribute to the extant knowledge on the affect-LMX relationship through a multilevel nest­ed research design. Mainly, study aims to investigate the person-level and intra-individual level effects of positive affect (PA) and negative affect (NA) on LMX. The direct (cross-level) effect and moderator effect of trait emotional intelligence (EI) and emotional labor (EL) (hiding, faking, and deep acting) on this association are also examined. The multilevel investigation of the affective anteced­ents of daily LMX is critical to study due to several rea­sons. First, LMX is a determinant for numerous critical employee attitudes, behaviors, and organizational out­comes specially in health care organizations (Shickle et al., 2014; Ürek & Ugurluoglu, 2019). Thus, the research on the development and antecedents of this concept is pre­cious. Second, despite their importance for organizations, the affective antecedents of LMX remain understudied. The general stream of findings addresses state-affective constructs such as positive and negative affect as conse­quences of LMX, and the path from affect to LMX is often neglected. Third, addressing the gap above regarding the investigation of LMX as a dynamic structure and including the intra-individual variance in the analysis can enhance the information on the matter. Lastly, the examination of possible dynamics influencing the association between daily affect and LMX can further develop our understand­ing regarding the nature of the affect-LMX relationship. Built on the affective events theory, this study aims to address the aforementioned gaps and contribute to extant knowledge on the affect-LMX association through a multi­level research design where intra-individual, inter-individ­ual, and cross-level variations and effects are investigated. This way, the study examines affect as an antecedent of daily LMX and tests cross-level and moderator effects of EI and EL in this relationship. This provides a shift for both the causality in the affect-LMX relationship and the exploration of the day-level and person-level nature of this association. Moreover, this study aims to advance our knowledge of the daily and cross-level dynamics regulat­ing this relationship. 2 Theoretical Framework and Construction of Hypotheses 2.1 Positive and Negative Affect and LMX Affect can be defined as emotions people show or feel during their interaction with others and during deci­sion-making processes. It can be positive (e.g., pride, en­thusiasm, joy) or negative (e.g., fear, distress, sadness). In terms of duration, affect has been considered from two viewpoints: as a trait that reflects basic feelings and is more stable over time, and as a state that is defined as the present mood/emotion that changes within-person over time (Tellegen et al., 1999). Many studies stressed the im­portance of affect in the development of LMX processes. Consequently, the research interest in the affect-LMX re­lationship has been growing since 2000 (Tse et al., 2018). The LMX theory has its roots in Dansereau et al. (1975)’s efforts on the vertical dyad linkages theory. The authors of the theory suggested that dyadic relationships between superiors and subordinates can explain varied re­lationships in an organization. LMX addressed the change in the quality of the dyadic relationship between the leader and the follower over time. Dienesch & Liden (1986) sug­gested a multidimensional construct for LMX consisting of perceived contribution, loyalty, and liking. Professional respect was added as the fourth dimension later (Dienesch & Liden, 1986). LMX is a dyadic relationship where both parts of in­teraction influence the exchange quality. Initially, the vast majority of the research dedicated to LMX has focused on the leader’s characteristics, while follower attributes have attracted organizational behavior scientists in recent dec­ades (Uhl-Bien et al., 2014). In general, followers expressing positive affect (PA) are expected to develop high-quality LMX, whereas negative affect (NA) is very likely to deteriorate their relationship with leaders (Dulebohn et al., 2012). Because PA reflects the bright side of personal characteristics like enthusiasm, optimism, motivation, and engagement, leaders are likely to support followers with a high PA, and therefore, the re­lationship between them is expected to have higher qual­ity (Liu et al., 2020). On the other hand, NA is associated with different kinds of negative feelings like fear, hostility, distress, exhaustion, and apathy, which foster an adverse reaction on people, situations, and interactions with others in the surrounding environment, including the workplace (Glasø et al., 2018). Subsequently, the relationship be­tween a subordinate with high NA and his/her supervisor is assumed to be less favorable and less effective (Bernerth et al., 2007). 2.2 Affective Events Theory (AET) Individuals confront different situations within the work environment, and these conditions impact their emotions and moods (Weiss & Cropanzano, 1996). For instance, positive or negative affect may arise in the re­lationship between supervisors and subordinates during daily interactions that illustrate the LMX perception. The affective events are not bounded only by organizational in­cidents but may include extra-work situations that change the emotional state of a person (Ashton-James & Ash­kanasy, 2008). Everyday family-related issues, as well as positive events, arouse negative or positive moods within an individual. Unconsciously, individuals transfer emo­tions inherited in these non-work associated conditions to their job environments and subsequently react to heteroge­neous events occurring in the organization through the lens of earlier experienced emotional situations. Such affective events may influence the personal attitudes and behaviors of organizational members as well as their LMX percep­tions (Cropanzano et al., 2017b). A considerable number of the studies that address AET as the foundation of the affect-LMX relationship consider LMX an antecedent of affect and positions LMX experienc­es as work events. However, LMX can also be regarded as an affect-driven consequence because the theory indicates that employees’ behaviors are guided by their emotions. How individuals feel reflects their behaviors and attitudes, which are fundamental elements of LMX. Cropanzano et al. (2017a), using AET for explaining the affect-LMX association, emphasized the importance of affect for the LMX development. They also highlighted the existence of alternative paths and a two-way causality where interac­tions with the leader shape the affective states and affects influence the LMX nature. In their metanalysis, Tse et al. (2018) demonstrated that LMX had been considered as both an antecedent and a consequence in different studies that are addressing the affect-LMX relationship. Based on the framework mentioned above and pat­tern of results regarding LMX and affect, we suggest that day-level affect will influence the fluctuations in subordi­nates’ LMX perception and propose the following hypoth­eses: H1: Daily positive affect positively influences the daily LMX perception of the employees. H2: Daily negative affect negatively influences the dai­ly LMX perception of the employees. 2.3 Emotional Labor, Positive Affect, Negative Affect, and Leader-Member Exchange Despite the different definitions of emotional labor (EL) in the organizational behavior literature, on a broad­er sense, it represents the act of showing the socially de­sired emotions in the service transaction processes (Hoch­schield, 1983). Interactions between different actors at work (client-officer, leader-subordinate, etc.) are subject to emotional experiences, and how emotions are managed during these interactions define the nature and quality of the EL construct. Two main strategies of EL are discussed in the academic literature: “Deep acting” and “surface act­ing” which are further divided into subgroups by different authors (Fouquereau et al., 2019). Surface acting involves emotion simulation by ex­pressing outward appearance like voice tone, gestures, or facial expression. The part of the surface acting when ac­tual feelings are masked and manipulated to show social­ly expected expression represents the “faking emotions” dimension. When an employee does not show his/her true emotions (and not changing how it appears), it refers to “hiding feelings”. Deep acting, in contrast, reflects the aligning of inner feelings by the situation. Sincere emo­tions, authenticity, or expression of genuine emotion is a way of the employee’s natural behavior in various ex­change situations (Grandey, 2000). Different individual attributes, job features as well as organizational characteristics influence the EL strategies of a person. Therefore, affect and LMX may be consid­ered as antecedents of EL. The affective state of an indi­vidual may define his/her EL (Lee & Madera, 2019), and high-quality LMX can positively regulate it (Medler-Li­raz, 2014). Affective events theory (AET) is one of the most frequently used theories in the research addressing emotional constructs, and it suggests that affective expe­riences can impact the emotional labor preferences of em­ployees (Lee & Madera, 2019). EL required in the workplace can have positive or neg­ative effects on employees. Whereas a majority of research stressed the negative influence of EL on subordinates’ psychological condition and work performance (Morris & Feldman, 1996; Wang et al., 2019), some studies un­derlined a positive outcome of EL on job satisfaction as well as on stress level (Xu et al., 2020). In general, extant studies reveal that deep acting promotes largely positive outcomes like job satisfaction, sincerity, or a sense of ful­fillment. On the other side, surface acting triggers mainly adverse outcomes such as emotional exhaustion or dissat­isfaction (Fouquereau et al., 2019). Moreover, EL strate­gies mitigate interpersonal processes within an organiza­tion, including a leader-member exchange. For instance, when individuals experience negative affect in the work­place, how they express or reflect it is very much related to their EL. If employees use deep acting, their NA would reflect their inner processes and their relationships with their leaders in a less destructive manner. If they engage in hiding or faking their emotions, their NA would create stronger negativity regarding their LMX. H3: Daily EL moderates the relationship between dai­ly NA and daily LMX, where deep acting weakens the re­lationship, hiding feelings and faking emotions strengthen the link. 2.4 Emotional Intelligence, Positive Affect, Negative Affect, and Leader-Member Exchange There are two main pillars of research and approach in EI literature. One view conceptualizes EI as an abili­ty and assesses it via objective performance tests such as the performance on solving an emotional problem (e.g., Freudenthaler & Neubauer, 2007). Other approach consid­ers EI as a personality trait and measure it with self-report forms on behavioral tendencies, capacities, or personal­ity characteristics (e.g., Petrides, 2009). A discussion on which operational definition or measurement is more valid has been controversial and discussed among scholars. The former approach criticized ’trait EI’ as being not distinct from personality variables and suggested ‘ability EI’ as a better predictor for achievement. The latter conceptualiza­tion criticized the ‘ability EI’ approach for having some flaws and problems and suggested the ‘trait approach’ to EI as a better alternative. Petrides and Furnham (2001) operationalize these two concepts (trait EI and ability EI) as distinct structures instead of considering them as two different approaches to measure the same variable. They conceptualized trait EI as a composite and distinct person­ality construct that stands for dispositions and behavioral tendencies regarding emotional characteristics. O’Connor and Little (2003)’s and Warwick and Nettelbeck (2004)’s findings support Petrides and Furnham (2001)’s distinction in the operationalization of ability EI and trait EI as they indicate different patterns of correlations for each variable with cognitive ability and personality constructs. Trait EI can be defined as the ability, and disposition individuals perceive about themselves regarding realizing, under­standing, expressing, adapting, and managing emotions effectively (Petrides and Furnham, 2001). In the current study, following Petrides and Furnham (2000, 2001), EI is conceptualized as a personality trait and assessed as a self-report person level (level 2) variable. Despite the disputes in the social science communi­ty about the EI construct and its measurement, this phe­nomenon is widely applied in the relationship as well as leadership studies. The positive relationship between EI and employee engagement, work performance, job satis­faction, and commitment was demonstrated in different studies (Jordan & Troth, 2011; Li et al., 2018; Wen et al., 2019). EI can impact how individuals’ positive or negative affective states during the work processes relate to other outcomes. For instance, EI is shown to intensify the posi­tive affect and decrease the influence of negative affect on organizational citizenship (Miao et al., 2020). Alternative­ly, people with a positive affective state are expected to regulate their EI more effectively. In contrast, a negative mood may reduce the ability of a person to understand and cope with the emotions of others. EI is considered as an essential characteristic of leaders as well as followers in leadership research (Dasborough, 2006; Troth et al., 2018; Tse et al., 2018) and, in general, the development of high-quality social exchange between parties. Additionally, LMX has its moderating/mediating effect on EI and job outcomes (Lee et al., 2018; Ardabili, 2020). As EI refers to the capacity to realize and manage one’s emotions, employees experiencing negative affect can better adjust and control their feelings to avoid negative interactions or impressions to create better LMX. Based on that assumption, this study proposes a moderating impact of EI on the relationship between NA and LMX. H4: Trait EI will moderate the relationship between daily NA and daily LMX; as the level of trait EI increas­es, the negative link between daily NA and daily LMX will be weaker. In other words, daily LMX levels of employees with higher levels of trait EI will be less affected by daily NA. 3 Method 3.1 Participants and Procedure The data for this research were collected from employ­ees working in an administrative department of the public health organization operating in Istanbul, Turkey. A con­venience sampling approach was used. The department is responsible for managing the operational and strategic functions of large-scale public hospital in the European side of Istanbul. The units are divided into three sections such as medical, financial, and administrative services. Personnel working in these units are defined as general administrative services, health services, technical services, and auxiliary services. Participants were informed about the aim and procedure of the research. Anonymity and confidentiality issues regarding the data collection process were underscored. A multilevel research design where daily affect and LMX measures were nested in individu­als was used. A general (one-time) survey that contained questions asking information regarding demographics, the measure for assessing trait EI, and a code to match this sur­vey with day-level surveys were given prior to the shorter day-level questionnaires. Two weeks after the collection of the first questionnaire, the participants were asked to fill the short surveys at the same time every afternoon during five consecutive workdays. The daily surveys collected data on LMX, EL, and negative and positive affect (with the matching code). All surveys were hand-collected in en­velopes by the authors. 73 employees willingly participat­ed in the first, one-time (second-level) survey. The surveys (daily and person-level) were matched by a unique per­sonal code produced by the respondents. 70 participants provided usable five consecutive day-level responses and a person a level response each. We used 350 day-level (70 participants x 5 days) and 70 second-level (personal-level) responses in our analysis (repeated daily measures (days) were nested within individuals). The research model of the study is depicted in figure 1. The mean age of respondents (N= 70) was 35 (min. 23, max. 50), the average tenure of participants was approximately seven (min.1, max.30) years, and 69% were female. About 13% had a high school degree, 14% completed vocational school, 41% completed graduate studies, and 32% completed post-graduate stud­ies. 3.2 Measures We chose the scales to assess the study variables through certain criteria. First, extant studies provided evidence for the reliability and validity of all measures. Second, measures were successfully used in several coun­tries, and they were also validated to Turkish language and context by extant studies. Last, they were consistent and enabling for multilevel analysis. All reliability scores calculated for the scales were acceptable and high as pre­sented under the “analysis and results” title in Table 1. All measures used in the study used a six-point Likert-type scale. All measures except the trait emotional intelligence questionnaire were daily surveys. Day-level scales were adapted (when necessary) with minor changes for captur­ing daily experiences (e.g., adding “today” or changing present tense to past tense). Original forms and slightly modified forms (for day level usage) are given in the ap­pendix. Positive and negative affect were measured using the PANAS scale developed by Watson et al. (1988) and val­idated for Turkey by Gençöz (2000). The original survey contains 20 items to assess NA and PA. Authors often pre­fer to select some of the items from the scales in daily and longitudinal research designs to keep the daily surveys as short as possible (e.g., Tadic et al., 2015; Tadic Vujcic et al., 2017). Items that are more relevant for the job nature of the sample and items that are more likely to happen on a daily basis with higher within-person variability (e.g., Wichers et al., 2012) are usually selected. Items with the highest factor loadings (e.g., Petrou et al., 2012) or items within a sub-cluster of the concept are also preferred (e.g., Ouyang et al., 2019). A review of the extant studies addressing the daily form of the concept also guides the formation of the structure. Following the aforementioned path, PA items were taken from pleasant activated (high activation) pos­itive affect cluster (Watson & Tellegen, 1985; Barrett & Russell, 1998). Also, one item from each cluster (anoth­er classification) of PA (see Egloff et al., 2003), joy (en­thusiastic), interest (strong), and activation (active) were taken for representativeness. Unpleasant activated (high activation) negative affect items (“distressed”, “upset”, “nervous”, and “irritated”) were used for assessing NA. The measure used a six-point scale ranging from “not at all” to “very often”. EL was assessed using the revised version of the Emo­tional Labor Scale developed by Brotheridge and Lee (2003) and revised by Lee and Brotheridge (2011) by changing the construct components from “deep acting” and “surface acting” to “deep acting”, “hiding feelings”, and “faking emotions”. This version was tested for the Turkish context by Dursun et al. (2014). The original scale consists of 9 items (3 items for each dimension). Six items of the initial questionnaire (2 items with the highest factor loadings for each dimension) were used for measuring the variable. Each of the three subscales had two items. The sample items from this scale for deep acting is “Today I made an effort to actually feel the emotions that I need to display to others”, for faking emotions “Today I pretended to have emotions that I didn’t really have”; and for hid­ing feelings “Today I resisted to express my true feelings”. The measure uses a six-point scale ranging from “not at all” to “very often”. Leader-Member Exchange was assessed using items from the LMX-7 scale that is developed by Scandura and Graen (1984). LMX-7 is a frequently used scale in the multilevel longitudinal analysis. Several studies (e.g., Gutermann et al., 2017) choose to use some of the items of the scale instead of giving all of the items due to feasibility reasons in multilevel studies. Three items with the highest factor loadings in the literature and the validation studies in the Turkish context were chosen from the original seven items (e.g., Joseph et al., 2011; Caliskan, 2015). A sample item from this scale is “How well did your leader under­stand your job problems and needs today?”. The measure uses a six-point Likert-type scale ranging from “strongly disagree” to “strongly agree”. In the current study, the re­liabilities calculated for the measure were between .85 - .89 across measurement days, indicating high reliabilities (Table 1). The EI levels of the respondents were assessed by uti­lizing the short form of the Trait Emotional Intelligence Questionnaire. The complete form (153 items) was devel­oped by Petrides and Furnham (2000, 2001), and the short form (30 items) of the scale (Petrides, 2009) was validated for the Turkish by Deniz et al. (2013). As EI was measured in the person level with a one-time longer questionnaire, we used all 30 items (same as the original). Petrides (2009) and Petrides et al. (2010) provided adequate and high reli­ability scores (Cronbach’s Alpha) for the short form scale (.69 and .87). Deniz et al. (2013) reported a .81 reliabil­ity score for the Turkish form of the scale. The measure adopts a six-point Likert scale ranging from “strongly dis­agree” to “strongly agree”. The reliability score calculated for the scale in the present study was .85. In addition, we controlled our results for demographic variables; gender, age, education, and tenure. 4 Results Demographic variables were used as second-level control variables in the current study. 70 respondents pro­vided 350 day level (daily measures of PA, NA, EL, and LMX for five consecutive workdays) and 70 person level responses (EI and demographics). Respondents’ ages were ranging between 23 to 50 (mean=35). Their tenure ranged between 1 to 30. 48 of the 70 respondents were female. 73% completed a university or higher level degree. We calculated reliability scores for the scales and their sub-dimensions that are used in the study. Within-person reliability may be misleading and tends to be high be­cause of repeated measures. As a solution to this problem, some authors (e.g., Beal & Ghandour, 2011) preferred to measure reliabilities of daily measures for each day the surveys were conducted. Following this approach, we cal­culated Cronbach’s alpha scores of day-level variables for each measurement day. Reliabilities across days and var­iables varying from .647 to .924 support the reliability of the day-level measures (Table 1). Besides, the reliability scores for the trait EI scale (person-level) are within the expectable limits. 4.1 Correlations and Descriptive Statistics Day-level and person-level correlations and descriptive statistics are presented in Table 2. Standard deviations and correlations were calculated for person-level and day-lev­el separately. Inter-individual correlations (person-level) were calculated by aggregating the five-day scores for each individual. The person-level variable EI was added by matching aggregated scores (across days) and person-level EI. Intra-individual correlations were calculated via HLM by using the differences in the variance comparing when the level one predictor was included in the model and not included in the model (null model) for explaining the other level-one variable. The aggregated person-level analyses demonstrate that LMX is positively and significantly asso­ciated with PA, Deep acting, and trait EI while negatively and significantly linked with NA. Also, EI is significantly and positively related to PA (aggregated across five days). 4.2 Analysis In the current study, repeated daily measures (days) are nested within individuals (see, Nezlek & Plesko, 2003; Xanthopoulou et al., 2008). To test our hypotheses, we utilized Hierarchical Linear Modelling Software (HLM), which is suitable for analyzing nested data. The Hierar­chical Linear Modeling approach enables researchers to simultaneously test intra and inter-individual associations. Before testing our hypotheses using a multilevel anal­ysis design, we investigated if there was a significant in­tra-individual variance in the daily measured variables of the study. We created null models explaining each daily variable with only intercepts from two levels. The calcula­tion of the percentages of intra-individual variance demon­strated that there was a substantial intra-individual level variance for all day-level variables, justifying and necessi­tating the multilevel analysis to test our hypotheses. 45% to 61% of the variance in the daily measured constructs was within-individuals (Table 3). Partitioning the variance of day-level variables showed that there is a significant amount of variation both for intra-individual and inter-in­dividual levels. 4.3 Tests of Hypotheses Following methodological suggestions in the liter­ature (Peugh & Enders, 2005; Enders & Tofighi, 2007), studies investigating day level repeated data that are nest­ed within-person, adopt group centering approach for the level 1 (day level) variables and grand centering for level 2 (person level) variables (e.g., Ilies et al., 2011; Tadic et al. 2015; Ouyang et al., 2019). Moreover, there is not only one best way for centering in nested models (Nezlek and Allen, 2006). The purpose of the analysis and the research design determine the centering method that should be used in the model. For testing the current study’s gener­al hypotheses, we chose to center level 1 variables on the respective person mean (group centered) and level 2 var­iables on the sample mean (group centered). Group cen­tering level 1 variables allows us to understand the effects of the variance in how participants felt each day. In other words, this centering approach enables us to investigate the effect of variance in an individual’s level of daily affect compared to the mean level of affect the same individu­al experiences on other days (the mean of the individual across measurement days – not the whole sample) on dai­ly LMX. For instance, an employee may be experiencing higher or lower levels of NA/PA compared to others in general; still, what happens when the employee experienc­es higher or level NA/PA compared to his/her own average is another question. This approach allows us to see how does employees’ LMX change on the days they feel higher NA/PA or lower NA/PA (independent from their variation from the sample). Thus, EL structures and daily affect var­iables were group-centered, and the person-level variable EI was grand centered. The dependent level-one variable LMX remained uncentered. Hypothesis 1 suggested a positive link between dai­ly PA and LMX, while hypothesis 2 suggested a negative association between daily NA and LMX. Hypothesis 3 predicted that EL moderates the relationship between NA and LMX. To test these hypotheses, a model with only person-level control variables, gender, age, tenure, educa­tion, and trait EI was created in the first step (Table 4). EI was entered as grand mean-centered. This model revealed that EI had a significant cross-level main effect on LMX (B=0.40, p<0.05). None of the demographic control var­iables had any significant effects on the independent var­iable. In the second model, we added day-level variables with previous control variables. All day-level variables were group-centered. This way, by eliminating the poten­tial effects of the difference between participants’ means, the analysis demonstrated how the day-level intra-person change in the participants’ NA, PA, and EL was associ­ated with daily, intra-person changes in the LMX expe­rience. PA significantly predicted LMX in the proposed way (B=0.43, p<0.001), supporting the first hypothesis of the study. NA also demonstrated a positive and significant association with LMX (B=0.10, p<0.05) in the opposite direction proposed by hypothesis 2. None of the EL di­mensions demonstrated significant direct effects on LMX. In model three, we tested the cross-level moderator effect of EI on PA and NA slopes. Trait EI showed no signifi­cant moderator effects. Hypothesis 4 was not supported. For level-one interaction effects (for the third hypothesis that suggests EL as a moderator on the relationship be­tween NA and LMX), we added product terms of NA and all EL dimensions. We multiplied standardized scores of each variable for creating product terms to avoid possible multicollinearity problems (Aiken &West, 1991); thus we entered them as uncentered (as they were already centered) in the model. The interaction terms indicated no signifi­cant moderation. The moderation hypotheses regarding day-level variables were rejected for the intra-individual variance. Results of correlation analysis demonstrated a negative link between NA and LMX on the aggregated level. Taken together with the results of regression analysis (through within-person variation), employees with higher NA levels (compared to others) have lower levels of LMX while they have higher levels of LMX on the days they feel higher NA (compared to their own NA mean). On the grounds of this pattern of results and research suggesting nonlinear links between NA and LMX (e.g., Hochwarter, 2005), we tested possible curvilinear effects of NA by adding NA square in the model after controlling for NA and second-level vari­ables. U-shaped relationship regarding NA and LMX was not significant. Still, the curve estimation of the quadratic model for NA was significant. We tested whether the re­lationships regarding NA change when respondents expe­rience high levels and low levels of NA. We created two different files by splitting through the responses lower and higher than the median of NA. For the low NA group, NA did not relate with LMX significantly. For the high NA group, NA was associated negatively and significantly with LMX. As we split the file through the occasions (not people) we used grand centering for testing the moderator hypotheses regarding the high-level NA group. The results were summarized in Table 5. NA was negatively related to LMX. EI showed no moderator effect on this relationship but had a direct cross-level effect on LMX. Deep acting and hiding feelings significantly moderated the relation­ship between NA and LMX. To understand the nature of the moderation, we graphi­cally depicted interaction effects by using one standard de­viation above and below the mean for variables. The plots demonstrate that the negative relationship is stronger when participants experienced lower levels of deep acting and higher levels of hiding feelings. 5 Discussion and Conclusion 5.1 Discussion This study investigated the association between daily affect and daily LMX and the moderator roles of EI and EL on this relationship. Seventy participants provided 70 per­son-level data (via a survey assessing demographics and EI) and 350 day level data (through a survey measuring daily forms of NA, PA, EL, and LMX for five consecu­tive days). Hypotheses were tested in a multilevel setting where the days were nested in employees. The multilevel analysis results using group-centered level-one variables indicated that both PA and NA were positively related to the dependent variable. This means, on the days employees experienced higher levels of PA and NA, they also perceived higher levels of LMX. In other words, higher levels of daily NA and PA (compared to the mean of individuals’ affect level across five days) increased the daily LMX experience. The PA part of these findings supports our propositions and is consistent with the extant literature (e.g., Hochwarter, 2005; Liu et al., 2020). How­ever, the NA part of these findings contrasts with our prop­ositions and most of the present research (e.g., Sniderman et al., 2016). This pattern of findings regarding the NA-LMX association can be explained by the Conservation of Resources Theory (Hobfoll, 2011) as the theory suggests that employees try to seek, increase and sustain resourc­es in demanding situations. On the days employees feel higher NA levels, they seek quality leadership interaction to alter their resources. The study measured affect that can be caused by any life events and towards anything instead of solely assessing affect towards the leader. Employees can use LMX as a coping mechanism (Major & Morgan­son, 2011) to reduce their NA produced by various factors. Besides, most studies that have addressed the relationship between NA and LMX did not adopt a multilevel with­in-person research design. Instead, they analyzed the be­tween-person variance addressing the one-time measured general experience of NA and the general perception of LMX. Hence, the current study findings regarding the per­son-level (aggregated) NA and LMX are consistent with the general stream of findings in the literature (as they in­dicate a negative association). Moreover, investigation of daily intra-individual relationships between the constructs remains understudied, and findings of such research can provide new perspectives on the matter. Findings underscore the fact that between-person and intra-person associations may operate in different ways. For instance, an employee may experience high-quality LMX in the days he/she feels comparatively higher af­fect, regardless of his/her between-person general affect or LMX level. The same person may face low-quality LMX than others when average scores are compared (inter-per­sonal level). Thus, between-person and intra-person var­iances can show different outcomes. Using a multilevel approach enabled us to investigate both variances that pro­vide essential insights into how affect is related to LMX. Therefore, we also tested the relationships regarding the between-employee variance. The analysis conducted with the aggregated scores of days for each employee showed a pattern of result that is opposite to the findings gathered through within-person variances (group-mean centering). Employees experience lower levels of LMX in gener­al when they have higher NA levels compared to others (person-level). In contrast, on the within-person level, they have higher levels of LMX on the days they experience higher NA. Trait EI shows a direct positive cross-level im­pact on day-level LMX in this study. The employees with higher Trait EI experience significantly higher levels of daily LMX. However, despite its direct effect, Trait EI has no significant influence on the affect-LMX slopes. The ex­tant EI literature either addresses LMX as an antecedent (e.g., Clarke & Mahadi, 2011) or focuses on the leader’s EI (e.g., Humphrey, 2012). Therefore, the current study’s findings serve as a platform and a call for future research to enlighten the role of followers’ EI on the affect-LMX association. None of the EL constructs demonstrated any signifi­cant moderating effects in the preliminary analysis. How­ever, when we tested for high NA and low NA groups sep­arately, deep acting and hiding feelings showed significant moderation between NA and LMX in the proposed way. The independent testing of high and low affect levels may further enhance our understanding of the nature of their influence on the outcomes (e.g., Hochwarter, 2005). We tested the relationship between variables on the days NA was higher than the sample average, and on the days it was lower. The negative relationship was stronger when participants experienced lower levels of deep acting and higher levels of hiding feelings for the high NA sample. These findings align with the existing studies that report­ed significant links between EL and LMX (e.g., Liden & Maslyn, 1998). 5.2 Theoretical and Practical Implications This investigation contributes to current literature in several ways. First, the study furthers the knowledge on the affect-LMX relationship by shifting the search of cau­sality by casting affect an antecedent role. It reveals that daily affective states significantly influence daily LMX. This is important to provide a platform and a call for more research to shed light on the affect-to-LMX pathway. Sec­ond, the study further elaborates the complex nature of the association between concepts as it demonstrates that they relate differently on the person and day level. This is critical because the extant knowledge regarding this rela­tionship is dominantly a product of the general stream of findings addressing concepts as static structures and ne­glecting intra-individual variance. The current study also enhances the information about the affect-LMX link by demonstrating that high and low levels of affect can have different natures of associations with LMX. Moreover, the significance of the moderating roles of EL constructs can also be different in high and low levels of affect. These findings provide significant managerial impli­cations. Enhancing our knowledge of the antecedents of LMX is crucial for the performance of public health in­stitutions. Recent studies conducted in the Turkish con­text (e.g., Ürek & Ugurluoglu, 2019) report findings that associate LMX with positive organizational outcomes in public healthcare organizations. The nature of the job and the workplace can elevate the levels of affect experienced by public health care professionals. Findings of the cur­rent study indicate that employees can seek and engage in higher levels of exchange with their leaders on the days they experience higher levels of both positive and nega­tive affect. Leaders should acknowledge that and respond with higher leadership support and involvement. However, employees’ general levels of affect act differently where­as positive affect increases LMX on the person-level, and negative affect is reversely related to LMX. Besides, the intensity of negative affect can impact LMX in different ways, and these effects are regulated by EL. When neg­ative affect is high, employees experience lower LMX if they choose to hide their emotions. Contrary, deep acting weakens the adverse association between negative affect and LMX. Therefore, efforts encouraging deep acting can mitigate the undesirable impact of negative affect on LMX. The findings of the present study underscore the HRM practices that address the usage of more positive emotional regulation styles, such as training and development efforts and workplace counseling. Also, strengthening leadership development processes for fostering leadership styles that would provide better LMX quality by including daily emo­tions such as emphatic leadership (Kock et al., 2019) can serve as a significant HRM policy. Creating a positive cli­mate where employees do not hide their feelings and sup­porting the psychological capacities of employees for con­structive regulation of emotions can alleviate the effects of negative affect on LMX. Given the positive cross-level effects trait EI on daily LMX, investing in the EI capacities and including EI as a selection criterion in recruitment pro­cesses can enhance LMX in organizations. 5.3 Limitations and Directions for Future Research The study data was provided by the employees work­ing in the public health organization in Turkey. Due to difficulties in data collection for daily multilevel research designs, we could not enlarge our sample to other work settings. Readers should consider this for the generaliz­ability of the findings. In a cross-cultural study (among ten countries), Aycan et al. (2000) presented that manag­ers’ assumptions regarding HRM practices and employee characteristics are significantly affected by the culture. The study also revealed that Turkey depicts a highly pa­ternalistic culture that can affect perceptions such as loy­alty and malleability, which are essential aspects of LMX development. Thus, the findings of the current should be considered with the cultural characteristics of the sample. Tightly-knit and close relational characteristics build on loyalty and paternalism can strengthen the bonds between leaders and members, fostering the crossover or effects of daily affect on LMX. All the variables analyzed in this research were as­sessed through self-rating measures. Although using group-centered forms of these variables eliminated the between-person variance, self-rating may create biases, especially for grand-mean centered forms of the variables. Despite two week-gap between the person-level gener­al questionnaire and the first day-level questionnaire, all day-level measures were given daily together. As the scale was kept very short and daily research designs are demand­ing for respondents, we did not separate the daily measures (for each variable) or give them at different times during the days. This can create suspicions for common method bias. For that reason, we conducted Harman’s single factor test. When daily measures were allowed to factor freely, they explained 68% of the variance, and when they were forced to factor under one factor, they could only explain 27% of the variance. One factor structure explaining less than 50% (threshold) of the variance indicates no problems regarding common method bias (Eichhorn, 2014). The study data was collected from one unit; therefore, team-level variances could not be reflected in our research design. For further studies, three-level research designs in­cluding organizational units may enhance our knowledge of the relationship between affect and LMX. 5.4 Conclusion The investigation of LMX in public healthcare organi­zations is crucial because LMX serves as a significant pre­dictor for several organizational and individual outcomes. The affective pathway to LMX is significant and complex. Affect relates to LMX differently at intra-individual and inter-individual levels. Besides, high and low affect lev­els can demonstrate different patterns of associations with LMX when the affective moderators are included in the model. The findings of the current study underscore the so­phisticated nature of the affect-LMX relationship and the significance of further research on the matter. Literature 1Received: 4th February 2021; revised: 6th May 2021; accepted: 11th May 2021 Figure 1: The Research Model of the Study Table 1: Reliabilities of the measures Day LMX Deep Acting Faking Emotions Hiding Feelings Emotional Labor Negative Affect Positive Affect Trait Emotional Intelligence 1 ,857 ,860 ,924 ,778 ,860 ,737 ,843 - 2 ,861 ,903 ,815 ,807 ,878 ,809 ,778 - 3 ,855 ,889 ,757 ,647 ,873 ,792 ,756 - 4 ,882 ,859 ,754 ,734 ,870 ,801 ,712 - 5 ,857 ,889 ,861 ,767 ,871 ,809 ,860 - Total ,861 ,879 ,751 ,824 ,870 ,792 ,802 ,851 Note: reliabilities for day-level variables are calculated for each measurement day. Total reliabilities stand for reliabilities calculated for all day-level data (N=350). Trait emotional intelligence is measured for one time (person-level). Table 2: Descriptive statistics and between and within-individual correlations among study variables Mean SDw SDb 1 2 3 4 5 6 7 1. Leader-Member Exchange 3,60 1,19 ,89 1 ,002 ,149** -,028 -,218** ,418** ,252* 2. Faking Emotions 2,28 1,13 ,93 ,052 1 ,565** ,699** ,337** -,077 -,090 3. Deep Acting 2,99 1,32 1,10 ,005 ,245** 1 ,482** ,105 ,071 ,015 4. Hiding Feelings 2,43 1,19 ,98 -,053 ,296** ,194** 1 ,372** -,142** -,208 5. Negative Affect 2,59 1,08 ,83 -,040 ,122 -,046 ,133* 1 -,363** -,170 6. Positive Affect 4,04 1,10 ,83 ,322** -,043 -,022 -,047 -,311** 1 ,442** 7. Trait Emotional Intelligence (L2) 4,48 - ,63 - - - - - - 1 Note: Correlations above the diagonal depict person-level correlations (means across days, N=70). Correlations below the diagonal show day-level correlations calculated via HLM (N=350). SDw= Within-person standard deviation, SDb= Between-person standard deviation. LMX= Leader-Member Exchange. L2= Level 2 *p<0.05 and **p<0.01 Table 3: Partitioning variance components of within-person variables Variable Intercept Within-person variance Between-person variance Percent of Within-person variance Leader-Member Exchange 3.60** 0.64 0.79 45 Positive Affect 4.04** 0.55 0.66 45 Negative Affect 2.59** 0.57 0.61 48 Faking Emotions 2.28** 0.75 0.53 59 Deep Acting 2.99** 1.07 0.69 61 Hiding Feelings 2.43** 0.92 0.76 55 The percent of within-person variance was as computed as s2/(s2 + t 00). **p < 0.01. Table 4: Multilevel estimates for models predicting the daily LMX Model 1 Model II Model III Model IV Variable Est SE T Est SE T Est SE T Est SE t Intercept 4.76 0.59 8.00** 4.76 0.59 8.00** 4.76 0.59 8.00** 4.73 0.58 8.11** Trait Emotional Intelligence 0.40 0.16 2.43* 0.40 0.16 2.43* 0.40 0.16 2.43* 0.40 0.16 2.50* Faking Emotions -0.04 0.09 -0.49 -0.04 0.09 -0.53 -0.02 0.08 -0.32 Deep Acting 0.05 0.07 0.75 0.05 0.07 0.74 0.05 0.07 0.82 Hiding Feelings -0.01 0.06 -0.25 0.02 0.06 0.32 0.00 0.07 0.07 Negative Affect 0.10 0.04 2.42* 0.10 0.04 2.21* 0.10 0.04 2.03* Positive Affect 0.43 0.05 7.72** 0.42 0.05 7.82** 0.44 0.05 7.98** Trait Emotional Intel­ligence on Positive Affect slope -0.06 0.07 -0.88 Trait Emotional Intel­ligence on Negative Affect slope -0.00 0.06 -0.06 Negative Affect*Fak­ing Emotions -0.04 0.04 -0.95 Negative Af­fect*Deep Acting 0.06 0.05 1.18 Negative Affect*Hid­ing Feelings 0.06 0.06 -1.07 Notes: N=All level 1 variables (except the dependent variable) were group centered relative to the individuals’ means. Level 2 variables were entered grand mean. Interaction variables were standardized before multiplication. **_p<0.01, *p<0.05 Table 5: Multilevel estimates for predicting LMX for high NA cluster Variable Est T Main effects Intercept 4.51 8.76** Faking Emotions -0.18 -0.89 Deep Acting 0.09 0.05 Hiding Feelings 0.31 1.60 Negative Affect 0.29 -2.16* Cross Level Interaction Trait Emotional Intelligence effect on Intercept 0.53 3.51** Negative Affect slope -0.01 -0.11 Level 1 Interaction Negative Affect*Faking Emotions 0.06 0.45 Negative Affect*Deep Acting 0.33 2.28* Negative Affect*Hiding Feelings -0.36 -2.03* Note: Analysis conducted in high-level Negative Affect group (responses higher than the median of Negative Affect across the sample). All predictor scores were grand mean centered to include interpersonal variance. *p<0.05, **p<0.01. Figure 2: The first level moderating effect of deep acting for high NA sample. Figure 3: The first level moderating effect of hiding feelings for high NA sample Aiken, L. S., West, S. G., & Reno, R. R. (1991). Multiple regression: Testing and interpreting interactions. sage. Ardabili, F. S. (2020). Moderating-mediating Effects of Leader Member Exchange, Self-efficacy and Psy­chological Empowerment on Work Outcomes among Nurses. Organizacija, 53(3), 246-258. https://doi.org/10.2478/orga-2020-0016 Ashton-James, C. E., & Ashkanasy, N. M. (2008). Chap­ter 1 Affective events theory: a strategic perspective. In Research on Emotion in Organizations, 4(1), 1–34. https://doi.org/10.1016/S1746-9791(08)04001-7 Aycan, Z., Kanungo, R., Mendonca, M., Yu, K., Deller, J., Stahl, G., & Kurshid, A. (2000). Impact of culture on human resource management practices: A 10-coun­try comparison. Applied Psychology, 49(1), 192-221. https://doi.org/10.1111/1464-0597.00010 Barrett, L. F., & Russell, J. A. (1998). Independence and Bipolarity in the Structure of Current Affect. Journal of Personality and Social Psychology, 74(4), 967-984. https://doi.org/10.1037/0022-3514.74.4.967 Beal, D. J., & Ghandour, L. (2011). Stability, change, and the stability of change in daily workplace affect.  Jour­nal of Organizational Behavior, 32(4), 526-546. https://doi.org/10.1002/job.713 Bener, A., N. Alayoglu, F. Çatan, P. Torun, and E. S. Yil­maz. (2019). Health Services Management in Turkey: Failure or Success?. International journal of preven­tive medicine 10(2019): 30. https://doi.org/10.4103/ijpvm.ijpvm_422_17 Bernerth, J. B., Armenakis, A. A., Feild, H. S., Giles, W. F., & Jack Walker, H. (2007). Is personality associated with perceptions of LMX? An empirical study. Leader­ship & Organization Development Journal, 28(7), 613–631. https://doi.org/10.1108/01437730710823879 Brief, A. P., & Weiss, H. M. (2002). Organizational Be­havior: Affect in the Workplace. Annual Review of Psychology, 53(1), 279–307. https://doi.org/10.1146/annurev.psych.53.100901.135156 Brotheridge, C. M., & Lee, R. T. (2003). Development and validation of the emotional labour scale. Journal of Oc­cupational and Organizational Psychology, 76(3), 365-379. https://doi.org/10.1348/096317903769647229 Caliskan, G. (2015). An examination of coach and player relationships according to the adapted LMX 7 scale: A validity and reliability study. Measurement in Physical Education and Exercise Science, 19(1), 22-33. https://doi.org/10.1080/1091367x.2014.977996 Clarke, N., & Mahadi, N. (2011). Chapter 9 Emotional Intelligence as a Moderator of the Quality of Lead­er–Member Exchange and Work-Related Outcomes. In What Have We Learned? Ten Years On  (pp. 227-254). Emerald Group Publishing Limited. https://doi.org/10.1108/s1746-9791(2011)0000007014 Costakis, H. R., Gruhlke, H., & Su, Y. (2021). Implications of emotional labor on work outcomes of service work­ers in not-for-profit human service organizations. Hu­man Service Organizations: Management, Leadership & Governance, 45(1), 29-48. https://doi.org/10.1080/23303131.2020.1818157 Cropanzano, R., Anthony, E. L., Daniels, S. R., & Hall, A. V. (2017a). Social exchange theory: A critical re­view with theoretical remedies. Academy of Manage­ment Annals, 1(1), 479-516. https://doi.org/10.5465/annals.2015.0099 Cropanzano, R., Dasborough, M. T., & Weiss, H. M. (2017b). Affective events and the development of leader-member exchange. Academy of Management Review, 42(2), 233–258. https://doi.org/10.5465/amr.2014.0384 Dansereau, F., Graen, G., & Haga, W. J. (1975). A vertical dyad linkage approach to leadership within formal or­ganizations. Organizational Behavior and Human Per­formance, 13(1), 46–78. https://doi.org/10.1016/0030-5073(75)90005-7 Dasborough, M. T. (2006). Cognitive asymmetry in em­ployee emotional reactions to leadership behaviors. Leadership Quarterly, 17(2), 163–178. https://doi.org/10.1016/j.leaqua.2005.12.004 Deniz, M. E., Özer, E., & Isik, E. (2013). Trait emotional intelligence questionnaire-short form: Validity and re­liability studies. Egitim ve Bilim, 38(169). Dienesch, R. M., & Liden, R. C. (1986). Leader-Member Exchange Model of Leadership: A Critique and Further Development. Academy of Management Review, 11(3), 618–634. https://doi.org/10.5465/amr.1986.4306242 Dulebohn, J. H., Bommer, W. H., Liden, R. C., Brouer, R. L., & Ferris, G. R. (2012). A Meta-Analysis of An­tecedents and Consequences of Leader-Member Ex­change. Journal of Management, 38(6), 1715–1759. https://doi.org/10.1177/0149206311415280 Dundar, M., Uzak, A. S., & Karabulut, Y. (2010). Health­care in overview of Turkey. EPMA Journal, 1(4), 587-594. https://doi.org/10.1007/s13167-010-0049-7 Dursun, S., Aytaç, S., & Bayram, N. (2014). Duygusal Emek Ölçegi Türkçe Formunun Geçerlilik Ve Güvenil­irlik Çalismasi. (Validation and Reliability Study of the Turkish Form of the Emotional Labor Scale) Is-Güç Endüstri Iliskileri ve Insan Kaynaklari Dergisi, 1683, 10-18. https://doi.org/10.4026/1303-2860.2014.0252 Egloff, B., Schmukle, S. C., Burns, L. R., Kohlmann, C. W., & Hock, M. (2003). Facets of dynamic positive affect: differentiating joy, interest, and activation in the positive and negative affect schedule (PANAS). Jour­nal of personality and social psychology, 85(3), 528. https://doi.org/10.1037/0022-3514.85.3.528 Eichhorn, B. R. (2014). Common method variance tech­niques. Cleveland State University, Department of Operations & Supply Chain Management. Cleveland, OH: SAS Institute Inc, 1-11. Ellis, A. M., Bauer, T. N., Erdogan, B., & Truxillo, D. M. (2019). Daily perceptions of relationship quality with leaders: implications for follower well-being. Work and Stress, 33(2), 119–136. https://doi.org/10.1080/02678373.2018.1445670 Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in cross-sectional multilevel models: A new look at an old issue. Psychological Methods, 12, 121–138. https://doi.org/10.1037/1082-989x.12.2.121 Fouquereau, E., Morin, A. J. S., Lapointe, É., Mokounko­lo, R., & Gillet, N. (2019). Emotional labour profiles: Associations with key predictors and outcomes. Work and Stress, 33(3), 268-294. https://doi.org/10.1080/02678373.2018.1502835 Freudenthaler, H. H., & Neubauer, A. C. (2007). Mea­suring emotional management abilities: Further evi­dence of the importance to distinguish between typ­ical and maximum performance. Personality and Individual Differences,  42(8), 1561-1572. https://doi.org/10.1016/j.paid.2006.10.031 Gençöz, T. (2000). Positive and Negative Affect Schedule: A study of validity and reliability. Türk Psikoloji Der­gisi. Glasø, L., Skogstad, A., Notelaers, G., & Einarsen, S. (2018). Leadership, affect and outcomes: symmetrical and asymmetrical relationships. Leadership and Orga­nization Development Journal. https://doi.org/10.1108/LODJ-08-2016-0194 Grandey, A. A. (2000). Emotion regulation in the work­place: a new way to conceptualize emotional labor. Journal of Occupational Health Psychology, 5(1), 95–110. https://doi.org/10.1037//1076-8998.5.1.95 Gutermann, D., Lehmann-Willenbrock, N., Boer, D., Born, M., & Voelpel, S. C. (2017). How leaders affect followers’ work engagement and performance: Inte­grating leader- member exchange and crossover the­ory. British Journal of Management, 28(2), 299-314. https://doi.org/10.1111/1467-8551.12214 Hobfoll, S. E. (2011). Conservation of resource caravans and engaged settings. Journal of occupational and or­ganizational psychology, 84(1), 116-122. https://doi.org/10.1111/j.2044-8325.2010.02016.x Hochschield, A. R. (1983). The Managed Heart. Working in America: Continuity, Conflict, and Change in a New Economic Era. Hochwarter, W. (2005). LMX and job tension: Linear and non-linear effects and affectivity. Journal of Busi­ness and Psychology, 19(4), 505–520. https://doi.org/10.1007/s10869-005-4522-6 Hofmans, J., Dóci, E., Solinger, O. N., Choi, W., & Judge, T. A. (2019). Capturing the dynamics of leader–fol­lower interactions: Stalemates and future theoretical progress. Journal of Organizational Behavior, 40(3), 382-385. https://doi.org/10.1002/job.2317 Humphrey, R. H. (2012). How do leaders use emotional la­bor?. Journal of Organizational Behavior, 33(5), 740-744. https://doi.org/10.1002/job.1791 Ilies, R., Johnson, M. D., Judge, T. A., & Keeney, J. (2011). A within-individual study of interpersonal conflict as a work stressor: Dispositional and situational moder­ators. Journal of Organizational Behavior, 32(1), 44-64. https://doi.org/10.1002/job.677 Jordan, P. J., & Troth, A. (2011). Emotional intelligence and leader member exchange: The relationship with em­ployee turnover intentions and job satisfaction. Lead­ership and Organization Development Journal, 32(3), 260–280. https://doi.org/10.1108/01437731111123915 Joseph, D. L., Newman, D. A., & Sin, H. P. (2011). Lead­er–Member Exchange (LMX) measurement: evidence for consensus, construct breadth, and discriminant va­lidity. In Building methodological bridges. Emerald Group Publishing Limited. https://doi.org/10.1108/s1479-8387(2011)0000006012 Kock, N., Mayfield, M., Mayfield, J., Sexton, S., & De La Garza, L. M. (2019). Empathetic leadership: How leader emotional support and understanding influenc­es follower performance. Journal of Leadership & Organizational Studies, 26(2), 217-236. https://doi.org/10.1177/1548051818806290 Lee, L., & Madera, J. M. (2019). A systematic literature review of emotional labor research from the hospi­tality and tourism literature. International Journal of Contemporary Hospitality Management, 31(7), 2808-2826. https://doi.org/10.1108/IJCHM-05-2018-0395 Lee, M., Mayfield, C. O., Hinojosa, A. S., & Im, Y. (2018). A Dyadic Approach to Examining the Emotional Intel­ligence–Work Outcome Relationship: The Mediating Role of LMX. Organization Management Journal, 15(1), 1-16. https://doi.org/10.1080/15416518.2018.1427539 Lee, R. T., & Brotheridge, C. M. (2011). Words from the heart speak to the heart. Career Develop­ment International, 16(4), 401-420. https://doi.org/10.1108/13620431111158805 Li, M., Pérez-Díaz, P. A., Mao, Y., & Petrides, K. V. (2018). A Multilevel Model of Teachers’ Job Perfor­mance: Understanding the Effects of Trait Emotion­al Intelligence, Job Satisfaction, and Organizational Trust. Frontiers in Psychology, 9, 1–13. https://doi.org/10.3389/fpsyg.2018.02420 Liden, R. C., & Maslyn, J. M. (1998). Multidimension­ality of leader-member exchange: An empirical as­sessment through scale development. Journal of Management, 24(1), 43-72. https://doi.org/10.1016/s0149-2063(99)80053-1 Liu, C., Xu, D., Wang, C., Zhang, C., & Qi, W. (2020). Can Positive Emotion Benefit to Leader-Member Ex­change? The Role of Leader Identification. In 2020 International Conference on Wireless Communica­tions and Smart Grid (ICWCSG) (pp. 412-415). IEEE. https://doi.org/10.1109/icwcsg50807.2020.00094 Major, D. A., & Morganson, V. J. (2011). Coping with work-family conflict: A leader-member exchange per­spective. Journal of Occupational Health Psycholo­gy, 16(1), 126-138 https://doi.org/10.1037/a0021727 Mayer, J. D., & Salovey, P. (1997). What is Emotional In­telligence? In P. and Salovey & D. J. Sluyter (Eds.), Emotional Development and Emotional Intelligence: Educational Implications (pp. 3–31). New York: Basic Books. https://doi.org/10.1037/t05047-000 Medler-Liraz, H. (2014). Negative affectivity and tipping: The moderating role of emotional labor strategies and leader-member exchange. International Journal of Hospitality Management, 36, 63–72. https://doi.org/10.1016/j.ijhm.2013.08.010 Miao, C., Humphrey, R. H., & Qian, S. (2020). The cross-cultural moderators of the influence of emotional intelligence on organizational citizenship behavior and counterproductive work behavior. Human Resource Development Quarterly, 31(2), 213-233. https://doi.org/10.1002/hrdq.21385 Morris, J. A., & Feldman, D. C. (1996). The dimen­sions, antecedents, and consequences of emotional labor. Academy of Management Review,  21(4), 986-1010. https://doi.org/10.5465/amr.1996.9704071861 Nahrgang, J. D., & Seo, J. J. (2016). How and Why High Leader–Member Exchange (LMX) Relationships De­velop: Examining the Antecedents of LMX. In The Ox­ford Handbook of Leader-Member Exchange. https://doi.org/10.1093/oxfordhb/9780199326174.013.3 Nezlek, J. B., & Allen, M. R. (2006). Social support as a moderator of day-to-day relationships between dai­ly negative events and daily psychological well-be­ing. European Journal of Personality: Published for the European Association of Personality Psycholo­gy, 20(1), 53-68. https://doi.org/10.1002/per.566 Nezlek, J. B., & Plesko, R. M. (2003). Affect-and self-based models of relationships between daily events and daily well-being. Personality and So­cial Psychology Bulletin, 29(5), 584-596. https://doi.org/10.1177/0146167203029005004 O’Connor Jr, R. M., & Little, I. S. (2003). Revisiting the predictive validity of emotional intelligence: Self-re­port versus ability-based measures. Personality and Individual differences, 35(8), 1893-1902. https://doi.org/10.1016/s0191-8869(03)00038-2 Ouyang, K., Cheng, B. H., Lam, W., & Parker, S. K. (2019). Enjoy your evening, be proactive tomorrow: How off-job experiences shape daily proactivity. Jour­nal of Applied Psychology, 104(8), 1003. https://doi.org/10.1037/apl0000391 Petrides, K. V. (2009). Psychometric properties of the trait emotional intelligence questionnaire (TEIQue). In As­sessing emotional intelligence (pp. 85-101). Springer, Boston, MA. 85-101 https://doi.org/10.1007/978-0-387-88370-0_5 Petrides, K. V., & Furnham, A. (2000). On the dimen­sional structure of emotional intelligence. Personality and Individual Differences, 29, 313-320. https://doi.org/10.1016/s0191-8869(99)00195-6 Petrides, K. V., & Furnham, A. (2001). Trait emotional in­telligence: Psychometric investigation with reference to established trait taxonomies. European Journal of Personality, 15(6), 425-448. https://doi.org/10.1002/per.416 Petrides, K. V., Vernon, P. A., Schermer, J. A., Ligthart, L., Boomsma, D. I., & Veselka, L. (2010). Relation­ships between trait emotional intelligence and the Big Five in the Netherlands. Personality and Individual Differences, 48, 906-910. https://doi.org/10.1016/j.paid.2010.02.019 Petrou, P., Demerouti, E., Peeters, M. C., Schaufeli, W. B., & Hetland, J. (2012). Crafting a job on a daily basis: Contextual correlates and the link to work en­gagement. Journal of Organizational Behavior, 33(8), 1120-1141. https://doi.org/10.1002/job.1783 Peugh, J. L., & Enders, C. K. (2005). Using the SPSS mixed procedure to fit cross-sectional and longitudinal multi­level models. Educational and Psychological Measure­ment, 65(5), 717–741. doi:10.1177/0013164405278558 Scandura, T. A., & Graen, G. B. (1984). Moderating ef­fects of initial leader–member exchange status on the effects of a leadership intervention. Journal of applied psychology, 69(3), 428. https://doi.org/10.1037/0021-9010.69.3.428 Shickle, D., Day, M., Smith, K., Zakariasen, K., Moskol, J., & Oliver, T. (2014). Mind the public health leader­ship gap: the opportunities and challenges of engag­ing high-profile individuals in the public health agen­da. Journal of Public Health, 36(4), 562-567. https://doi.org/10.1093/pubmed/fdu003 Sniderman, P., Fenton-O’Creevy, M., & Searle, R. (2016). Effects of managerial communication as moderated by LMX and trait NA. Journal of Managerial Psycholo­gy, 31(6), 1074-1090. https://doi.org/10.1108/jmp-04-2015-0120 Tadic Vujcic, M., Oerlemans, W. G., & Bakker, A. B. (2017). How challenging was your work today? The role of autonomous work motivation. European Jour­nal of Work and Organizational Psychology,  26(1), 81-93. https://doi.org/10.1080/1359432x.2016.1208653 Tadic, M., Bakker, A. B., & Oerlemans, W. G. (2015). Challenge versus hindrance job demands and well-be­ing: A diary study on the moderating role of job re­sources. Journal of Occupational and Organizational Psychology,  88(4), 702-725. https://doi.org/10.1111/joop.12094 Tellegen, A., Watson, D., & Clark, L. A. (1999). On the Dimensional and Hierarchical Structure of Affect. Psychological Science, 10(4), 297–303. https://doi.org/10.1111/1467-9280.00157 Troth, A. C., Lawrence, S. A., Jordan, P. J., & Ashkanasy, N. M. (2018). Interpersonal Emotion Regulation in the Workplace: A Conceptual and Operational Review and Future Research Agenda. International Journal of Management Reviews, 20(2), 523–543. https://doi.org/10.1111/ijmr.12144 Tse, H. H. M., Dasborough, M. T., & Ashkanasy, N. M. (2008). A multi-level analysis of team climate and interpersonal exchange relationships at work.  The Leadership Quarterly, 19(2), 195-211. https://doi.org/10.1016/j.leaqua.2008.01.005 Tse, H. H. M., Troth, A. C., Ashkanasy, N. M., & Collins, A. L. (2018). Affect and leader-member exchange in the new millennium: A state-of-art review and guid­ing framework. Leadership Quarterly, 29(1), 135–149. https://doi.org/10.1016/j.leaqua.2017.10.002 Uhl-Bien, M., Riggio, R. E., Lowe, K. B., & Carsten, M. K. (2014). Followership theory: A review and research agenda. The Leadership Quarterly, 25(1), 83–104. https://doi.org/10.1016/j.leaqua.2013.11.007 Ürek D., & Ugurluoglu Ö. (2019). Examining the Impact of Leader–Member Exchange on Organizational Cit­izenship Behaviour in Healthcare Sector. Is ve Insan Dergisi, 6(1), 81-92. Wang, H., Hall, N. C., & Taxer, J. L. (2019). Antecedents and Consequences of Teachers’ Emotional Labor: a Systematic Review and Meta-analytic Investigation. Educational Psychology Review. 1-36 https://doi.org/10.1007/s10648-019-09475-3 Warwick, J., & Nettelbeck, T. (2004). Emotional intel­ligence is…?. Personality and Individual Differ­ences, 37(5), 1091-1100. https://doi.org/10.1016/j.paid.2003.12.003 Watson, D., & Tellegen, A. (1985). Toward a consensual structure of mood. Psychological Bulletin, 98(2), 219. https://doi.org/10.1037/0033-2909.98.2.219 Watson, D., Clark, L. A., & Carey, G. (1988). Positive and Negative Affectivity and Their Relation to Anxiety and Depressive Disorders. Journal of Abnormal Psychol­ogy, 97(3), 346–353. https://doi.org/10.1037/0021-843X.97.3.346 Weiss, H. M., & Cropanzano, R. S. (1996). Affective events theory: A theoretical discussion of the structure, causes and consequences of affective experiences at Work. In Research in organizational Behavior, 18(1), 1–74). https://doi.org/1-55938-938-9 Wen, J., Huang, S., & Hou, P. (2019). Emotional intel­ligence, emotional labor, perceived organizational support, and job satisfaction: A moderated mediation model. International Journal of Hospitality Man­agement, 81(December 2018), 120–130. https://doi.org/10.1016/j.ijhm.2019.01.009 Wichers, M., Peeters, F., Rutten, B. P., Jacobs, N., Derom, C., Thiery, E., ... & van Os, J. (2012). A time-lagged momentary assessment study on daily life physical activity and affect. Health Psychology, 31(2), 135. https://doi.org/10.1037/a0025688 Xanthopoulou, D., Baker, A. B., Heuven, E., Demer­outi, E., & Schaufeli, W. B. (2008). Working in the sky: a diary study on work engagement among flight attendants. Journal of Occupational Health Psy­chology, 13(4), 345. https://doi.org/10.1037/1076-8998.13.4.345 Xu, S. T., Cao, Z. C., & Huo, Y. (2020). Antecedents and out­comes of emotional labour in hospitality and tourism: A meta-analysis. Tourism Management, 79, 104099. https://doi.org/10.1016/j.tourman.2020.104099 Mehmet Çetin, PhD in organizational behavior. Currently works as an academician and the vice manager of the Graduate Education Institute in Istanbul Sabahattin Zaim University. He has ten years of academic experience. He also has worked as a strategic management consultant. His primary research areas are leadership, wellbeing, workplace spirituality, organizational culture, proactive behavior, and workplace loneliness. Gulmira Samenova, PhD candidate at Istanbul Sabahattin Zaim University. She has extensive practical knowledge in the business and management of trade companies. Her areas of interest include leadership, HRM, marketing management, and business processes in agricultural enterprises. Filiz Türkkan, PhD candidate, continues her education at Istanbul Sabahattin Zaim University Business Administration Doctorate Program. She currently works as a Branch Manager in Küçükçekmece District Health Directorate. She produces academic work on management and organization, organizational behavior, public health, and quality management. Ceylan Karatas, PhD student, continues her education at Istanbul Sabahattin Zaim University Business Administration Doctorate Program. She works as the Deputy Manager of Administrative and Financial Services at a city Hospital. She conducts research on organizational behavior, public health, management, and organization. Vloga vsakodnevnega vpliva na izmenjavo med vodjo in timom v javnih zdravstvenih zavodih Ozadje in namen: Ceprav je kljucna vloga vplivov v odnosih med vodjem in clani tima (Leader Member Exchange – KMX) splošno znana, je le malo študij raziskovalo vpliv dnevnih sprememb znotraj posameznih clanov tima na vsakodnevno izmenjavo med vodjo in timom (LMX) ali obravnavalo morebitne posamezne moderatorje tega vpliva. Študija je preucuje pozitivne in negativne vplive clanov tima na njihov dnevni LMX v javnih zdravstvenih organizaci­jah. Raziskovali smo tudi moderatorsko vlogo custvene inteligence. Metodologija: Izvedena je bila vecstopenjsko zasnovana raziskava, kjer je 70 udeležencev, ki delajo v javni zdra­vstveni organizaciji v Istanbulu, pet zaporednih delovnih dni vsak dan izpolnilo dnevni vprašalnik (350 odzivov na dnevni ravni), in splošni vprašalnik en teden po obdobju vsakodnevnega zbiranja podatkov (70 odzivov na ravni posameznika). Hipoteze smo preverjali z uporabo hierarhicnega linearnega modeliranja (HLM). Rezultati: Pozitivni in negativni vpliv sta bila pozitivno povezana z LMX (dnevna raven), negativni ucinek pa je imel negativno povezavo z LMX na medosebni ravni (tedensko poprecje rezultatov po dnevih). Ceprav je custvena inteligenca lastnosti pokazala vpliv ucinek na vec nivojih, nobena od predpostavljenih posrednih povezav ni bila statisticno pomembna. Zakljucek: Vloga dnevnega vpliva LMX je kljucnega pomena in ima zapleteno strukturo. Ugotovitve poudarjajo pomen vecstopenjskih raziskav za razumevanje razmerja vpliv-LMX, saj prikazujejo razlicne slike po dnevih in po posameznikih. Kljucne besede: Custveno delo, Izmenjava med vodjo in timom, Custvena inteligenca, Dnevna interakcija Appendix A: Leader Member Exchange Questionnaire - Scandura and Graen (1984) The original items (3 of the total 7) Instructions: This questionnaire contains items that ask you to describe your relationship with either your leader or one of your subordinates. For each of the items, indicate the degree to which you think the item is true for you by circling one of the responses that appear below the item. 1. How well does your leader (follower) understand your job problems and needs? 2. How well does your leader (follower) recognize your potential? 3. How would you characterize your working relationship with your leader (follower)? Slightly Modified (for day level use) Versions Used in the Study (moditifations are in italics): Instructions: This questionnaire contains items that ask you to describe your relationship with your leader. For each of the items, indicate the degree to which you think the item is true for you by circling one of the responses that appear below the item. Please consider your relationship with your leader today (not in general). 1. How well did your leader understand your job problems and needs today? 2. How well did your leader recognize your potential today? 3. How would you characterize your working relationship with your leader today? Appendix B: Emotional Labor Questionnaire – Lee & Brotheridge (2011) The original items: 1. Pretend to have emotions that I don’t really have 2. Show emotions that are expected rather than what I feel 3. Really try to feel the emotions I have to show as part of my job 4. Make an effort to actually feel the emotions that I need to display to others 5. Hide my true feelings about a situation 6. Resist expressing my true feelings Slightly Modified (for day level use) Versions Used in the Study (moditifations are in italics): 1. Today I pretended to have emotions that I didn’t really have 2. Today I showed emotions that are expected rather than what I feel 3. Today I really tried to feel the emotions I have to show as part of my job 4. Today I made an effort to actually feel the emotions that I need to display to others 5. Today I hided my true feelings about a situation 6. Today I resisted expressing my true feelings Appendix C: Positive Affect Negative Affect Scale - PANAS – Watson et al. (1988) Just the wording in the instructions was modified to ask the frequency regarding today. 1. Distressed 2. Upset 3. Enthusiastic 4. Nervous 5. Irritated 6. Strong 7. Active Appendix D: Trait Emotional Intelligence Questionnaire- TEIQue-SF – (Person level) Petrides & Furnham (2001), Petrides (2009) Just the wording in the instructions was modified to ask the frequency regarding today. 1. Expressing my emotions with words is not a problem for me. 2. I often find it difficult to see things from another person’s viewpoint. 3. On the whole, I’m a highly motivated person. 4. I usually find it difficult to regulate my emotions. 5. I generally don’t find life enjoyable. 6. I can deal effectively with people. 7. I tend to change my mind frequently. 8. Many times, I can’t figure out what emotion I’m feeling. 9. I feel that I have a number of good qualities. 10. I often find it difficult to stand up for my rights. 11. I’m usually able to influence the way other people feel. 12. On the whole, I have a gloomy perspective on most things. 13. Those close to me often complain that I don’t treat them right. 14. I often find it difficult to adjust my life according to the circumstances. 15. On the whole, I’m able to deal with stress. 16. I often find it difficult to show my affection to those close to me. 17. I’m normally able to “get into someone’s shoes” and experience their emotions. 18. I normally find it difficult to keep myself motivated. 19. I’m usually able to find ways to control my emotions when I want to. 20. On the whole, I’m pleased with my life. 21. I would describe myself as a good negotiator. 22. I tend to get involved in things I later wish I could get out of. 23. I often pause and think about my feelings. 24. I believe I’m full of personal strengths. 25. I tend to “back down” even if I know I’m right. 26. I don’t seem to have any power at all over other people’s feelings. 27. I generally believe that things will work out fine in my life. 28. I find it difficult to bond well even with those close to me. 29. Generally, I’m able to adapt to new environments. 30. Others admire me for being relaxed. DOI: 10.2478/orga-2021-0009 Effect of Management Commitment to Service Quality on Car-hailing Drivers’ Service Behaviors: The Case of GrabCar in Vietnam Phuong TRAN HUY, Hong Chuong PHAM National Economics University, Hanoi, Vietnam, phuongth@neu.edu.vn; chuongph@neu.edu.vn Background and Purpose: Management Commitment to Service Quality (MSCQ) has been found to positively pre­dict employee’s service quality and service behaviors in different service industries. In the context of sharing econ­omy, the relationship between company and service providers is different from traditional employment relationship. For car-hailing service, drivers are mainly classified as contractors rather than employees. It is, therefore, necessary to understand whether MSCQ influences drivers’ service quality in a car-hailing context. Design/Methodology/Approach: Data were collected from 214 GrabCar drivers in Vietnam using online and offline survey. Partial Least Squares Structural Equation Modeling (PLS-SEM) was used for data analysis. Results: The findings suggest that three dimensions of MCSQ, namely reward system, technology support and or­ganizational support exert significant direct impact on drivers’ service behaviors. In addition, job involvement plays an intermediary role in the relationship between MCSQ and service behaviors. Conclusion: This study expands previous research on MCSQ to the car-haling service and confirms the role of job involvement as an important mechanism to improve service quality provided by drivers. Due to the characteristics of the company-service providers’ relationship in the sharing economy, the mechanisms through which MCSQ influ­ences providers’ service performance need to be investigated in further details. Keywords: Management commitment to service quality (MCSQ), Service behaviors, Job involvement, Car-hailing, Sharing economy, Vietnam. 1 Introduction In the context of sharing economy, a number of busi­nesses have emerged and proliferated (Peticca-Harris et al., 2018). In transportation, car-hailing service companies such as Uber, Lyft and Zipcar have expanded and gained significant market share throughout the world. Ride-shar­ing or Car-hailing typically refers to hire services where a passenger books for a ride using an online/ mobile plat­form. This type of sharing can happen when a driver shares his private vehicle with a passenger in a more open taxi-like system (Amey et al., 2011). By 2017, Uber was in use in 507 cities throughout the world and has an estimated market value of over $68 billion (Çetin, 2017). In China, Didi Chuxing (DiDi), have been successful in the transpor­tation market, overtaking Uber’s four billion rides in 2017 (Zuo, et al., 2019). In other markets, Transport Network Companies (TNCs) such as GrabCar and Go-jek have also achieved remarkable market growth (Hamenda, 2019). The focus of the current study is on car-hailing ser­vice, namely GrabCar, provided by Grab Inc in Vietnam. Formerly known as MyTeksi and GrabTaxi, Grab is a Singapore based ride-sharing company. The company was originally founded in Malaysia and later moved its headquarters to Singapore. Grab is an app-based service provider that offers ride-hailing, ride-sharing, and logistics services through its app. It now operates in the Southeast Asian countries of Singapore, Malaysia, Indonesia, Philip­pines, Vietnam, Thailand, Myanmar, Cambodia and Japan. Grab has been operating in Vietnam since 2014 with the original name Grabtaxi. Current transportation services in­clude GrabCar, GrabTaxi and GrabBike. According to ABI research (2019), Grab dominates the car-hailing market in Vietnam with about 73% market share, 5 times more than that of its closest rival. In 2019, it is estimated that Grab has attracted about 135,000 drivers, more than 50,000 of whom are car drivers. Despite its high potential, car-hailing services are not without problems, with service quality being the most prominent (Rayle, et al., 2016). Rayle et al. (2016) con­cerned that poor service quality may result because riders cannot compare information on price or service quality before choosing a vehicle. Yang et al. (2017) indicated that low industry expertise, lack of professional training, and loose obligations with the sharing service organizers are among the sources for low quality service provided by TNCs. Both Katz (2015) and Zuo et al. (2019) seemed to agree that, it is the business model, in which TNCs such as Uber have no employment relationship with direct service providers (drivers), and no financial stake in any particular transaction that cause service problems. Because service quality is the most critical success factor for service (Frei, 2008), TNCs have find ways to provide better services to their customers. However, be­cause of the loose nature of the relationship between TNCs and drivers, managing the ride service quality provided by contracted drivers is a challenging tasks (Shokoohyar, Sobhani, and Nargesi, 2020). The current study aims to ad­dress the following research questions: (a) “What should online platforms do to improve service quality offered to customers by their drivers?” and (b) “Through what chan­nels do management’ actions influence drivers’ service delivery?” The current study proposes that a TNC’s management commitment to service quality (MCSQ) would influence car-hailing driver’s service, despite the striking difference between car-hailing and traditional employment context. In addition, unlike in traditional employment context, where organizational commitment may act as a mediator of the relationship between firm commitment to service quality and provided service, this paper advocates that in the car-hailing context, job involvement would mediate the above relationship. The contribution of this paper is threefold. First, while service quality has already attracted plentiful research in­terests, Priporas et al. (2017) and Zuo et al. (2019) pro­posed that empirical research about service quality in the sharing economy context in general, and online car-hail­ing industry, in particular, are still scarce. Furthermore, most extant studies examined the dimensions of service quality or investigated it as an antecedent of consumer satisfaction (Ahrholdt, et al. 2017; Priporas et al., 2017; An et al., 2019). To date, little research investigates the antecedents of service quality in the sharing economy. In addition, while many studies observe service quality from the users’ perspective (Dias et al., 2017; Ju et al., 2018; Priporas et al., 2017), few studies take the perspective of the direct service providers. Extant studies that focus on the supply side of car-hailing service mostly focus on fac­tors that influence drivers’ supply of service such as fare policy (Shokoohyar, 2018) or socio-economics factors and transportation infrastructures (Shokoohyar, Sobhani and Sobhani, 2020). This research takes a different focus by in­vestigating service quality provided by car-hailing drivers. Second, while the impact of management commitment to service quality on service performance has been inves­tigated in the universal organizational context (Ashill et al., 2008; Babakus et al. 2003; Kim et al., 2009), few re­search focus on the sharing economy context in general and car-hailing service in particular. In addition, Lin et al., (2020) argued that because the determinants of workers’ performance in the sharing economy have been scarce, more research attention to explore the driving mechanism of such workers’ performance is needed. Similarly, Gleim et al. (2019) acknowledged a clear gap in our understand­ing of the perceptions and work outcomes of sharing econ­omy workers and emphasize the importance of addressing this gap as the sharing economy continues to grow. Third, this research explores the mediating impact of job involvement on the relationship between MCSQ and service performance. In the car-hailing context, because of the usually temporary and short term, non-employment relationship between TNCs and their direct service pro­viders (Yang et al., 2017; Karn & Hutson, 2019) we ex­pect that organizational commitment would not act as an intermediary variable of the MCSQ-service performance linkage. Peticca-Harris, et al. (2018) studied the motiva­tions and experiences of Uber drivers in Canada. Data collected from 31 drivers, divided into part-time, full-time non-professional and full time professional group, indicate that drivers do not think about a long-term future with Uber. Instead, the drivers reported that driving for Uber is a temporary solution. Specifically, part-time drivers con­sider working for Uber as a part-time job for some extra income, while looking for more attractive future opportu­nities elsewhere. They also regard “driver” as an unwanted occupation and they prefer Uber because it is easier to quit Uber than traditional taxi. Full-time non-professional driv­ers do not regard driving as an occupational choice, yet they have to drive to make ends meet, showing some ex­tent of continuance commitment to the company. Finally, while full-time professional drivers accept driving as their profession, they only opt for Uber for flexibility and po­tentially better income in comparison with traditional taxi companies. In a recent study, Lin et al. (2020) proposed that because of the flexibility that allows drivers to work for multiple online platforms simultaneously, commitment should not be viewed from organizational perspective. In­stead, they advocated the use of career commitment, which refers to the “relative strength of an individual’s identifi­cation with and involvement” with car-hailing sector in the sharing economy. The current study, examines job in­volvement of car-hailing drivers as a possible mechanism to deliver the impact of MSCQ on service quality. 2 Literature review 2.1 Management commitment to service quality Extant literature has confirmed the role of manageri­al practices in fostering service behaviors among service workers (Kandampully & Menguc, 2000; Ueno, 2008). Managerial practices that contribute to excellent service quality can be conceptualized as management commit­ment to service quality (Babakus et al., 2003; Karatepe & Karadas, 2012; Rod & Ashill, 2010). Babakus et al. (2003, p. 275) defined management commitment to service qual­ity (MCSQ) as “employees’ appraisal of an organization’s commitment to nurture, develop, support, and reward its employees to achieve service excellence”. Previous re­search suggested that MCSQ comprises practices such as training, empowerment and reward (Babakus et al., 2003), service technology, supportive management and servant leadership (Ashill et al., 2008), customer service orien­tation (Rod and Ashill, 2010), and organizational support (Kim et al., 2009). MCSQ has been found to positive­ly predict employee’s service recovery performance and service behaviors in different service industries including banking (Babakus, et al., 2003), hospitality (Karatepe & Karadas, 2012), healthcare (Rod & Ashill, 2010), and pub­lic service (Ashill et al. 2008). In the context of car-hail­ing, because the drivers are mainly classified as contrac­tors rather than employees (Rosenblat & Stark, 2016; Rosenblat, et al., 2017), they have total autonomy in their work, receive no service training and do not report to a particular leader. Therefore, MCSQ in this context should not include training, leadership and empowerment. In the current study, MCSQ includes service performance-based reward system, service technology and perceived organi­zational support. 2.2 Job involvement Job involvement refers to the degree to which individu­als psychologically identify with their present job (Kanun­go, 1982). According to Kanungo (1982), job involvement and work involvement are different constructs. While job involvement focuses on one’s identification with the cur­rent job and is influenced by the extent to which the current job can satisfy present needs, work involvement refers to the value of work in an individual’s entire life and is a function of past experience. Because car-hailing drivers do not regard the car-hailing job as a career or a long term commitment (Karn & Hutson, 2019), and most of them have worked in other jobs (Valente, Patrus, & Guimarães, 2019), this study focuses on job involvement of GrabCar drivers. Research provides evidences to support the desirability of job involvement in the workplace. For example, people with high job involvement tend to develop strong tie with their jobs and are willing invest “personal resources” in their current job (Dimitriades, 2007; Kanungo, 1982). On the other hand, studies have found that employees with a low degree of job involvement tend to lose focus on their job (Hogan et al., 2013), and are less imaginative and open to new concepts (Abdallah et al., 2016). 3 Theoretical framework and hypotheses 3.1 Impact of management commitment on service quality and service behavior Reward system The impact of organizational reward on service perfor­mance has been well documented in the literature. Para­suraman (1987) argued that in order to distinguish a firm’s services from others, rewarding innovative thinking, out­standing service behaviors, and excellence in service de­livery is vital. Similarly, Yavas et al. (2003) proposed that reward is not only important in fostering excellent service but also crucial in motivating direct service providers to solve service complaints. In ride-hailing service, TNCs implement positive and punitive reward systems toward direct service providers. For instance, Uber offers rewards to its drivers if they achieve a pre-specified number of trips per week (Shokoo­hyar, 2018). In addition, Uber introduces reward program such as Uber Pro which offers such benefit as priority pickups, supports on the road, and family benefit based on numbers of trips and rider ratings (Uber, 2020). Similar re­ward programs have been applied by Grab who bases driv­er benefits on monthly rides, completion rate and star rat­ing (Grab, 2019). On the other hand, car hailing companies also offer punitive reward program. For example, ride-hail companies require a driver to maintain a minimum rating, completion rate and driver timeout. Drivers are penalized if they go below the required completion rate. If a driv­er excessively ignores or cancels booking requests, they are locked out of the platform for a temporary period of time. Drivers with passenger complaints and low star rat­ings are also flagged and suspended (Grab, 2019). While many TNCs use riders’ rating to base their reward system, drivers complain that they are not aware of what influence their ratings (Rosenblat & Stark, 2016), and are doubtful about the link between ratings and benefits (Rosenblat et al., 2017). Ashill et al. (2008) advocated that if direct ser­vice providers believe that rewards systems and policies truly reflect commitment to service quality by manage­ment, they tend to engage themselves in service quality efforts. Thus, we can conclude that in order to improve service performance, reward structure should be perceived by drivers to be contingent upon service performance. H1a: Service performance-based reward system is pos­itively linked with service behaviors of car-hailing drivers Service technology support Service technology may aid service workers in pro­viding better service to customers by optimizing pro­cess, making procedures accurate and efficient, removing time-consuming routine tasks and reducing human errors (Ashill et al., 2008; Lee et al., 2003). In car hailing service, TNCs provide drivers with technological support in the form of rudimentary dispatching and navigation through the company-provided mobile apps. According to Karn and Hutson (2019), while the app-based system is consid­ered reliable, issues such as difficulty with driver-passen­ger meet-up due to GPS failures, or billing problems that neither the driver nor rider could resolve through the app’s functions may reduce service performance. Thus, we can expect that the availability of technology service support will enable the drivers to focus on delivery transportation service. Based on the above arguments, it is hypothesized that: H1b: Service technology support is positively linked with service behaviors of car-hailing drivers Perceived Organizational support Organizational support refers to an individual’s per­ception that the organization is concerned about their success and well-being (Eisenberger et al., 1986). Organi­zational support has been a strong predictor of service per­formance (Ashill et al., 2008; Kim et al., 2009; Susskind et al., 2000). Ashill et al. (2008) argued that, based on the norm of reciprocity, service employees who receive sup­port from their organization are likely to exert more efforts on their job in return. Susskind et al. (2000) indicated that organizational support is of great importance when dealing with service complaints. Because POS reflects workers’ assessment of self-value by the hiring organization (Al­balawi et al., 2019), service workers who receive support in such situations tend to build trust with the organization, which in turn, influence service efforts. In the context of car-hailing, support received by the drivers from the shar­ing platforms is one important area of interest. Karn and Hutson (2019) found that drivers receive little or no sup­port in many aspects of work including vehicle ownership and maintenance, expenses tracking and deductions, and business finances. Karn and Hutson (2019) proposed that these lack of support influence drivers socially, behavio­rally, financially, and emotionally. Shokoohyar (2018) also indicated that operational costs and lack of job security are the main concerns of drivers and suggested that providing more supportive incentive contract to compensate driv­ers’ operational costs can improve drivers job satisfaction. Thus, we can expect that organizational support will foster efforts in service behaviors. H1c: Perceived Organizational Support is positively linked with service behaviors of car-hailing drivers 3.2 The impact of MCSQ on job involvement Reward system Kanungo (1982, p.342) defined job involvement as “the degree that an individual psychologically identifies with his or her present job”, and suggested that because job involvement concerns the present job, it is influenced by how much the present job can satisfy current needs. In car-hailing scenario, Standing et al. (2019) suggested that income, being independent, and lack of job opportuni­ties are the main reasons to provide service. Valente et al. (2019) found that the main reason for becoming a car-hail­ing driver is to solve unemployment issues and to gener­ate more income, rather than to make an adjustment to the market or a to search for more flexible and complementary type of work. Because reward structure that is contingent upon service performance can satisfy the current needs of generating income, it is expected to be positively linked with job involvement. H2a: Service performance-based reward system posi­tively influences a driver’s job involvement Service technology support Technology support can help service employees carry out their job duties successfully. It frees the employees of routine repetitive tasks that may frustrate the employees. With the help of technology, the service workers can de­vote their time to deliver excellent service to the custom­ers (Lytle et al., 1998). Technology support also makes the jobs less demanding and taxing by reducing job stress. Therefore, technology is expected to positively linked with job involvement. It is hypothesized that: H2b: Service technology support positively influences a driver’s job involvement Perceived organizational support Researchers contended that the work environment has a significant impact on job involvement (Lawler & Hall, 1970; Lawler, 1992). Lambert and Paoline (2012) sug­gested that positive aspects of work environment such as organizational support, can reduce job stress and make the job more enjoyable, thereby increases job involvement. In addition, when employees perceive that the organiza­tion is supportive and concerned about their success and well-being, they may consider it as a signal from the or­ganization that it respects and values the jobs that they are doing. Organizational support may also contribute toward the fulfillment of “socio-emotional needs” of employees (Rhoades & Eisenberger, 2002). Being successful and val­ued increases the likelihood that a person will identify with his or her job. In addition, fulfillment of needs also contrib­utes to the feeling of attachment with the job. Therefore, it is hypothesized that: H2c: Perceived organizational support positively in­fluences a driver’s job involvement 3.3 The mediating role of job involvement Job involvement is identified as a predictor of impor­tant work outcomes, such as organizational commitment, job satisfaction, increased work effort, job performance, and organizational citizenship behaviors (Chen & Chiu, 2009; Diefendorff, Brown, Kamin, & Lord, 2002). Wood (1974) suggested that job involvement results from in­trinsic motivation and therefore, is likely to be linked to job performance that provides intrinsic rewards. Abhari et al. (2019) study collaborative innovation in the sharing economy and found that beside extrinsic motivation such as financial gain, intrinsic motivation such as enjoyment and altruism are important determinants of engagement in social product development projects. Lin et al. (2020) research food delivery workers in the sharing economy and found that enjoyment attached with the job was an important work value perceived by these workers. In ad­dition, they found a positive relationship between drivers’ work centrality and work engagement, which in turn, fos­ter positive work behaviors. Thus it can be expected that, in the context of sharing economy, job involvement is an important determinant of positive job outcomes among participants. Furthermore, the mediating impact of job in­volvement on the relationship between job characteristics and job performance has been confirmed in other contex­tual settings (Chen & Chiu, 2009; Mrayyan & Al-Faouri, 2008). Following these above argument, it is hypothesized that: H3a-c: Job involvement mediates the relationship be­tween reward system, service technology and perceived organizational support and service performance. 4 Method 4.1 Data collection Preliminary interviews with 10 drivers were conducted by the author during 30-minute rides with Grab to under­stand the characteristics of the relationship between driv­ers and the TNC, the availability of technological support and job involvement of the drivers. Information obtained from interviews were used to develop questionnaires for a subsequent quantitative survey. Quantitative data were collected via survey based on convenient sampling and snowball techniques due to unavailability of drivers’ information. Questionnaires in paper-based or QR-coded versions were administered to Grab drivers by our research collaborators during their rides with Grab. Then, the drivers were requested to intro­duce other fellow drivers who are driving for Grab. Based on this information, we contacted other drivers and sent the QR-coded/online versions of the questionnaires. These drivers may choose whether to participate in the research. Out of 450 questionnaires sent, 214 have been filled by Grab drivers and were used for data analysis. 4.2 Measure All items are measured on a 5-point Likert scale rang­ing from 1. Strongly disagree to 5. Strongly agree, unless stated otherwise. Service performance-based Reward system: respond­ents are asked to evaluate the degree to which reward are based on performance at Grab Vietnam. Three items used by Ashill et al. (2008) were adopted to measure employee reward. A sample item is “My income depends on the qual­ity of service I deliver”. Service Technology Support: Drivers are requested to rate the extent to which technology support is offered by the company. This scale was measured using a four-item scale used by Ashill et al. (2008). The items include: “Grab has the necessary technology support to serve my custom­ers effectively”. Perceived organizational support (POS): The drivers are asked to provide their self-perception of the support offered to them by Grab company. Eight items selected from the 16-item SPOS scale developed by Eisenberger et al. (1986) were used to measure POS in car-hailing back­ground. Uses of shorter version was suggested by Rhoades and Eisenberger (2002) as acceptable. Shorter version of the scale has been used by Hameed et al. (2019). The items include “Grab is willing to help me when I need a special favor”. Job involvement: This variable measures the extent to which a driver is identified with their driving jobs. The 10-item scale developed by Kanungo (1982) was adopted in this research. These items include: “The most important things that happen to me involve my present job” and “I live, eat and breathe my job”. Service behavior (SB): This variable measures the drivers’ delivery of service to the customers. Five items adapted from Zhang et al. (2019) were used to measure service behavior. One sample item was “I provide trans­portation services according to the job responsibilities prescribed by Grab”. The scale items are presented in the Appendix. 5 Results 5.1 Sample description Among 214 drivers, 210 are men and only 4 are wom­en. The average age of the drivers is 32.4 years. On av­erage, the drivers have been in partnership with Grab for two and a half years. About 75% of the drivers bought a car to drive for Grab while the remaining 25% used their under-utilized cars. Also, more than two-thirds of the re­spondents are working full-time as drivers. More than half of the respondents also drive for other ride-hailing plat­forms. 5.2 Validity and reliability Before testing the hypotheses, we check the reliability and validity of the research variables. As indicated in figure 1, all indicators showed significant loading values on their specified construct, and the standardized loadings ranged from .54 to .89 (t values = 7.093 to 39.77; p < .001). For reliability and convergence validity, Composite Reliability (CR) of all latent variables were calculated and compared to the threshold value of .7. The average variance extracted (AVE) for all variables should also exceed the threshold value of .5 (Hair, Black, Babin, and Anderson, 2010). For discriminant validity, the square roots of the all AVE scores should be higher than the correlation coeffi­cients between the constructs (Fornell & Larcker, 1981). Furthermore, Heterotrait-monotrait (HTMT) ratios must be smaller than 0.8. The figures shown in Table 3 and the fact that the highest HTMT ratio is 0.639 confirm the reli­ability and validity of the constructs. Common method bias The Harman’s one factor test is used to examine the potential problem of common method variance. All items were entered into EFA which is forced to produce one fac­tor. The result suggested that the factor accounts for only 21.25% of the total variance. Therefore, common method bias does not pose a significant problem in this study. 5.3 Hypothesis testing Hypotheses H1a to H1c concern the direct impact of management commitment to service quality on car-hailing drivers’ service behaviors. Specifically, reward, technolo­gy support and organizational support are each expected to positively and significantly predict service behaviors. These hypotheses were tested based on the significance of the structural path coefficients. As indicated in Table 4, the path coefficients from reward (ß=.198, p<.05), technolo­gy support (ß=.370, p<.001) and organizational support (ß=.158, p<.05) are all positive and significant. Therefore, hypotheses from H1a to H1c are supported. Table 4 showed that the path coefficients from reward, technology support and organizational support to Job in­volvement are all positive and significant, supporting hy­potheses H2a to H2c. Hypotheses H3a to H3c suggest that Job involvement mediates the relationship between three dimensions of MCSQ and Service behaviors. To test the hypotheses, the 4 steps suggested by Baron and Kenny (1986) are followed. The conditions for mediation need to be checked before making conclusion. The first two conditions for mediation are that the independent variables must be significantly associated with the mediator (job involvement) and the dependent variable (service behaviors). These conditions have been satisfied with the support of Hypothesis H1a to H1c and H2a to H2c. The third condition was also met when the mediator significantly influences the dependent variable (ß=.208, p<.05). The indirect effects of reward, technology support and organizational support on service behaviors via job involvement are indicated in Table 5. The data provide support for H3b and H3c but not for H3a. 6 Discussion and contribution 6.1 Discussion This study aims to investigate the impacts of MCSQ on service behaviors among GrabCar drivers. Because of the generally agreed non-employment relationship between Transportation Network Companies and their drivers, it is important to understand how these TNCs can improve service quality by influencing their drivers. The result showed that reward is associated with ser­vice behaviors among car-hailing drivers. Specifically, the impact of management commitment to service quality, manifesting in the reward system that is contingent upon service delivery, is positively influence the delivery of ser­vice as expected by the TNCs. In the car-hailing context, income is among the primary reasons for individuals to be­come a driver (Standing et al., 2019; Valente et al., 2019). If the drivers perceive that their income is directly linked with their service delivery, they are likely to devote more effort in providing excellent service. The results are in line with the findings of previous research in the car-hailing context. Fielbaum and Tirachini (2020) found that wage level and the transparency in wage determination are the top two predictors of job satisfaction among ride-hailing drivers in Chile. On the other hand, inadequate pay and perceived pay injustice are among the causes for drivers’ dissatisfaction (Karn & Hutson, 2019). The availability of technology support has been found to be an important determinant of service behavior among car-hailing drivers. Because Car-hailing service is a tech­nology-based transportation service, the role of technol­ogy support is paramount. Technology helps connect the drivers and passengers, calculate fares and locate the best possible routes. However, despite the overall utility of the app, problems usually arise due to GPS failure or slow up­date. In addition, drivers report that they did not receive much support in dealing with customers beyond the scope of the app (Rosenblat et al., 2017). Other studies highlight the role of trip characteristics, weather conditions, and socioeconomics and demographic factors on car-hailing service accessibility, as well as, service performance in­dicators such as pick-up waiting time, trip duration, and fare increase (Shokoohyar, Sobhani, & Sobhani, 2020; Shokoohyar, Sobhani, & Nargesi, 2020). The findings em­phasize the importance of technology support in regulating supply of car-hailing service in order to improve service performance. Despite the loose relationship between TNCs and driv­ers, this study suggests that organizational support is a cru­cial factor in fostering service delivery among drivers. In the context of sharing economy, Liu et al. (2020) found that organizational support, in the forms of psycholog­ical contract fulfillment, directly influences gig workers’ task performance. Similarly, Gleim et al. (2019) reported that organizational trust is an important predictor of job outcome status and satisfaction among sharing economy workers. Furunes and Mkono (2019) also pointed out that lack of organizational support is the main challenges for food-delivery workers, especially in case of service fail­ure. Job involvement refers to the personal identification with the current job. Factors that make the job more in­teresting and enjoyable and those that reduces obstacles and strains in doing the job are expected to increase job in­volvement. The findings add to previous evidences regard­ing the antecedents of job involvement in sharing econo­my context. Fielbaum and Tirachini (2020) indicate that “enjoy driving” was an important predictor of car-hailing drivers’ job satisfaction, second only to flexibility. Job involvement partially mediates the impact of per­ceived organizational support and technology support on service behaviors. Individuals with high job involvement tend to find the satisfaction of needs in the job. Delivery excellence service is an important mechanism to satisfy both current and long-term needs. Although reward is significantly related to both job involvement and service behaviors, and job involvement affects service behaviors, the indirect effect of reward on service behavior is not significant. A possible explanation lies in that job involve­ment refers to the degree to which the person places the job at the center of his life. Wood (1974) suggested that job involvement results from intrinsic motivation and there­fore, is likely to be linked to factors that provides intrin­sic rewards. Lin et al. (2020) also concluded that although many workers in the sharing economy aim for earning quick money, it is often their obligation norm rather than the monetary benefits which determine their career com­mitment. Therefore, although extrinsic reward may have impact on job involvement, other characteristics of the job and the job-environment may contribute much more to job involvement. 6.2 Theoretical contributions The current research concerns the relationship between MCSQ and direct service providers’ service delivery in a sharing economy context. While previous research ex­plores the impact of MCSQ in traditional employment settings, few looks to test these impacts in a non-tradition­al employment context such as car-hailing service. This study suggests that MCSQ can exert positive influence on frontline employees in other service industries. Second, the current study is one of the few attempts to investigate the mediating role of job involvement in the relationships between MCSQ and service behaviors. Oth­er research examines the mediation role of organizational commitment on the above linkage. The nature of the rela­tionship between TNCs and their drivers undermines the mediating role of such variables as organizational commit­ment. Because independence and flexibility distinguishes car-hailing service from other transportation services, the role of job involvement in service quality may be of great­er interests than organizational commitment. 6.3 Practical Implications In order to motivate drivers to improve service quali­ty, which, in turn lead to customer satisfaction and profit, Grab can take several courses of action as follows. First, upgrading the platform and software to enhance the trans­parency of the rating system and allow drivers’ participa­tion in the performance evaluation process may enhance the perception that the reward system is contingent upon service performance. Currently, although many platforms use rider’s rating to measure service quality, many drivers do not believe such ratings influence their incomes. Sec­ond, technology support should be improved in order to as­sist the drivers in dealing with service issues and to prevent service failure. The need to contact call centers or filling paperwork could reduce the attractiveness of this form of work and cause role strain or job stress, which eventually reduce service performance. Finally, providing other kind of support such as family oriented support, service train­ing, and insurance could enhance the drivers’ involvement in the job. Highly involved drivers are likely to provide excellent service. 7 Conclusions and limitations 7.1 Conclusions This study expands previous research on MCSQ from traditional service industries to the context of sharing economy. In car-hailing context, the direct service provid­ers are responsible for the delivery of high quality service to end users and receive performance feedback from end users via the form of star-ratings (Grab, 2019). The end-us­ers are in fact, evaluate the quality of service delivered by drivers rather the quality of service provided by Transport Network Companies. TNCs, in turn, influence service quality by adopting reward and punitive policies based on star-ratings (Shokoohyar, 2018). Although, customers do not directly rate a TNC’s service, it is likely that they will switch to a TNC’s competitors if they constantly receive poor service from drivers associated with this TNC. There­fore, TNCs should encourage their drivers to improve their service quality. In addition, because TNCs are competing for the service of drivers, they need provide support for their drivers. This research reveals that by showing their commit­ment to service quality, TNCs can influence service quali­ty provided by their affiliated drivers. Furthermore, due to the loosely tied nature of the relationship between a TNC and its drivers, providing long-term career, enhancing or­ganizational commitment and creating loyalty are not ap­propriate. By fostering job involvement among drivers via various forms of technological and psychological support, TNCs could help drivers in improving service quality. Our results can provide insights for future research about service quality in other services in the sharing economy context. 7.2 Limitations and directions for future research The current study seeks to expand current understand­ing of MCSQ’s influence on service behaviors to sharing economy context. However, the findings of this research only focuses on car-hailing service, thus the generaliza­bility of the results is limited. Future research may seek to advance the focus to other sharing economy business models. Cross-sectional design and single data source are the second limitation of the current research. Because job in­volvement refers to a state of psychological identification with work, observing it at one point in time may not cap­ture the influence of MSCQ on job involvement overtime. Future research may adopt a longitudinal research design to avoid this limitation. In addition, although there has been no severe effect of common method bias in the cur­rent paper, it would be desirable to collect data from mul­tiple sources such as management of TNCs, peer drivers, and riders in order to yield more robust results. The current study investigates the mediating role of job involvement in the MSCQ-service behaviors linkage. Be­cause literature has suggested that people participate in the sharing economy for both extrinsic and intrinsic motiva­tion, exploring other mechanisms that transfers the impact of sharing platform management practices on participants’ behaviors is, therefore, require further research attentions. Literature 1Received: 7th November 2020; revised: 23rd February 2021; accepted: 11th March 2021 Table 1: Summary of MCSQ studies Study MSCQ practices Dependent variable Mediator Sample Kandampully and Menguc (2000) Quality control practices; quality measurement prac­tices; and service mainte­nance practices Service quality None Service firms Babakus et al. (2003) Training, empowerment, and rewards Service recovery perfor­mance Affective com­mitment Job satisfaction Frontline bank employees Yavas, Karatepe, Avci, and Tekinkus (2003) Service orientation, re­ward, training, empower­ment, and teamwork Service recovery perfor­mance None Frontline bank employees Lee, Barker, and Kandampully (2003) Technology support Service quality Hotel managers Ashill, Rod, and Carruthers (2008) Training, empowerment, rewards, supportive management, servant leadership, and service technology. Service recovery perfor­mance and turnover intentions. Job satisfaction and organiza­tional commit­ment Public sector service frontline employees Kim, Tavitiyaman, and Kim (2009) Organizational support, rewards, empowerment, and training Service behaviors Job satisfaction Hotel employees Rod and Ashill (2010) Reward, training, em­powerment, and service orientation. Service recovery perfor­mance Organizational commitment. Frontline hospital employees Karatepe and Karadas (2010) Training, empowerment, and rewards Service recovery perfor­mance and extra-role customer service Job embedded­ness Frontline hotel employees Figure 1: Research model Table 2: Respondent profile Variable Frequency Percentage Gender of driver Men 210 98.1% Women 4 1.9% Age of driver From 18 to 25 64 29.9% From 26 to 35 75 35.1% From 36 to 45 41 19.2% Above 46 34 15.8% Year of working with Grab Less than 1 year 101 47.2% From one to 3 years 74 34.6% More than 3 years 39 18.2% Type of vehicle Under utilized 156 72.9% Newly bought 58 27.1% Multiple TNCs Drive only for Grab 93 43.5% Drive for other TNCs 121 56.5% Figure 2: Factor loadings Table 3: Correlations, reliability and validity of the constructs Variable 1 2 3 4 5 a C.R AVE Job Involvement .710 .891 .91 .504 Organizational support .404 .782 .905 .925 .612 Reward .291 .136 .837 .785 .875 .701 Service Behavior .458 .318 .447 .790 .849 .892 .624 Technology Support .346 .131 .452 .552 .802 .816 .877 .643 Note: CR: composite reliability; AVE: average variance extracted; a: Cronbach’s alpha. Square-root of AVE for each construct in bold. Table 4: Direct effects of management commitment to service quality on service behavior and job involvement Direct effect Service behavior Job Involvement Reward .198* .124* Technology support .370*** .239** Organizational support .158* .355*** Note: CR: composite reliability; AVE: average variance extracted; a: Cronbach’s alpha. Square-root of AVE for each construct in bold. Table 5: Indirect effect of MCSQ on service behaviors via job involvement Indirect effect Coefficient t p value Reward -> Job Involvement -> Service behaviors .028 1.485 .138 Technology support -> Job Involvement -> Service behaviors .050 1.978 .049 Organizational support -> Job Involvement -> Service behaviors .074 2.302 .022 Figure 3: Path coefficients and p-value Abdallah, A.B., Phan, A.C., & Matsui, Y. (2016). Inves­tigating the effects of managerial and techno­logical innovations on operational performance and customer satisfaction of manufacturing compa­nies. International Journal of Business Innovation and Research, 10 (2/3), 153–183. https://doi.org/10.1504/IJBIR.2016.074824 Abhari, K., Davidson, E.J., & Xiao, B. (2019). Collabo­rative innovation in the sharing economy: Profiling social product development actors through classifica­tion modeling. Internet Research, 29(5), 1014-1039. https://doi.org/10.1108/INTR-03-2018-0129 Ahrholdt, D. C., Gudergan, S. P., & Ringle, C. M. (2017). Enhancing Service Loyalty: The Roles of Delight, Satisfaction, and Service Quality. Jour­nal of Travel Research, 56(4), 436–450. https://doi.org/10.1177/0047287516649058 Albalawi, A.S., Naughton, S., Elayan, M.B., & Sleimi, M.T. (2019). Perceived Organizational Support, Alter­native Job Opportunity Organizational Commitment, Job Satisfaction and Turnover Intention: A Moderat­ed-mediated Model. Organizacija, 52(4), 310-324. https://doi.org/10.2478/orga-2019-0019. Amey, A., Attanucci, J., & Mishalani, R. (2011). Re­al-Time Ridesharing: Opportunities and Challenges in Using Mobile Phone Technology to Improve Rideshare Services. Transportation Research Record, 2217(1), 103–110. https://doi.org/10.3141/2217-13 An, S., Suh, J., & Eck, T. (2019). Examining structural relationships among service quality, perceived value, satisfaction and revisit intention for Airbnb guests. In­ternational Journal of Tourism Sciences, 19(3), 145-165. https://doi.org/10.1080/15980634.2019.1663980 Ashill, N.J., Rod, M., & Carruthers, J. (2008). The Ef­fect of Management Commitment to Service Qual­ity on Frontline Employees’ Job Attitudes, Turno­ver Intentions and Service Recovery Performance in a New Public Management Context. Journal of Strategic Marketing, 16(5), 437-462. https://doi.org/10.1080/09652540802480944 Babakus, E., Yavas, U., Karatepe, O. M., & Avci, T. (2003). The effect of management commitment to ser­vice quality on employees’ affective and performance outcomes. Journal of the Academy of Marketing Sci­ence, 31(3), 272–286. http://dx.doi.org/10.1177/0092070303031003005 Baron, R. M., & Kenny, D. A. (1986). The moderator-me­diator variable distinction in social psychological re­search: Conceptual, strategic, and statistical consider­ations. Journal of Personality and Social Psychology, 51(6), 1173-1182. Çetin, T. (2017). The Rise of Ride Sharing in Urban Trans­port: Threat or Opportunity? in H. Yaghoubi (Ed.), Ur­ban Transport Systems. Rijeka: InTech, pp. 191-202. https://doi.org/10.5772/66918 Chen, C. C., & Chiu, S. F. (2009). The mediating role of job involvement in the relationship between job char­acteristics and organizational citizenship behavior. The Journal of Social Psychology, 149(4), 474-494. https://doi.org/10.3200/SOCP.149.4.474-494 Dabholkar, P.A. (2000). Technology in service delivery: Implications for self-service and service support, in T.A. Swartz and D. Iacobucci (Eds.), Handbook of services marketing and management, Thousand Oaks, CA: Sage, 103–110. Dias, F.F., Lavieri, P.S., Garikapati, V.M., Astroza, S., Pen­dyala, R.M., & Bhat, C.R. (2017). A behavioral choice model of the use of car-sharing and ride-sourcing ser­vices. Transportation, 44(6), 1307–1323. https://doi.org/10.1007/s11116-017-9797-8 Diefendorff, J. M., Brown, D. J., Kamin, A. M., & Lord, R. G. (2002). Examining the roles of job involvement and work centrality in predicting organizational citi­zenship behaviors and job performance. Journal of Organizational Behavior, 23(1), 93–108. https://doi.org/10.1002/job.123. Dimitriades, Z. S. (2007). The influence of service climate and job involvement on customer oriented organiza­tional citizenship behavior in Greek service organiza­tions: A survey. Employee Relations, 29(5), 469–491. https://doi.org/10.1108/01425450710776290 Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. (1986). Perceived organizational support. Journal of Applied Psychology, 71(3), 500-507. https://doi.org/10.1037/0021-9010.71.3.500 Fielbaum, A., & Tirachini, A. (2020). The sharing econo­my and the job market: the case of ride-hailing driv­ers in Chile. Transportation, (June 2020). https://doi.org/10.1007/s11116-020-10127-7 Fornell, C., & Larcker, D. F. (1981). Evaluating structur­al equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.2307/3151312 Frei, F.X. (2008). The four things a service business must get right. Harvard Business Review, 86(4), pp. 70-80. Furunes, T., & Mkono, M. (2019). Service-delivery suc­cess and failure under the sharing economy. Interna­tional Journal of Contemporary Hospitality Man­agement, 31(8), 3352-3370. https://doi.org/10.1108/IJCHM-06-2018-0532 Gleim, M.R., Johnson, C.M., & Lawson, S.J. (2019). Sharers and sellers: A multi-group examination of gig economy workers’ perceptions. Journal of Busi­ness Research, 98 (May 2019), 142-152. https://doi.org/10.1016/j.jbusres.2019.01.041 Grab, (2019). Update on Cancellation Policy. Retrieved from https://www.grab.com/ph/blog/update-on-can­ cellation-policy. April 26th, 2019. Hair, J.F., Black, W.C., Babin, B.J. & Anderson, R.E. (2010). Multivariate Data Analysis. 7th Edition, Pear­son, New York. Hamenda, A. (2019). An integrated model of service quality, price fairness, ethical practice and customer perceived values for customer satisfaction of sharing economy platform. International Journal of Business and Society, 19(3), 709-724. Hogan, N. L., Lambert, E. G., & Griffin, M. L. (2013). Loyalty, Love, and Investments: The Impact of Job Outcomes on the Organizational Commitment of Cor­rectional Staff. Criminal Justice and Behavior, 40(4), 355–375. https://doi.org/10.1177/0093854812469944 Ju, Y., Back, K.J, Choi, Y., & Lee, J.S. (2019). Exploring Airbnb service quality attributes and their asymmetric effects on customer satisfaction. International Journal of Hospitality Management, 77(1), 342-352. https://doi.org/10.1016/j.ijhm.2018.07.014 Kandampully, J., & Menguc, B. (2000). Managerial prac­tices to sustain service quality: an empirical investi­gation of New Zealand service firms. Marketing In­telligence and Planning, 18(4), 175-184. https://doi.org/10.1108/02634500010333299 Kanungo, R. (1982). Measurement of job and work in­volvement. Journal of Applied Psychology, 67(3), 341-349. https://doi.org/10.1037/0021-9010.67.3.341 Karatepe, O.M., & Karadas, G. (2012). The effect of man­agement commitment to service quality on job embed­dedness and performance outcomes. Journal of Busi­ness Economics and Management, 13(4), 614-636. https://doi.org/10.3846/16111699.2011.620159 Karn, K.S., & Hutson, W.E. (2019). Getting Us There Ride-Hailing Systems from the Drivers’ Perspec­tives. Ethnographic Praxis in Industry Conference Proceedings, 302–317. https://doi.org/10.1111/1559-8918.2019.01287 Katz, V. (2015) ‘Regulating the sharing economy’, Berke­ley Technology Law Journal, 30(385), 1067-1126. Kim, H. J., Tavitiyaman, P., & Kim, W. G. (2009). The effect of management commitment to service on employee service behaviors: the mediating role of job satisfaction. Journal of Hospitality and Tour­ism Research, 33(3), 369–390. https://dx.doi.org/10.1177/1096348009338530 Lambert, E. G., & Paoline, E. A. III. (2012). Explor­ing potential antecedents of job involvement: An exploratory study among jail staff. Criminal Jus­tice and Behavior, 39(3), 264–286. https://doi.org/10.1177/0093854811433533 Lawler, E., & Hall, D. (1970). Relationship of job char­acteristics to job involvement, satisfaction, and intrin­sic motivation. Journal of Applied Psychology, 54(4), 305-312. https://doi.org/10.1037/h0029692 Lawler, E. (1992). The ultimate advantage: Creating the high-involvement organization, San Francisco: Jos­sey-Bass. Lee, S., Barker, S., & Kandampully, J. (2003). Technology, service quality, and customer loyalty in hotels: Austral­ian managerial perspectives. Managing Service Qual­ity: An International Journal, 13(5), 423-432. https://doi.org/10.1108/09604520310495886 Lin, Pearl. M.C, Au, W.C., Leung, Vicky. T.Y., & Peng, K.L. (2020). Exploring the meaning of work within the sharing economy: A case of food-delivery workers. International Journal of Hospitality Management, 91 (102686). https://doi.org/10.1016/j.ijhm.2020.102686. Liu W, He C, Jiang Y, Ji R., & Zhai X. (2020). Effect of Gig Workers’ Psychological Contract Fulfillment on Their Task Performance in a Sharing Economy—A Perspective from the Mediation of Organizational Identification and the Moderation of Length of Service. International Journal of Environmental Research and Public Health, 17(7):2208. https://doi.org/10.3390/ijerph17072208 Lytle, R.S., Hom, P.W., & Mowka, M.P. (1998). SERV*OR: Managerial measures of organizational service orien­tation. Journal of Retailing, 74(4), 1–15. https://doi.org/10.1016/S0022-4359(99)80104-3 Mrayyan, M.T., & Al-Faouri, I. (2008). Predictors of ca­reer commitment and job performance of Jordanian nurses. Journal of Nursing Management, 16, 246-256. https://doi.org/10.1111/j.1365-2834.2007.00797.x Parasuraman, A. (1987). Customer-oriented corporate cul­tures are crucial to services marketing success. Jour­nal of Services Marketing, 1(1), 39-46. https://doi.org/10.1108/eb059587 Peticca-Harris, A., deGama, N., & Ravishankar, M. N. (2020). Post-capitalist precarious work and those in the ‘drivers’ seat: Exploring the motivations and lived expe­riences of Uber drivers in Canada. Organization, 27(1), 36–59. https://doi.org/10.1177/1350508418757332 Priporas, C.V., Stylos, N. Vedanthachari, L.N., & Santi­watana, P. (2017). Service quality, satisfaction, and customer loyalty in Airbnb accommodation in Thai­land. International Journal of Tourism Research, 19(6), 693-704. https://doi.org/10.1002/jtr.2141 Priporas, C.V., Stylos, N., Rahimi, R., & Vedanthachari, L. (2017). Unraveling the diverse nature of service quality in a sharing economy: A social exchange the­ory perspective of Airbnb accommodation. Interna­tional Journal of Contemporary Hospitality Man­agement, 29(9), 2279-2301. https://doi.org/10.1108/IJCHM-08-2016-0420 Rayle, L., Dai, D., Chan, N., Cervero, R., & Shaheen, S. (2016). Just a better taxi? A survey-based comparison of taxis, transit, and ride-sourcing services in San Fran­cisco. Transport Policy, 45(1), 168–178. https://doi.org/10.1016/j.tranpol.2015.10.004 Rhoades, L., & Eisenberger, R. (2002). Perceived organ­izational support: A review of the literature. Journal of Applied Psychology, 87(4), 698–714. https://doi.org/10.1037/0021-9010.87.4.698 Rod, M., & Ashill, N.J. (2010). Management commitment to service quality and service recovery performance: A study of frontline employees in public and private hospitals. International Journal of Pharmaceutical and Healthcare Marketing, 4(1), 84-103. https://doi.org/10.1108/17506121011036042 Rosenblat, A., & L. Stark. (2016). Algorithmic Labor and Information Asymmetries: A Case Study of Uber’s Drivers. International Journal of Communication, 10, 3758–3784. Rosenblat, A., Levy, K.E., Barocas, S., & Hwang, T. (2017). Discriminating Tastes: Uber’s Customer Rat­ings as Vehicles for Workplace Discrimination. Poli­cy & Internet, 9(3), 256-279. https://doi.org/10.1002/poi3.153 Shokoohyar, S. (2018). Ride-sharing platforms from drivers’ perspective: Evidence from Uber and Lyft drivers. International Journal of Data and Net­work Science, 2(4), 89–98. https://doi.org/10.5267/j.ijdns.2018.10.001 Shokoohyar, S, Sobhani, A., & Ramezanpour Nargesi, SR (2020). On the determinants of Uber accessibility and its spatial distribution: Evidence from Uber in Phila­delphia. WIREs Data Mining Knowledge Discovery, 10(4), 1-15. https://doi.org/10.1002/widm.1362 Shokoohyar, S, Sobhani, A., & Sobhani, A. (2020). Im­pacts of trip characteristics and weather condition on ride-sourcing network: Evidence from Uber and Lyft. Research in Transportation Economics, 80 (100820). https://doi.org/10.1016/j.retrec.2020.100820 Standing, C., Standing, S., & Biermann, S. (2019). The im­plications of the sharing economy for transport. Trans­port Reviews, 39(2), 226-242. https://doi.org/10.1080/01441647.2018.1450307 Susskind, A. M., Borchgrevink, C. P., Kacmar, K. M., & Brymer, R. A. (2000). Customer service employees’ behavioral intentions and attitudes: An examination of construct validity and a path model. Internation­al Journal of Hospitality Management, 19(1), 53-77. https://doi.org/10.1016/S0278-4319(99)00030-4 Uber (2019). Introducing Uber Pro. Retrieved from https://www.uber.com/us/en/drive/uber-pro. Accessed on 29 Feb. 2020. Ueno, A. (2008). Is empowerment really a contributory factor to service quality? The Service Industries Journal, 28(9), 1321-1337. https://doi.org/10.1080/02642060802230262 Valente, E., Patrus, R., & Córdova Guimarães, R. (2019). Sharing economy: becoming an Uber driver in a de­veloping country. Revista de Gestão, 26(2), 143-160. https://doi.org/10.1108/REGE-08-2018-0088 Wood, D. A. (1974). Effects of worker orientation differ­ences on job attitude correlates. Journal of Applied Psychology, 59(1), 54–60. https://doi.org/10.1037/h0035829 Yang, S., Song, Y., Chen, S., & Xia, X. (2017). Why are customers loyal in sharing-economy services? A rela­tional benefit perspective. Journal of Services Mar­keting, 31(1), 48-62. https://doi.org/10.1108/JSM-01-2016-0042 Yavas, U., Karatepe, O.M., Avci, T., & Tekinkus, M. (2003). Antecedents and outcomes of service re­covery performance: an empirical study of frontline employees in Turkish banks. The International Jour­nal of Bank Marketing, 21(4/5), 255-65. https://doi.org/10.1108/02652320310488439 Zervas, G., Proserpio, D., & Byers, J. W. (2017). The Rise of the Sharing Economy: Estimating the Impact of Airbnb on the Hotel Industry. Journal of Market­ing Research, 54(5), 687-705. https://doi.org/10.1509/jmr.15.0204 Zhang, N., Li, M., Gong, Z., & Xu, D. (2019). Ef­fects of ethical leadership on nurses’ service behav­iors. Nursing Ethics, 26(6), 1861–1872. https://doi.org/10.1177/0969733018787220 Zuo, W., Zhu, W., Chen, S., & He, X. (2019). Service quality management of online car-hailing based on PCN in the sharing economy. Electronic Commerce Research and Application, 34, Article 100827. https://doi.org/10.1016/j.elerap.2019.100827 Phuong Tran Huy is an assistant professor who currently works at Faculty of Human Resource Economics and Management, National Economics University, Hanoi Vietnam. Phuong holds a Ph.D. degree in management from Kobe University, Hyogo, Japan. Phuong does research in Organizational Behavior, Human Resources Management and Labor Economics. Phuong’s current project is “High performance work system in Vietnam”. Hong Chuong Pham is an Associate Professor, who has been working for more than 30 years at the National Economics University (NEU) in Vietnam as a lecturer in Tourism and Hospitality Management. Chuong has authored and co-authored numerous papers in Tourism Economics, Tourism and Hospitality Management and Tourism Development. Vpliv zavezanosti vodstva k kakovosti storitev na vedenje voznikov pri storitvah vožnje: primer GrabCar v Vietnamu Ozadje in namen: Znano je, da zavezanost vodstva h kakovosti storitev (Management Commitment to Service Qua­lity - MSCQ) pozitivno vpliva na kakovost storitev in ravnanje zaposlenih v razlicnih storitvenih panogah. V kontekstu ekonomije delitve se odnos med podjetjem in ponudniki storitev razlikuje od tradicionalnega delovnega razmerja. Pri storitvah vožnje so vozniki v glavnem izvajalci in ne kot zaposleni. Zato je treba razumeti, ali MSCQ vpliva na kakovost v kontekstu storitve vožnje. Metodologija: Anketirali smo 214 voznikov GrabCar v Vietnamu z uporabo spletne in klasicne ankete. Za analizo podatkov je bilo uporabljeno modeliranje strukturnih enacb z delnimi najmanjšimi kvadrati (PLS-SEM). Rezultati: Ugotovitve kažejo, da tri razsežnosti MCSQ, in sicer sistem nagrajevanja, tehnološka podpora in orga­nizacijska podpora, mocno vplivajo na vedenje voznikov pri storitvah. Poleg tega ima zaposlitev posredni vpliv v razmerju med MCSQ in vedenjem voznikov. Zakljucek: Ta študija razširja prejšnje raziskave o MCSQ na storitev vožnje in potrjuje vlogo zaposlitve kot po­membnega mehanizma za izboljšanje kakovosti storitev, ki jih zagotavljajo vozniki. Zaradi znacilnosti odnosa med podjetjem in ponudniki storitev v ekonomiji delitve je treba podrobneje preuciti mehanizme s katerimi MCSQ vpliva na kvaliteto storitev ponudnikov. Kljucne besede: Zavezanost vodstva kakovosti storitev (MCSQ), Storitveno vedenje, Zaposlitev, Storitev vožnje, Ekonomija delitve, Vietnam. Appendix Service-based reward system 1. I receive positive recognition from Grab when I excel in serving customers. 2. Grab has financial incentives for service excellence. 3. My income depends on the quality of service I deliver. Service technology support 1. Grab has the necessary technology support to serve my customers effectively. 2. Grab has ‘state of the art’ technology to enhance our service quality. 3. Grab allocates a significant amount of money on technology to support my efforts to deliver effective service. 4. Grab works hard to make our systems and processes more customer friendly. Perceived organizational support 1. Grab values my contribution to its well-being. 2. Grab fails to appreciate any extra effort from me. (Reverse) 3. Grab would ignore any complaint from me. (Reverse) 4. Grab really cares about my well-being. 5. Even if I did the best job possible, Grab would fail to notice. (Reverse) 6. Grab cares about my general satisfaction at work. 7. Grab shows very little concern for me. (Reverse) 8. The organization takes pride in my accomplishments at work Job involvement 1. The most important things that happen to me involve my present job. 2. To me, my job is only a small part of who I am. (Reverse) 3. I am very much involved personally in my job. 4. I live, eat and breathe my job. 5. Most of my interests are centered around my job. 6. I have very strong ties with my present job which would be very difficult to break. 7. Usually I feel detached from my job. (Reverse) 8. Most of my personal life goals are job-oriented. 9. I consider my job to be very central to my life. 10. I like to be really involved in my job most of the time. Service behavior 1. I provide transportation services according to the job responsibilities prescribed by Grab 2. I am familiar with the service procedures of different car-hailing requests. 3. I can independently fulfill responsibilities to customers as specified in my job description. 4. I can satisfy the needs of customers with pleasure in my duties. 5. I have the good manners as a driver with neat and professional appearance. DOI: 10.2478/orga-2021-0010 The Impact of Internal Knowledge Sharing on Sales Department’s Innovativeness and New Product Commercialization Erik RUŽIC, Dragan BENAZIC Juraj Dobrila University of Pula, Faculty of Economics and Tourism “Dr. Mijo Mirkovic”, Pula, Croatia, erik.ruzic@unipu.hr, dragan.benazic@unipu.hr Background and Purpose: Innovativeness and new product commercialization are highly important for companies. Therefore, a deep understanding of the impact of all potentially influential drivers of success is critical. The purpose of the paper is to explore the impact of internal knowledge sharing on new product selling and sales innovativeness as well as the impact of empowerment on internal knowledge sharing and, indirectly, on new product selling and sales innovativeness. Design/Methodology/Approach: The research encompassed 101 salespeople working at the top 1000 value add­ed creators in Croatia. The questionnaire was developed and adapted using four scales, to assess internal knowl­edge sharing, new product selling, sales innovativeness, and empowerment. The data was analyzed by using the PLS-SEM method to examine the relationships between constructs. Results: As evidenced by the survey results, internal knowledge sharing positively impacts new product selling and sales department’s innovativeness, and empowerment is positively linked to internal knowledge sharing and, indirectly, to new product selling and sales department’s innovativeness. Conclusion: Managers should underpin different activities in order to enhance empowerment and internal knowl­edge sharing with the aim to affect companies’ performance in commercialization of a new product and sales depart­ment’s innovativeness. Future research could include moderator variables between the empowerment construct and the internal knowledge sharing construct and deepen the insight into the type of information shared, the dynamics of sharing and the barriers in the process, and other factors that positively affect knowledge sharing. Keywords: New product selling, Sales department’s innovativeness, Internal knowledge sharing, Empowerment, Salespeople. 1 Introduction Although the failure rate of new products is high, in­tense global competition and demanding customers put pressure on companies and force them to launch new prod­ucts (Sharma & Sagar, 2017). However, 40 to 90% of all new products fail in the marketplace, causing a substantial financial loss to companies (Borgh & Schepers, 2017). Yet at the same time, newly launched products and services generate the highest revenues for companies (Petrariu et al., 2013). As a result, despite all obstacles and high poten­tial losses, market circumstances and potential revenues push companies to develop new products. An important stage in the new product development process is the last phase, commercialization, and although successful new product selling is one of the most important missions of a company (Ahearne, et al., 2010), it is also one of the most challenging tasks for salespeople. In fact, they are not prone to selling a new product; they prefer selling the existing ones (Wieseke et al., 2007). On the other hand, in a world full of similar and easy-to-copy products and services, innovativeness may be a differentiation tool (Healy et al., 2018) for salespeople and a source of competitive advantage. Innovativeness can work well within the sales department and is especially important when a new product is brought to the market and salespeople undertake the challenging task of selling it. In fact, innovativeness within a department is seen as a work environment that encourages employees’ innovative behavior within the department itself (Matsuo, 2005). In order to improve and facilitate the commercial­ization phase (i.e. new product selling) and to enhance salespeople’s creativity and flexibility (i.e. sales innova­tiveness), it becomes important to investigate the impact of a potentially significant driver of success. In fact, new product selling could be positively affected by knowledge sharing among salespeople as well as between sales force and other departments. Moreover, this openness and readi­ness to share information could have the potential to foster creativity and flexibility which, in turn, positively impact innovativeness within a sales department. Lastly, an environment in which employees and man­agers share power and the organizational culture encour­ages such sharing, (i.e. employees are empowered) can create an environment of openness and of people willing to share knowledge across the organization. And by doing this, indirectly and/or in synergy, it could lead to better innovation and facilitate new product selling. The main aim of the study is to investigate the impact of internal knowledge sharing on new product selling and sales department’s innovativeness as well as the impact of empowerment on internal knowledge sharing and, indi­rectly, on new product selling and sale innovativeness in an emerging and small market context such as Croatian. In fact, innovations have proven to be an important contrib­utor to competitiveness and economic growth of countries (Petrariu et al., 2013). Moreover, previous studies (Švarc & Dabic, 2019) pointed out many weaknesses in technolo­gy transfers and innovation efforts in the Republic of Cro­atia, whose R&D intensity (i.e. expenditure on R&D as a percentage of GDP) is only 0.97%. There is a large body of research on internal knowl­edge sharing, empowerment, new product selling (Wiese­ke et al., 2007; Sharma & Sagar, 2017; Hohenberg & Hahn, 2018) and on the innovation topic conducted in less developed countries (Central and Eastern Europe) and in emerging innovation systems (Petrariu et al., 2013; Sto­jcic et al., 2018; Stojcic, 2020). However, to the best of authors’ knowledge, there is little evidence on the linkages between these constructs among salespeople in the spe­cific context of an emerging market. Hence, the present research will fill this gap. Sales people were chosen to be the focus of this study because of their role as boundary spanners and because of their critical importance in the last phase of the new prod­uct development process (i.e. commercialization). Following the introduction, Section 2 provides the literature review and outlines the development of the hy­potheses. Section 3 presents the methodology. Section 4 describes the empirical data and analysis, the measurement model and the structural model. Section 5 discusses the empirical results and implications. Finally, Section 6 pro­vides the concluding remarks, limitations and directions for further research. 2 Literature Review and Hypothesis Development 2.1 Internal Knowledge Sharing, New Product Selling and Sales Department’s Innovativeness In today’s ultra-competitive environment, knowledge is seen as a strategic resource whose transfer, if performed well, provides multiple benefits to companies, such as gaining competitive advantage (Hume & Hume, 2015), strengthening entrepreneurial orientation (DeClerq, et al. , 2013), and avoiding mistakes and redundancy (Rafiq & Ahmed, 2006). Moreover, it affects organizational perfor­mance and effectiveness (Kim & Lee, 2006), reduces time to market (Rafiq & Ahmed, 2006), and strengthens a com­pany’s ability to meet customer needs (Kim & Lee, 2006). In fact, many authors pointed out the positive effects of cross-functional integration and information sharing on achieving new product success (Arfi et al., 2019). Furthermore, studies showed the importance of knowl­edge sharing in NPD processes, within the sales department and between sales and other departments (van den Berg et al., 2014) due to first-hand information and insights into customer requirements gained by sales staff. Crawford and DiBenedetto (2011) pointed out that information collect­ed from the marketplace is an important source of infor­mation in the problem-solving process (in relation to the new product) and Matsuo (2018) saw salespeople as the central unit of interaction among different stakeholders. While Sharma and Sagar (2017) highlighted their crucial role once the product is launched, Ahearne et al. (2010) emphasized the unique position of salespeople in generat­ing knowledge due to their regular interactions with cus­tomers. Although these interactions are under threat due to technology and social media which negatively impact the opportunities for direct interactivity between salesforce and potential customers Sharma and Sagar (2017), they were referred to as “knowledge brokers” by Verbeke and Masih (2020). The quality of such interactions is especial­ly important in the commercialization phase. In fact, it is in this phase that salespeople often encounter resistance and their role as the primary source of information is to over­come resistance and facilitate the adoption of the product. Moreover, it was suggested (Ahearne et al., 2010) that salespeople can adjust their planning and behavior based on the knowledge and insights acquired and might even be innovative (if allowed). Zhou and Li (2009) emphasized that internal knowledge exchange is more relevant for rad­ical innovation than external knowledge acquisition if the organizational knowledge base is broad (the opposite ap­plies if the knowledge base is deep). Tsai (2001) pointed out that better knowledge access by an organizational unit, through knowledge exchange and transfer, affects its inno­vation and performance. A positive relationship between knowledge sharing and performance as well as between knowledge sharing and innovation was confirmed by dif­ferent studies (Wang, & Wang, 2012; Yiu et al., 2020). Moreover, Caloghirou et al. (2004) highlighted that read­iness for knowledge exchange and internal ability to rec­ognize and exploit external knowledge improve innovative performance. Yiu et al. (2020) showed that service-orient­ed companies have higher innovation performance thanks to their learning from partner relationships. Matsuo (2005) in his study highlighted the need for knowledge management systems in order to enable in­formation sharing among salespeople and throughout the entire organization. Hohenberg and Hahn (2018) in their study pointed out that internalized new product selling motivation is crucial for performance. Borgh and Schepers (2017) highlighted the need for appropriate information sharing by a sales manager with the sales force once a new product is launched. However, knowledge management, as well as the NPD process (Helmy et al., 2019), generates costs, so it must be related to economic benefits (either revenue growth or decreased costs) or competitive advan­tage for the firm in order to be adopted. Van den Berg et.al. (2014) argued that in today’s knowledge-based economy, knowledge development and transfer within the company act as a booster for sales performance and strongly moti­vate salespeople to new product selling, which in the end influences sales and overall success as well. On the other hand, Haas et al. (2007) suggested that in some cases the salesforce may be overwhelmed by knowledge sharing, which negatively affects the success of new product sell­ing. Tang and Marinova (2020) pointed out the mixed ev­idence regarding the effects of knowledge sharing in NPD processes. Mulyana et al. (2019) stated that new products and services, as well as new ways of their promotion and dis­tribution, will represent the crucial factors for companies’ success, therefore marketing innovation is seen as a vital strategic orientation. According to Hendar et al. (2018), marketing innovativeness is crucial for performance. Studies conducted in various European countries have shown different challenges in the innovation process. Ac­cording to Švarc and Dabic (2019), the development of post-socialist economies depends on their ability to gen­erate and exploit innovation and their ability to transfer knowledge from scientific institutions to companies. Moreover, Švarc and Dabic (2019) highlighted that in Croatia, as a typical transition country with a specific transition process (socialism to capitalism) in comparison to other EU member states, technology transfer happens slowly. Prokop and Stejskal (2017), in their study based on CIS data and conducted in Germany, Portugal, Bulgaria and Slovenia, highlighted that companies put the impetus on gaining information from various partners and docu­ments and use it to innovate. Furthermore, Prokop and Stejskal (2017) analyzed the impact of soft knowledge infrastructure and HRST (Human Resources in Science and Technology) on the economic development of EU 28 economies in 2012 and of CEE economies between 2002 and 2012. The study showed that CEE countries were less effective than the rest of EU countries in using selected de­terminants of economic development. Stojcic et al. (2018), in their study conducted in Croatia, discussed different im­pacts of creativity in the innovation process. Hohenberg and Hahn (2018) stated that research on salesforce related factors during the commercialization phase is relatively scarce. In the light of the above, and with the aim of broaden­ing the knowledge of the yet insufficiently explored links between the described constructs among salespeople in the specific context of an emerging market, we hypothesize: Hypothesis 1: Internal knowledge sharing is positively related to new product selling. Hypothesis 2: Internal knowledge sharing is positively related to sales department’s innovativeness. 2.2 Empowerment and Internal Knowledge Sharing Despite the above-mentioned importance of knowl­edge transfer, there are a number of knowledge exchange barriers to consider in the attempt to improve internal shar­ing (Haas et al., 2007). Different sources (i.e. barriers), such as knowledge transferring tools, provider, receiver and context, were found threatening the cross-functional knowledge sharing during the NPD process (Hunag et al., 2008). DeClerq et al. (2013) mentioned that despite the many benefits of internal knowledge sharing, it is chal­lenging when it occurs between different organizational functions as is the case in NPD process which involves a large number of employees (Crawford & DiBenedetto, 2011). Researchers noted the key role of motivation in the process of knowledge sharing (Gressgard, 2015). More­over, the motivation for knowledge sharing derives from the individual’s beliefs and from the organization’s factors (shared values, norms, practices) that shape the individu­al’s beliefs (Bock et al., 2005). Furthermore, it was pointed out that knowledge shar­ing within organizations depends on the employee’s in­dividual behavior (Bock et al., 2005) and ability to share (Gressgard, 2015). Kim & Lee (2006) in their study on the three elements of organizational culture (i.e. vision and goals, trust, and social networks) pointed out that social networks positively impact employee knowledge-sharing capabilities. Similarly, Mu et al. (2016) argued that besides knowledge generation, a networking ability is required too. Chen et al. (2014) in their study on knowledge sharing in virtual community showed that trust positively affects the intention to share knowledge and that high altruism makes stronger the relationship between trust and knowl­edge sharing. Moreover, DeClerq et al. (2013) found that trust and goal congruence (between different functions) result in more internal knowledge sharing. Hume & Hume (2015) argued that for KM to be successful, trust, personal relevance and personnel satisfaction are needed. Besides, authors Bock et al. (2005, p. 88) pointed out that individuals are not prone to share knowledge; in fact, sharing knowledge is more an exception than a rule within organizations. In their study on barriers in cross-functional knowledge sharing during the NPD process, Huang et al. (2008) highlighted that the use of appropriate strategies (processual and classical) can reduce these barriers and by doing this reduce the related costs. Hohenberg and Hahn (2018) stressed that new product selling bonuses can en­hance the relationship between salespeople’s performance predisposition and new product financial performance, but weaken the relationship between salespeople’s learning predisposition and new product financial performance. Moreover, they showed that a periodic review strengthens the relationship between salespeople’s learning predispo­sition and new product financial performance. Wang et al. (2014) pointed out that evaluation and evaluation with reward have positive impact on knowledge sharing. They also stated that knowledge sharing is influenced by the interaction between evaluation plus reward and conscien­tiousness, neuroticism and openness to experience. Shar­ma and Sagar (2017) pointed out that salespeople when selling new products often face, among others, inefficient information flow, so sales managers should effectively cre­ate and spread knowledge about the new product. Gressgard (2015) in his study on motivational and organizational factors affecting knowledge sharing high­lighted empowerment as an important intrinsic motivation factor of knowledge exchange. Empowerment has been defined in many different ways. Conger & Kanungo (1988, p. 473) defined empow­ering as a managerial strategy or technique that strengthens self-determination need or self-efficacy belief of employ­ees and makes them feel more powerful. The employee’s empowered „state of mind“ may (Rafiq & Ahmed, 2006) consequently lead to positive work behavior (Abbasi et al., 2020). Empowerment is also seen as a multifaceted motivational factor which can lead to altruistic behavior such as knowledge sharing (Wang et al., 2019). Kang et al. (2017) in their study showed that positive and proactive knowledge sharing behavior occurs as a consequence of knowledge management system user empowerment. As mentioned above, empowerment is multifaceted and according to Spreitzer (1995), psychological empow­erment construct merits a special critical inquiry. Spreitzer (1995) pointed out that psychological empowerment refers to the individual state of strong motivation, sense of au­thority and ability to perform job duties. It is in fact an active motivational orientation and it motivates people in their performance while executing different proactive tasks (Kang et al., 2017). Abbasi et al. (2020) highlighted that psychological empowerment mediates the relation­ship between high performance work system and knowl­edge sharing behavior. Al-Omari et al. (2020) in their study pointed out the positive linkages between employ­ee empowerment and overall company performance and customer satisfaction. Grošelj et al. (2020) suggested that psychological empowerment moderates the relationship between leadership and innovative work behavior. Kang et al. (2017) considered the positive impact of psychological empowerment on different outcomes, such as effective­ness. According to Spreitzer (1995), two consequences of empowerment are effectiveness and innovative behavior. Arfi et al. (2019) examined a case in which empowering employees, among other changes in management style, fa­cilitated knowledge exchange and firm’s open innovation and performance. Furthermore, Helmy et al. (2019) point­ed out that knowledge sharing mediates the relationship between the dimension of psychological empowerment (competence, impact, self-determination) and innovative work behavior. Lastly, Kang et al. (2017) stated that there has been little research investigating what motivators im­pact proactive knowledge sharing. For a deeper understanding of the influence of empow­erment on knowledge sharing and its indirect effects on new product selling and innovativeness of salesforce, we posit the following hypotheses: Hypothesis 3: Empowerment is positively related to in­ternal knowledge sharing. H3a: Empowerment is positively related to new prod­uct selling through internal knowledge sharing H3b: Empowerment is positively related to sales de­partment’s innovativeness through internal knowledge sharing 3 Methodology 3.1 Measurement Scales The data for the present research was taken from a broader study conducted among salespeople in Croatia. The questionnaire was developed using scales validated in previous research and tested by the respective authors. The measurement items used to assess Internal Knowledge were the ones developed by DeClerq et al. (2013) (7 items). In the questions related to Internal Knowledge Sharing, re­spondents were asked to provide their attitudes toward the relationship and collaboration between their commercial function and technically oriented functions. The Sales De­partment’s Innovativeness level was measured using the scales developed by Matsuo (2005) (6 items), while for the assessment of the New product selling attitudes we used the van den Berg et al. (2014) scales (3 items). The meas­urement items used to measure Empowerment were the ones developed and tested by Spreitzer (1995) (12 items). All the scales were translated into Croatian language by an expert linguist and demographic questions were added. Respondents were asked to rate all the statements using the 7-point Likert scale (1-strongly disagree to 7-strongly agree). The data for quantitative research was collected in April and May of 2015. 3.2 Sample The research on the impact of internal knowledge shar­ing on new product selling and sales innovation and the impact of empowerment on internal knowledge sharing was conducted on a sample of 101 key informants. The key informants were sales staff working at different lev­els in businesses operating in different sectors. The key informants were identified from a secondary source of in­formation, i.e. a publication issued by a business magazine listing the top 1000 value added creators in the Republic of Croatia. The research was conducted using a highly structured questionnaire sent to organizations randomly selected from the above-mentioned list. The questionnaire was sent via email asking for it to be forwarded to sales­people at different levels within the organization. The sam­ple structure regarding gender, age, level of education and years of sales experience of the key informants is shown in Table 1. Since the questioning method was used and all the questions in the questionnaire were answered by the same key informants (common source), there was a possibility of common method variance. Therefore, when designing the questionnaire, several procedural and statistical rem­edies according to Podsakoff et al. (2003) were applied to avoid the common method variance issue. The key in­formants were guaranteed anonymity and the instructions emphasized that there were no right or wrong answers. Furthermore, questions relating to individual constructs were neatly divided by subheadings. In terms of the sta­tistical techniques used, the Harman’s single-factor test was conducted by performing an unrotated exploratory factor analysis. According to the test performed, all indi­cator variables resulted in one general factor accounting for 42.46% of the total variance in all variables, which is less than 50%, suggesting that common method variance would not be a major concern in this study. In addition, a confirmatory factor analysis was conducted, according to which all indicator variables were combined in one la­tent variable. The specified single-factor model was a poor fit to the data (RMSEA = 0.20, CFI =0.458; <0.90, TLI = 0.414; <0.90 and SRMR = 0.15; <0.08), thus further sup­porting a low likelihood of common method bias in this study. In the sections below, the psychometric properties of the measurement scales are tested, followed by an anal­ysis of the structural model in relation to the proposed re­search hypotheses. 4 Research Results 4.1 Measurement Model Analysis Before testing the hypotheses, we assessed the internal consistency reliability and the convergent and discrimi­nant validity of the measurement scales. Data was analyz­ed using the PLS-SEM method instead of the traditional CB-SEM analysis, because it offers greater flexibility re­garding non-compliance under the assumption of normal variable distribution and because the PLS method gives better results in case of small sample sizes. Moreover, the PLS method is more appropriate when the research is primarily aimed at determining the predictive ability of endogenous constructs (Hair et al., 2012, Hair et al., 2014), rather than for studies whose primary focus is theo­ry testing. Data was analyzed using SmartPLS 3.2.8. soft­ware (Ringle et al., 2015). The descriptive statistics of all items included in the analysis are reported in Table 2. The bootstrapping method with 5,000 subsamples, as recom­mended by Hair et al. (2019, p. 149), was used to evaluate the measurement scales and, subsequently, to analyze the structural models. All measurement models are specified as reflective measurement models – Mode A (Hair et al., 2019, p. 46) - based on previous research using the same measurement scales. Table 1 shows item outer loadings, CR and AVE indicators. The EMP6 (I have mastered the skills necessary for my job.), EMP7 (I have significant au­tonomy in determining how I do my job.), EMP8 (I can decide on my own how to go about doing my work.) and EMP9 (I have considerable opportunity for independence and freedom in how I do my job.) variables of the con­struct Empowerment were excluded from the analysis be­cause their factor loadings ranged from 0.4 to 0.6, and their exclusion led to an increase in Composite Reliability (CR) and Average Variance Extracted (AVE) above the thresh­old values of 0.8 and 0.5, respectively. As shown in Table 3 below, all investigated constructs had satisfactory Cronbach’s alpha, Composite Reliability (CR) and Average Variance Extracted (AVE) values. Cron­bach’s alpha and CR coefficients ranged between 0.8 and 0.96 and they all exceeded the recommended cut-off of 0.8. Besides, all of the AVE scores were above 0.5, ranging from 0.53 to 0.80. According to the below data, all the con­structs showed an acceptable level of internal consistency reliability and convergent validity. The cross-loading analysis as well as the Fornell-Larck­er criterion (1981) and the Heterotrait-Monotrait (HTMT) ratio confidence intervals were used to assess the discri­minant validity. The results of the cross-loading analysis showed that the outer loadings were high on their respec­tive constructs, but low on all other constructs. The For­nell-Larcker criterion and the HTMT ratio are shown in Table 4. According to the data presented in Table 4, the For­nell-Larcker criterion was met and the square root of the AVE for each construct was greater than the correlation between the construct and all other constructs. The HTMT values between constructs were all below 0.85, indicating that none of the HTMT confidence interval bias-correct­ed values included the value of 1. All the above data sug­gest satisfactory discriminant validity of the measurement scales. The next section provides an analysis of the struc­tural model. 4.2 Structural Model Analysis This research is based on the assumption of a positive impact of Empowerment through Internal Knowledge Sharing on New Product Development and Sales Innova­tion (Figure 1). A VIF test was used to check for multicollinearity among endogenous constructs. The VIF for each pair of endogenous constructs was below the threshold of 5, so it could be concluded that multicollinearity among endoge­nous constructs would not be a major concern in this study. The results of the structural model analysis are illustrated in Table 5. The results of the structural model analysis supported all the proposed hypotheses. Direct effects were analyzed within the framework of the model. In this respect, Em­powerment (EMP) had a statistically significant positive direct effect on Internal Knowledge Sharing (IKS) (H1: ß=0.555; t=7.192; p<0.05), which in return had a positive direct effect on New Product Selling (NPS) (H3: ß=0.343; t=3.106; p<0.05) and Sales Department’s Innovativeness (SALINNOV) (H4: ß=0.633; t=10.209; p<0.05). Likewise, two statistically significant positive indirect effects were found, namely Empowerment through Internal Knowledge Sharing positively affects Sales Department’s Innovative­ness (ßindirect=0.351; t=5.095, p<0.05) and New Product Selling (ßindirect=0.351; t=5.095, p<0.05). The effect size of the construct Empowerment was moderate, it explained 31% of Internal Knowledge Sharing construct variance (R2=0.31), while Internal Knowledge Sharing explained 12% of variance in the endogenous construct of New Prod­uct Selling (R2=0.12) and its effect size was moderate i.e. explained 40% of variance in Sales Innovation (R2=0.40). In terms of the effect size, according to Cohen (1988) there was a large effect size (f2=0.44) of the construct of Em­powerment on Internal Knowledge Sharing and a medium (f2=0.13) effect size of Internal Knowledge Sharing on New Product Selling, whereas the effect size of the latter construct on Sales Innovation was strong (f2=0.67). The predictive relevance of the model was examined using the blindfolding procedure. All Q2 values were greater than 0, thus providing support for the model’s predictive rel­evance in respect of all endogenous constructs. The next section discusses in more detail the theoretical and practi­cal implications of the research findings. 5 Discussion and Implications The present study has revealed that internal knowledge sharing is positively linked with new product selling and sales department’s innovativeness. The research results are in line with Tsai (2001), Wang and Wang, (2012) and Yiu et al. (2020); namely, they argued about the impact of knowledge sharing on innovation and performance. These links were confirmed within the sales department (i.e. among salespeople). Our research can be added to Prokop and Stejskal’s study (2017), as it shows that sharing of the knowledge acquired in the marketplace between sales and technical department in the specific context of Croa­tian economy can bring innovation (in sales) and enhance new product selling. Moreover, the proven positive effect of empowerment on internal knowledge sharing and, indi­rectly, on new product selling and sales innovativeness is in line with the research conducted by Kang et al. (2017), who pointed out the positive impact of empowerment on proactive tasks, and with van den Berg et al. (2014) study focusing on the role of knowledge transfer on performance and new product selling. A number of practical implications arise from this study. As stated above, regular knowledge sharing im­proves the final stage of the NPD process and positively af­fects the innovativeness of sales department. Based on this evidence, top management and sales management should jointly adopt appropriate internal knowledge practices, develop an internal knowledge sharing system and under­take different activities in order to facilitate and encour­age knowledge sharing, such as developing an appropriate organizational culture, rewarding knowledge sharing, or conducting internal marketing activities. Salespeople get first-hand feedback from customers, so it is important to enable and encourage sharing of such valuable informa­tion, especially with the technical department. Shared in­formation may allow faster response to customer needs. Managers should empower knowledge system users and encourage (i.e. empower) knowledge exchange between employees while selling new product, with the aim of fos­tering creativity and flexibility among them and bettering the process. However, this striving can be a challenge as well; there is a number of different obstacles to overcome in the knowledge sharing enhancement efforts. Attitudes, beliefs and motivation are important drivers for knowledge sharing and according to the present study, empowered people are more prone to share knowledge. Ac­cordingly, sales managers should allow discretion which can be formalized (documents, procedures) or implied (organizational culture) and encouraged through training and personal (i.e. managers’) example, meaning that by doing this they’ll impact employee’s knowledge-sharing behavior. Finally, empowerment will lead to better product selling and more creativity and flexibility among salespeo­ple. These outcomes will positively impact the company’s overall performance and contribute to the national econo­my. Therefore, salespeople should be empowered, but ac­cording to past studies (Sharma & Sagar, 2017; Homburg & Hahn, 2018), effective knowledge transfer from sales managers to their sales teams and an appropriate financial policy framework (i.e. bonuses) should be in place as well. The present study contributes to the body of knowledge with new insights on the topic from the sales department’s perspective. In fact, while previous research has shown the effects of empowerment and internal knowledge sharing in different settings and the effects of these factors with dif­ferent outcomes (Al-Omari et al., 2020; Yiu et al., 2020), the links between empowerment and internal knowledge sharing as well as between these constructs and new prod­uct selling and sales innovativeness have not been investi­gated in salespeople in a specific emerging market context. Because of the specific role of salespeople in the commer­cialization phase and the need for innovation within the sales department itself, many companies struggle with their sales force. Sales personnel as boundary spanners collect a large amount of information and play a crucial role in product commercialization. With the aim of addressing the above gap, we conducted the current research. This study’s findings corroborate previous research, but also contribute to the literature by highlighting the relevant links in the specific sales and economic context. 6 Conclusions, Limitations and Future Research Innovations are crucial both for companies and for national economies. Companies should innovate due to intense competition and customer expectancy and gov­ernments should innovate for the sake of their country’s economic growth. But there are a number of challenges on this path and all the steps and specificities should be examined in order to better this process. Our research dealt with the last stage of the process - new product selling and the innovativeness of people who sell the products. In fact, in this process, employees, especially those who are in di­rect contact with potential customers, play an important role. Their innovativeness and efforts can be particularly valuable when the product faces the market for the first time (i.e. in the commercialization phase). Salespeople’s role has already been recognized as important, but at the same time a number of issues faced by companies dealing with salesforce selling new product have been pointed out. Despite the above-mentioned importance of innovation, different challenges have been identified in the context of an emerging market economy. Lastly, because of a number of similar products on the market, creativity and flexibility are important for salespeople and they should as well be boosted. Based on the study’s evidence, this outcome can be influenced by knowledge sharing and empowerment, while knowledge sharing can be affected by adequate em­powerment of salespeople. The present findings may help companies in their striv­ing to avoid losses and launch products more successfully. Moreover, the addressed process characterized by high in­vestment and failure rates could be improved by the joint efforts of practitioners and scientists. The present study has some limitations that may serve as the basis for future re­search. A larger study sample with an appropriate structure of businesses in terms of size and sector would provide more valid and more reliable research findings and a better generalization of the results. Moreover, since in the larger portion of the sample only one salesperson per company was surveyed due to time and cost limitations, involving more employees at different levels in the sales team of an organization, i.e. the use of the triangulation method, would yield more accurate results. Furthermore, due to the sample size, the authors did not check the unobserved heterogeneity that may reduce the accuracy of the conclusions of the study in case of dif­ferent segments - groups of key informants. The study was conducted on cross-sectional data i.e. it lacks an analysis of the observed phenomena from a dynamic perspective, which should be covered by future research in order to obtain a better insight into the causal relationships among the investigated constructs. The knowledge gained within this study could be broadened in future research by including some moderat­ing variables between the Empowerment and the Internal Knowledge Sharing constructs, such as organizational cul­ture of the company, job satisfaction, personality traits of salespeople, selling style, sales staff motivation, job burn­out syndrome. In addition, it would be useful if future research en­compassed the types of information shared between man­agers, salesforce and other employees, and the impact of such information on sales process innovation and new product development. Acknowledgment “This paper is a result of the scientific project The Use and Perceptions of Holistic Marketing in Croatia support­ed by the Faculty of Economics and Tourism “Dr. Mijo Mirkovic”, Juraj Dobrila University of Pula. Any opinions, findings, and conclusions or recommendations expressed in this paper are those of the author(s) and do not neces­sarily reflect the views of the Faculty of Economics and Tourism “Dr. Mijo Mirkovic” Pula.” Literature 1Received: 2nd June 2020; revised: 8th March 2021; accepted: 19th March 2021 Table 1: The sample structure Category (%) Gender M 35.7 F 64.3 Age < 26 years 0.9 26-35 years 19.8 36-45 years 41.6 46-55 years 24.8 > 56 years 12.9 Educational Background High School Graduate 19.8 Junior College Degree 63.4 University Degree 16.8 MSc/PhD 7 Experience in Sales < 5 years 26.8 6 - 10 27.5 11-15 12.7 > 15 33 Source: Authors, n=101 Table 2: Descriptive statistics ITEM ITEM -code Mean Median Standard Deviation Excess Kurtosis Skewness EMPOWERMENT The work I do is very important to me.  EMP1 6.030 6.000 1.222 5.429 -2.105 My job activities are personally meaningful to me.  EMP2 6.099 6.000 1.058 5.494 -2.086 The work I do is meaningful to me.  EMP3 5.812 6.000 1.175 3.522 -1.634 I am confident about my ability to do my job.  EMP4 6.287 6.000 0.825 15.741 -2.842 I am self-assured about my capabilities to per­form my work activities.  EMP5 6.307 6.000 0.817 16.192 -2.835 I have mastered the skills necessary for my job.  EMP6 6.109 6.000 0.866 10.937 -2.255 I have significant autonomy in determining how I do my job.  EMP7 5.822 6.000 1.214 4.227 -1.809 I can decide on my own how to go about doing my work. EMP8 5.337 6.000 1.307 1.333 -1.162 I have considerable opportunity for indepen­dence and freedom in how I do my job.  EMP9 5.149 5.000 1.360 1.068 -1.065 My impact on what happens in my department is large.  EMP10 4.832 5.000 1.548 -0.194 -0.753 I have a great deal of control over what happens in my department.  EMP11 5.317 6.000 1.495 1.275 -1.298 I have significant influence on what happens in my department.  EMP12 4.842 5.000 1.565 0.133 -0.881 SALES DEPARTMENT’S INNOVATIVENESS Our ability to function creatively is respected by the leadership.  SALINN­OV1 4.980 6.000 1.528 0.659 -1.165 Creativity is encouraged here.  SALINN­OV2 4.614 5.000 1.688 -0.408 -0.826 Around here, people are allowed to try to solve the same problems in different ways.  SALINN­OV3 4.762 5.000 1.517 0.217 -1.023 This organization can be described as flexible and continually adapting to change.  SALINN­OV4 4.723 5.000 1.672 -0.554 -0.712 This organization is open and responsive to change.  SALINN­OV5 4.733 5.000 1.515 -0.487 -0.576 The reward system here encourages innovation.  SALINN­OV6 4.119 4.000 1.831 -1.331 -0.227 NEW PRODUCT SELLING I like to present my customers with our most in­novative products.  NPS1 5.881 6.000 1.065 3.769 -1.506 I like selling products that need me to explain in great detail just what is new and exciting about them.  NPS2 5.752 6.000 1.121 1.877 -1.338 I like to visit new accounts where I have to pres­ent what my company is selling.  NPS3 5.624 6.000 1.319 1.369 -1.193 INTERNAL KNOWLEDGE SHARING There is close interaction and collaboration be­tween people in the two functions.  IKS1 5.129 5.000 1.355 0.870 -1.014 There is open communication between people in the two functions.  IKS2 5.129 6.000 1.433 0.855 -1.153 There is high level of knowledge sharing be­tween people in the two functions.  IKS3 4.822 5.000 1.485 -0.031 -0.829 People in the two functions have great dialogues with each other.  IKS4 4.941 5.000 1.326 0.912 -0.949 People in the two functions regularly communi­cate with each other.  IKS5 5.129 5.000 1.376 1.343 -1.161 People in the two functions provide each other with a lot of feedback.  IKS6 4.792 5.000 1.430 0.368 -0.801 There is a lot of two-way communication be­tween people in the two functions.  IKS7 4.891 5.000 1.421 0.543 -1.004 Table 2: Descriptive statistics (continues) Source: Authors’ own calculation Table 3: Internal consistency reliability and convergent validity Construct/Item Original Sample (O) Standard Deviation (STDEV) T Statistics (|O/STDEV|) Cronbach’s Alpha Composite Reliability (CR) Average Variance Extracted (AVE) Internal Knowledge Sharing IKS1 <- IKS 0.895* 0.024 37.033 0.95 0.96 0.80 IKS2 <- IKS 0.927* 0.016 58.983 IKS3 <- IKS 0.889* 0.032 28.031 IKS4 <- IKS 0.898* 0.029 30.504 IKS5 <- IKS 0.887* 0.031 28.771 IKS6 <- IKS 0.888* 0.030 29.941 IKS7 <- IKS 0.874* 0.038 23.089 New Product Selling NPS1 <- NPS 0.896* 0.056 15.977 0.87 0.92 0.79 NPS2 <- NPS 0.899* 0.061 14.675 NPS3 <- NPS 0.879* 0.069 12.643 Sales Department’s Innovativeness SALINNOV1 <- SALIN­NOV 0.894* 0.023 38.698 0.92 0.94 0.73 SALINNOV2 <- SALIN­NOV 0.902* 0.022 40.177 SALINNOV3 <- SALIN­NOV 0.822* 0.046 17.790 SALINNOV4 <- SALIN­NOV 0.883* 0.031 28.491 SALINNOV5 <- SALIN­NOV 0.884* 0.030 29.910 SALINNOV6 <- SALIN­NOV 0.736* 0.060 12.222 Empowerment EMP1 <- EMP 0.786* 0.062 12.681 0.87 0.89 0.53 EMP2 <- EMP 0.752* 0.086 8.785 EMP3 <- EMP 0.870* 0.047 18.597 EMP4 <- EMP 0.614* 0.194 3.170 EMP5 <- EMP 0.600* 0.200 3.001 EMP10 <- EMP 0.693* 0.104 6.634 EMP11 <- EMP 0.750* 0.079 9.526 EMP12 <- EMP 0.697* 0.109 6.411 Table 3: Internal consistency reliability and convergent validity (continues) * p<0.05 Source: Authors’ own calculation Table 4: Fornell-Larcker criterion and HTMT ratio* Fornell–Larcker criterion ** HTMT ratio (95% bias corrected interval) EMP IKS NPS SALINNOV EMP IKS NPS EMP 0,725 IKS 0.555 0.894 0.569 (0.376-0.737) NPS 0.514 0.343 0.891 0.579 (0.424-0.724) 0.364 (0.139-0.603) SALINNOV 0.648 0.633 0.396 0.856 0.695 (0.551-0.792) 0.664 (0.518-0.781) 0.430 (0.251-0.593) *EMP – Empowerment, IKS – Internal Knowledge Sharing, NPS – New Product Selling, SALINNOV – Sales Department’s Innovativeness ** Construct correlation matrix with the square root of the AVE on the diagonal in bold Source: Authors’ own calculation Source: Authors Figure 1: Structural model Table 5: Direct and indirect effects Hypothesis Original Sample (O) Standard Devi­ation (STDEV) T Statistics (|O/STDEV|) 95% bias cor­rected interval R2 f2 Hypothesis Acceptance Direct and Total Effects H1: Internal Knowl­edge Sharing ->New Product Selling 0.343* 0.110 3.106 0.124-0.539 0.12 0.13 Accepted H2: Internal Knowl­edge Sharing ->Sales Depart­ment’s Innovative­ness 0.633* 0.062 10.209 0.488-0.745 0.40 0.67 Accepted H3: Empowerment ->Internal Knowl­edge Sharing 0.555* 0.077 7.192 0.373-0.687 0.31 0.44 Accepted Indirect Effects H3a: Empowerment -> Internal Knowledge Sharing -> Sales Department’s Inno­vativeness 0.351* 0.069 5.095 0.208-0.475 Accepted H3b:Empowerment -> Internal Knowl­edge Sharing – New Product Selling 0.190* 0.083 2.279 0.049-0.356 Accepted * p < 0.05 Source: Authors’ own calculation Abbasi, S.G., Shabbir, M.S., Abbas, M, Tahir, M.S. (2020). HPWS and knowledge sharing behavior: The role of psychological empowerment and organization­al identification in public sector banks. Journal Public Affairs, Article e2512. https://doi.org/10.1002/pa.2512 Ahearne, M., Rapp, A., Hughes, D. E. & Jindal, R. (2010). Managing Sales Force Product Perceptions and Con­trol Systems in the Success of New Product Introduc­tions. Journal of Marketing Research, 47(4), 764-776. https://doi.org/10.1509%2Fjmkr.47.4.764 Ahmed, P. K. & Rafiq, M. (2006). Internal Marketing: Tools and Concepts for Customer-Focused Manage­ment. Oxford: Butterworth-Heineman Al-Omari,Z.,S., Alomari K., A.,A., & Aljawarneh N.,M., (2020). The role of empowerment in improving in­ternal process, customer satisfaction, learning and growth. Management Science Letters, 10, 841-848. http://doi.org/10.5267/j.msl.2019.10.013 Arfi W. B., Enstrom R., Sahut J.M., & Hikkerova L. (2019). The significance of knowledge sharing plat­forms for open innovation success, Journal of Organ­izational Change Management, 32(5), 496-516. http://doi.org/10.1108/JOCM-09-2018-0256 Bock G., Zmud R., Kim Y., & Lee J. (2005). Behavioral Intention Formation in Knowledge Sharing: Examin­ing the Roles of Extrinsic Motivators, Social-Psycho­logical Forces, and Organizational Climate. MIS Quar­terly, 29 (1), 87-111. http://doi.org/10.2307/25148669 Borghl M., & Schepers J., (2017), Are conservative ap­proaches to new product selling a blessing in disguise?. The Journal of the Academy Marketing Science, 46, 857-878. http://doi.org/10.1007/s11747-017-0521-1 Caloghirou Y., Kastelli I., & Tsakanikas A. (2004). Internal capabilities and external knowledge sources: comple­ments or substitutes for innovative performance? Tech­novation, 24, 29-39. http://doi.org/10.1016/S0166-4972(02)00051-2 Crawford, M., & DiBenedetto, A. (2011). New Product Management. (10th ed.). New York, NY: McGraw-Hill Irwin Chen H., Fan H. & Tsai C. (2014). The Role of Commu­nity Trust and Altruism in Knowledge Sharing: An In­vestigation of a Virtual Community of Teacher Profes­sionals. Journal of Educational Technology & Society, 17(3), 168-179. Conger J.A. & Kanungo R.N. (1988). The Empowerment Process: Integrating Theory and Practice. The Acade­my of Management Review, 13 (3), 471-482. https://doi.org/10.5465/amr.1988.4306983 Grošelj, M., Cerne M., Penger S. & Grah B. (2020). Au­thentic and transformational leadership and innovative work behaviour: the moderating role of psychological empowerment. European Journal of Innovation Man­agement, 1460 - 1060. https://doi.org/10.1108/EJIM-10-2019-0294 De Clercq D., Thongpapanl N. & Dimov D. (2013). Or­ganizational Social Capital, Formalization, and Inter­nal Knowledge Sharing in Entrepreneurial Orientation Formation, Entrepreneurship Theory and Practice, 37(3), 505-537. https://doi.org/10.1111/etap.12021 Fornell, C. & Larcker, D.F., (1981). Evaluating structur­al equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39-50. https://doi.org/10.2307/3151312 Gressgård L. J. (2015). Antecedents of Knowledge Ex­change Systems Usage: Motivational and Work En­vironment Factors. Knowledge and Process Manage­ment. 22(2), 112–125. https://doi.org/10.1002/kpm.1471 Hair, J.F., Ringle, M. C. & Sarstedt, M. (2012). Partial Least Squares: The better approach to structural equa­tion modeling? Long Range Planning, 45(5-6), 312-319. http://dx.doi.org/10.1016/j.lrp.2013.01.001 Hair, J.F., Hult, M. T., Ringle, M. C. & Sarstedt, M. (2014). A Primer on Partial Least Squares Structural Equation Modeling. CA: SAGE Publications Ltd. Hair, J.F., Sarstedt, M., Risher, J.J. & Ringle, M. C. (2019). When to use and how to report the results of PLS-SEM. European Business Review, 31 (1), 2-24 Hass M. R. & Morten T. Hansen (2007) Different Knowl­edge, Different Benefits: Toward a Productivity Per­spective on Knowledge Sharing in Organizations. Strategic Management Journal, 28 (11), 1133-1153. https://doi.org/10.1002/smj.631 Healy, M., Cleary, P. & Walsh, E. (2018). Innovativeness and accounting practices: an empirical investigation. Qualitative Research in Accounting & Management, 1176-6093 https://doi.org/10.1108/QRAM-06-2017-0047 Helmy I., Adawiyah W.,R. & Banani A. (2019). Linking Psychological Empowerment, Knowledge Sharing, and Employees. Innovative Behavior in Indonesian SMEs. The Journal of Behavioral Science. 14(2), 66-79. Retrieved from https://so06.tci-thaijo.org/index.php/IJBS/article/view/172180 Homburg C., Hohenberg S. & Hahn A. (2019). Steering the Sales Force for New Product Selling: Why Is It Dif­ferent, and How Can Firms Motivate Different Sales Reps?, Journal of Product Innovation management, 36(3), 282-304. https://doi.org/10.1111/jpim.12476 Huang Ch-M., Chang J. & Henderson S. (2008) Knowl­edge Transfer Barriers Between Research and De­velopment and Marketing Groups Within Taiwanese Small- and Medium-Sized Enterprise High-Technolo­gy New Product Development Teams. Human Factors and Ergonomics in Manufacturing, 18 (6), 621–657. https://doi.org/10.1002/hfm.20130 Hume C. & Hume M. (2015).The Critical Role of Inter­nal Marketing in Knowledge Management in Not-for-Profit Organizations. Journal of Nonprofit & Public Sector Marketing, 27, 23–47. https://doi.org/10.1080/10495142.2014.934567 Kang Y.,J., Lee J., Y. & Kim H., W. (2017). A psycho­logical empowerment approach to online knowledge sharing. Computers in Human Behavior 74, 175 – 18. http://dx.doi.org/10.1016/j.chb.2017.04.039 Kim S. & Lee H. (2006). The Impact of Organizational Context and Information Technology on Employee Knowledge-Sharing Capabilities. Public Administra­tion Review, 66(3), 370-385.  https://doi.org/10.1111/j.1540-6210.2006.00595.x Matsuo M., (2018). Sales management: Learning and innovationin Japan, Journal of Marketing Chan­nels, 25(4), 241-244, https://doi.org/10.1080/1046669X.2019.1658014 Matsuo, M. (2005). The Influence of Sales Management Control on Innovativeness of Sales Departments. The Journal of Personal Selling and Sales Management, 29(4), 321-333. https://doi.org/10.2753/PSS0885-3134290402 Mu J., Thomas E., Peng G. & Di Benedetto A. (2017). Strategic orientation and new product development performance:The role of networking capability and networking ability, Industrial Marketing Management, 64(July), 187-201. http://dx.doi.org/10.1016/j.indmar­ man.2016.09.007 Mulyana M., Hendar H.,Zulfa M. & Ratnawati A. (2020). Marketing Innovativeness on Marketing Performance: Role of Religio-Centric Relational Marketing Strate­gy. Journal of Relationship Marketing, 19(1), 52-74. https://doi.org/10.1080/15332667.2019.1664869 Noorderhaven N. & Harzing A. (2009). Knowledge-Shar­ing and Social Interaction within MNEs. Journal of International Business Studies, 40(5), 719-741. Re­trieved from http://www.jstor.org/stable/40262806 Podsakoff P.M., MacKenzie S.B, Lee J-Y. & Podsakoff N. (2003). Common Method Biases in Behavioral Research: A Critical Review of the Literature and Recommended Remedies. Journal of Applied Psy­chology, 88(5), 879–903 https://doi.org/10.1037/0021-9010.88.5.87934 Prokop V. & Stejskal J. (2017). The Influence of Infor­mation Sharing and Collaboration on the Extent of Innovation activities. International Conference on Knowledge Engineering and Applications. https://doi.org/10.1109/ICKEA.2017.8169907 Petrariu I., R., Bumbac R. & Ciobanu R. (2013). Innova­tion: a path to competitiveness and economic growth.The case of CEE countries. Theoretical and Applied Economics. 5(582), 15-26. Ringle, C. M., Wende, S. & Becker, J.-M. (2015). “Smart­PLS 3.” Boenningstedt: SmartPLS GmbH, http://www.smartpls.com. Rossiter, 2002. Sharma A. & Sagar, M. (2018). New product selling chal­lenges (key insights in the ICT sector). Journal of In­dian Business Research, 10(3) 291-319. https://doi.org/10.1108/JIBR-11-2017-0216 Spreitzer M., G. (1995). Psychological Empowerment In The workplace: Dimensions, Measurement And Vali­dation. The Academy of Management Journal, 38(5), 42-1465. https://doi.org/10.2307/256865 Stojcic N. (2020). Collaborative innovation in emerging innovation systems: Evidence from Central and East­ern Europe. The Journal of Technology Transfer, 46, 531–562. https://doi.org/10.1007/s10961-020-09792-8 Stojcic N., Hashi I. & Aralica Z. (2018). Creativity, inno­vations and firm performance in an emerging transition economy. Ekonomski Pregled, 69(3), 203-228 Švarc J. & Dabic M. (2019). The Croatian path from so­cialism to European membership through the lens of technology transfer policies., The Journal of Technol­ ogy Transfer, 44, 1476–1504. https://doi.org/10.1007/s10961-019-09732-1 Song M. & Parry M. (1997). The Determinants of Japanese New Product Successes. Journal of Marketing Re­search, 34 (1), 64-76. https://doi.org/10.2307/3152065 Tang Y.E & Marinova D. (2020). When less is more: the downside of customer knowledge sharing in new product development teams, Journal of the Acade­my of Marketing Science, 48, 288–307. https://doi.org/10.1007/s11747-019-00646-w Tsai W. (2001). Knowledge Transfer in Intraorganization­al Networks: Effects of Network Position and Absorp­tive Capacity on Business Unit Innovation and Perfor­mance, The Academy of Management Journal, 44(5), 996-1004. https://doi.org/10.2307/3069443 Van den Berg, W. E., Verbeke, W., Bagozzi, R. P., Worm, L., de Jong, A. & Nijssen, E. (2014). Salespersons as Internal Knowledge Brokers and New Products Sell­ing: Discovering the Link to Genetic Makeup. The Journal of Product Innovation Management, 31(4), 695-709. https://doi.org/10.1111/jpim.12156 Verbeke W.J.M.I & Masih J. (2020). Evolutionary-shaped goal orientation in Homosapiens: how life sciences contribute to a better understanding of salespeople as knowledge brokers, Journal of Personal Selling& Sales Management, 40(1), 43-56, https://doi.org/10.1080/08853134.2019.1654389 Wang S., Noe R. A. & Wang Z. (2014). Motivat­ing Knowledge Sharing in Knowledge Manage­ment Systems: A Quasi–Field Experiment. Jour­nal of Management, 40(4), 978-1009. https://doi.org/10.1177/0149206311412192 Wang Z. & Wang N. (2012). Knowledge sharing, innova­tion and firm performance. Expert Systems with Appli­cations, 39(10), 8908 -8899. https://doi.org/10.1016/j.eswa.2012.02.017 Wieseke J., Homburg C. & Lee N. (2008). Understand­ing the adoption of new brands through sales peo­ple: a multilevel framework. Journal of the Academy of Marketing Science, 36(2), 278 - 291. https://doi.org/10.1007/s11747-007-0055-z Yiu H. L., Ngai E.W.T. & Lei C. F. (2020). Impact of Service-Dominant Orientation on the Innovation Per­formance of Technology Firms: Roles of Knowledge Sharing and Relationship Learning. A Journal of De­cision Sciences Institute, 51(3), 620 - 653. https://doi.org/10.1111/deci.12408 Zhou K.Z. & Li C.B. (2009). How knowledge affects radical innovation: Knowledge base, market knowl­edge acquisition, and internal knowledge. Strategic Management Journal, 33(3), 1090–1102. https://doi.org/10.1002/smj.1959 Erik Ružic, PhD, is an Associate Professor at the Faculty of Economics and Tourism “Dr. Mijo Mirkovic” in Pula, Croatia. His main fields of interest are sales management, personal selling and internal marketing. He used to work as a sales manager and has professional experience in different managerial positions. Dragan Benazic, PhD, is an Associate Professor at the the Faculty of Economics and Tourism “Dr. Mijo Mirkovic” in Pula, Croatia. He is author and co-author of numerous scientific articles. His research is focused on the fields of marketing research, price management and services marketing. Vpliv interne izmenjave znanja na inovativnost oddelka prodaje in komercializacijo izdelkov Ozadje in namen: Inovativnost in trženje novih izdelkov sta za podjetja zelo pomembni. Zato je dobro razumevanje vpliva vseh potencialno vplivnih dejavnikov uspeha kljucnega pomena. Namen prispevka je raziskati vpliv notranje izmenjave znanja na prodajo novih izdelkov in prodajno inovativnost ter vpliv opolnomocenja na notranjo izmenjavo znanja in posredno na prodajo novih izdelkov in prodajno inovativnost. Oblikovanje / metodologija / pristop: Raziskava je zajela 101 prodajalca, ki so delali pri enem od 1000 podjetij z najboljšo dodano vrednostjo na Hrvaškem. Vprašalnik je bil razvit in prilagojen s štirimi lestvicami za oceno notranje izmenjave znanja, prodaje novih izdelkov, prodajne inovativnosti in opolnomocenja. Podatki so bili analizirani z upo­rabo metode PLS-SEM za preucevanje povezav med konstrukti. Rezultati: Notranja izmenjava znanja pozitivno vpliva na inovativnost oddelka za prodajo in prodajo novih izdelkov, opolnomocenje pa je pozitivno povezano z notranjo izmenjavo znanja in posredno z inovativnostjo oddelka za pro­dajo in prodajo novih izdelkov. Zakljucek: Menedžerji bi morali podpirati razlicne dejavnosti, da bi okrepili opolnomocenje in notranjo izmenjavo znanja, da bi vplivali na uspešnost podjetij pri komercializaciji novega izdelka in inovativnosti oddelka prodaje. Pri­hodnje raziskave bi lahko vkljucevale moderatorske spremenljivke med konstruktom opolnomocenja in konstruktom notranje izmenjave znanja ter poglobile vpogled v vrsto izmenjanih informacij, dinamiko izmenjave in ovire v procesu ter druge dejavnike, ki pozitivno vplivajo na izmenjavo znanja. Kljucne besede: Prodaja novih izdelkov, Inovativnost prodajnega oddelka, Notranja izmenjava znanja, Opolnomo­cenje, Prodajalci DOI: 10.2478/orga-2021-0011 Optimisation of the rational proportion of intelligent technologies application in service organisations Anton IVASCHENKO1, Alfiya R. DIYAZITDINOVA2, Tatiyana NIKIFOROVA1 1 Samara State Technical University, Samara, Russia, anton.ivashenko@gmail.com, kolesnikova.t.v.163@gmail.com 2 Povolzhskiy State University of Telecommunications and Informatics, Samara, Russia, dijazitdinova@mail.ru Background and Purpose: The growing role and involvement of Artificial Intelligence in modern digital enterprises leads to a considerable reduction of personnel and reorientation of the remaining staff to new responsibilities. How­ever, in many areas like services and support the total elimination of the employed human resources still remains impossible. It is proposed to study the organisational problem of finding the optimal proportion of computer agents and human actors in the mixed collaborative environment. Methods: Using the technology of semantic and statistical analysis, we developed an original model of computer agents’ and human actors’ cooperative interaction and an optimization method, which is novel in considering the focus of the executors while calculating the compliance indicators. Results: The problem was studied by an example of service desk automation. Considering the semantics of the problem domain in the form of ontology introduces the logic for better distribution and automation of tasks. Conclusion: In a modern digital enterprise there exists and can be estimated a rational balance between the computer agents and human actors, which becomes a significant indicator of its performance. In general, human actors are preferable for processing unpredictable events in real time, while agents are better at the modelling and simulation. Keywords: Service organisations automation, Artificial intelligence, Human resources, Decision-making support. 1 Introduction Current trends in the automation of modern organisa­tional management require the widespread introduction of artificial intelligence. According to the strategy of total digitalisation, human resources should be replaced by ro­bots capable of performing business processes with higher efficiency and lower costs. In this context, the robots are not only machines but also computer agents with auton­omous behaviour that interact in the virtual world and provide self-organisation. Intelligent solutions of this kind stimulate high expectations. At the same time, their implementation in practice re­mains challenging. In most business cases it is impossible to completely replace humans by programmes. Sometimes it is hard to formalise all the options of the business pro­cess and maintain possible changes; more often, the per­sonnel remain, as they can provide greater adaptability. Therefore, in most real applications both humans and al­gorithms are involved. Human and robotic computer-simulated executors need to interact and cooperate in real time under the con­tinuous pressure of incoming data and on-going events, and require sophisticated coordination. People require ad­ditional training for acting in vastly automated spaces; al­gorithms need extra configuration to consider the possible influence of the human factor. In early applications, e.g. production line automation, attempts were made to bound and minimise interaction by physical borders. In modern organisations this is impracticable. Thereby a new problem appeared of looking for the best balance of artificial and human intelligence, consider­ing the specifics of certain business cases. It seems reason­able to completely replace the human executors in produc­tion and provide their maximum utilisation in services, but it is not so simple. Hence, when larger processes involve automation, the remaining personnel start looking for new functions and change their professional skillset, adapting it to the changing world. This process can either enhance the business or damage it. In this paper the problem is studied and solved for a common business case of service desk automation, which has often raised the question: “What is the best proportion of human and robot operators in an organisation?”. This paper originally brings up the question of the ne­cessity and sufficiency of the application of artificial intel­ligence to practice. The most relevant papers in this area are devoted to the problems of efficient substitution of hu­man resources with robots and improving the computer-human interaction. The authors propose for consideration an idea of the reasonable involvement of artificial intelli­gence based on the analysis of business processes specif­ics. Compared to the previous publications, new results of AI and HR balancing studied by an example of the service desk automation are presented. 2 State of the art Artificial intelligence is currently becoming one of the most promising areas for discussion and research in the modern world of digitalisation (Patel et al., 2000). By im­plementing these concepts, traditional enterprises are turn­ing into digitally minded organisations along the path of innovative transformation. A digital enterprise is an organ­isation that uses information technology as a competitive advantage in all areas of its business: manufacturing, busi­ness processes, marketing, and customer interaction. The product itself, offered by such an enterprise to the market, also becomes digital. Recent developments in the field of artificial intelli­gence (AI) and augmented reality provide great oppor­tunities in various areas of modern economics and busi­ness (One Internet, 2016). The most significant results are being achieved in the field of data visualisation and decision-making support, aimed at developing existing applications with new capabilities for data collection, pro­cessing and management. Cognitive computing (Kelly, 2015), the foundation of artificial intelligence, implies the simulation of human brain activity, which allows the ma­chines to process information, learn about the world and analyse events like people do, perhaps even more produc­tively. New trends in project management (Grzeszczyk, 2018) consider the influence of intelligent information systems, AI in decision-making support and new information and communication technologies. Mixed intelligence systems are applied for project evaluation, taking into account the current state of knowledge in the field of project manage­ment. The possible ways of considering the opportunities of artificial intelligence for pragmatic applications are pre­sented in (Bentley et al., 2018). Existing models of intelligent technologies imple­mentation in deep cooperation with human employees introduce the concept of the computer agent – an avatar with a pre-programmed profile, actions and communica­tion (Rosen, 2018). In the context of this paper, the term software agent can also be used as a synonym. Comput­er agents are involved in collaborative problem solving by sharing knowledge and understanding, organising the group work and monitoring the progress, taking actions to solve the problem, and providing constructive feedback to group members. The introduction of AI capabilities in business leads to a limitation in the same type of operations that are amenable to simple algorithms. This results in a reduction of the number of employees engaged in non-creative and formalised work. Despite the success of such solutions in corporate applications, the problem of the interaction of people and robots in the common information space of an enterprise is still challenging. The interaction of AI and HR intelligence is identified as mixed intelligence, and the environment in which they interact is combined. The results of a survey by (Grace et al., 2018) predict that AI will outperform humans in many activities that require both technical skills and creativity. A solution to bridge the gap between AI and human-robot interaction is proposed in (Khandelwal et al., 2017). The related ethical problems are overviewed in (Hagendorff, 2020). The existing approaches to the application of intelli­gent technologies to improve the staffing of modern organ­isations typically suggest applying AI in the area of HR management. Certainly, introducing intelligent systems for decision-making support in this area can significant­ly improve the efficiency and quality of projects staffing and management, recruitment and people development (Ivaschenko et al., 2019; Simonova et al., 2016). For example, the possible applications of Big Data technologies in enterprise HR management are widely discussed in recent literature (Zang et al., 2015; Barman et al., 2015). The challenge is in the use of multiple data sources to improve practices in training, recruitment, per­formance analysis and motivation. Big Data in HR sets out to evaluate and improve practices including talent acqui­sition, development, retention, and overall organisational performance. An interesting approach to the application of the AI concept in human communities is introduced in (Letouze et al., 2018). It is proposed to design future human sys­tems based on the principles of interaction and Big Data technologies. The strengths and weaknesses of human and machine collaboration are reviewed in (Wilson et al., 2018). This paper is close to the topic studied here. In particular, it states that there is no reason to fear that robots will re­place all human employees. This research involved 1500 firms in a range of industries and shows that the biggest performance improvements come when humans and smart machines work together, enhancing each other’s strengths. When people and AI agents cooperate, the business pro­cesses need to be redesigned. As a result, in addition to the implementation of innovative AI technologies, provision should always be made for the development of employ­ees in using the computer-human interfaces and changing their mindset. This new sphere is called a collaborative intelligence (Kragic et al., 2018; Baratta 2015). Despite the recom­mendation to build a cooperative IT solution based on the organisation of efficient interaction of robots and humans, no technical methodology is proposed on how to calculate the optimal rate and transfer it to the design of the business process. Collaborative robots are deeply studied in Industry 4.0 (Djuric et al., 2016; Teixeira et al., 2018). The active appli­cation of automation and robotics requires a realistic sense of the best time and place for their use. In the framework of this theory, collaborative robots specifically designed for direct interaction with humans are introduced; a collabo­rative workspace where robots and humans can perform tasks simultaneously and safely; and collaborative oper­ations in which purpose-designed robots can work safely together with humans. Yet the problems being solved have still always been limited to providing comfortable physical interaction, leaving the challenges of functional co-work­ing unaddressed. The rapid increase in the share of artificial intelligence in human-computer interaction complicates the search for a solution to the optimal interaction of AI with the em­ployees. A managerial decision should solve two prob­lems: what is the best ratio of AI and people in a mixed interaction, and what are the optimum areas of application of these solutions, taking into account the human factor. The solution of the AI and HR proportion and involvement in the business processes of a digital company needs to be built on the principles of management by conditions (Ivaschenko et al., 2015, 2019, 2020). As a result, the user is put inside the “loop” of com­puter-human interaction (Holzinger, 2014, 2016), which requires the user interfaces context and focus to be seen as dependent. The system should interlink all related data sets (e.g., images, text, measured values, scans) and offer visual scenes. It is proposed to involve the decision mak­er in the process of data processing and visualisation by means of continuously interacting with the system, which helps to optimise the learning behaviour of both humans and algorithms. The introduction of AI requires additional efforts to provide modified and improved business processes (Simek et al., 2019; Halaska et al., 2018). An orientation to the specifics of the executors is provided by the sub­ject-oriented approach for business processes management (S-BPM), which conceives a process as a collaboration of multiple subjects organised via structured communication (Fleischmann et al., 2013, 2015). These models enable the synchronisation of computer agents and human actors. But they still remain oriented to primarily human deci­sion-making, which becomes insufficient in the case of the predominance of robots over the human operators. In addition, the principle of decentralised decisions means the ability of a cyber-physical system to make de­cisions on its own and to perform tasks as autonomously as possible. The simulation and modelling of models of multi-agent technology (Wooldridge, 2002; Gorodetskii, 2012) allow the implementation of such autonomous be­haviour by software and thus provide the effect of self-or­ganisation. To coordinate computer agents and human actors on a semantic level, the mechanism of knowledge descrip­tion in the form of ontologies can be used (Pouchard et al., 2000; De Nicola et al., 2009). Ontology represents the semantic network, combining concepts, attributes, rela­tions and rules that describe the situation from a predefined point of view. The functioning of the organisation staff is highly influenced by the human factor, that is to say that people perceive different situations differently, although, for a certain problem domain, the employees tend to agree on a similar perception and understanding. Due to this fact, problem domain ontologies are treated as impersonal and objective. The ontologies of computer agents can be formally predefined and used to specify the definite and explicit logic of their behaviour. One of the potential problem domains that require optimisation of the proportion of intelligent technologies applied is service desk automation. Its process involves online collaboration and the provision of IT services to customers. Ensuring the constant high quality of IT servic­es and customer or user satisfaction is implemented based on the approach (Axelos, 2019; England, 2008) described by the IT Infrastructure Library (ITIL). It focuses on the achievement of goals, the analysis of key performance indicators, as well as the resources spent on achieving goals. ITIL includes a set of documents used for the prac­tical implementation of IT Service Management (ITSM) approaches, which will include the basic processes that support and provide IT services. Service management is implemented by IT providers by using the optimal com­bination of people, processes and information technology. In the structure of ITIL and ITSM processes, the ser­vice desk plays an important role, carrying out accounting, registration of tasks, solution and tracking of issues. Next, we will consider the flow of tasks received from users and their processing in the service desk system. Tasks arrive on the service desk at random times and form a task queue. Then, the tasks go to a group of specialists who take them out of the line independently, so we cannot uniquely iden­tify a specific executor. All personnel from the working group are interchangeable. 3 The formal model 1Received: 7th July 2020; revised: 5th March 2021; accepted: 26th March 2021 The following formal model is based on an ontological approach that allows the definition of the semantics of any object by a knowledge descriptor: (1) where is a tag (keyword), its weight, defines the relation to another descriptor , – the time of the descriptor’s applicability. Concepts relate to the objects, weighted tags corre­spond to the attributes, and their dependencies are de­scribed by relationships. The basic rule required in this context determines the criterion of proximity. Objects and are ontologically close when (2) where is an allowable level of semantic deviation, Let us study the combined environment of a digital enterprise that contains both human actors and computer agents . rep­resent the number of actors and agents respectively. Both actors and agents have corresponding knowledge descrip­tors: and . A typical service desk implements standardised busi­ness processes according to which actors and agents need to perform certain tasks or actions at specified mo­ments. Therefore the work process can be formulated by a scenario: , (3) where – is the requirement for the executor, – time for the task to start, – task appearance time, , – its normative duration. For similar operations, specified template scenarios for the generation of new states for every new order can be used. To start a task an appropriate executor should be avail­able and be assigned to it. An assignment event for the ac­tor is specified as , , (4) where the sequence of descriptors describes how the actor’s focus changes with time. An assignment event for the agent is specified cor­respondingly: , (5) The process of task flow distribution between the agents and actors in real time is characterised by the fol­lowing average indicators: • processing time; • downtime or waiting time (idle time when waiting for assignment and start); • the number of failures. In problem domains like service desk automation the process of tasks execution involves multiple negotiations with the consumer and is performed in real time. Therefore the difference between the processing time by computer agents or human actors does not differ. When processing the task flow, the system goal is to minimise the number of execution failures and the task processing timeout. Using the introduced indicators, we reduce the process of determining the optimal percentage ratio of agents and actors to solving the optimisation problem of discrete pro­gramming that includes the following objective functions. Minimum waiting time (downtime): , (6) Minimum number of failures: (7) The proposed model has the following main assump­tions and limitations. 1. The application of intelligent technologies in the modern organisation is concerned with the reduction of personnel and reorientation of the remaining staff to new responsibilities. Therefore, during the time of AI imple­mentation, agents and actors are considered interchange­able. 2. The quality of tasks being performed by human ac­tors is not compared to the quality of computer agents. If the task is carried out by any executor on time and with acceptable quality it is treated as successfully done. Other­wise it is treated as “rejected”. 3. The difference between the normative duration of task execution by a human employee or robot, as well as the difference in cost, is not taken into account here. These parameters are influenced by the number of frequently re­peated typical tasks. Computer agents are preferable for typical tasks, and human executors react better to unique occasional events. Therefore the problem of costs and time optimisation remains outside the scope of this study. 4. We consider that the efficiency characteristics of the computer agents and human actors are constant and do not change over time, and the number of tasks that need to be processed also remains unchanged. 4 Method The solution of the problem (6, 7) is formulated as a set of assignment events . In order to process the incoming event flow of tasks , the system should pro­vide available resources of a certain type (agents or actors) at the required moments of time. The main problem is that we can neither attract additional employees when needed nor keep additional staff in reserve. Therefore the proposed method contains three main stages: agents/actors ratio optimisation, compliance analy­sis and assignment. The generalised schema of the method is presented in Fig. 1. This method can be implemented using either standard software for simulation and analysis or a specifically designed software solution. The first stage is to determine the quantitative ratio of agents and actors for efficient processing of the flow of incoming tasks. In the case of low knowledge about the real amount and types of tasks, event flows are simulated. Figure 1: The method of optimisation of the rational proportion of intelligent technologies application The first stage is to determine the quantitative ratio of agents and actors for efficient processing of the flow of incoming tasks. In the case of low knowledge about the real amount and types of tasks, event flows are simulated. As a starting point, we consider an option with the number of actors = 1, and the proportion = 100%, which corresponds to the complete execution of tasks by people, without the use of automated robotic systems. Next, we consider the option in which the number of agents is = 1. It corresponds to fully automatic pro­cessing of incoming tasks. With this distribution, the wait­ing time for tasks in the queue is reduced, but the number of failures has the potential to increase, since the agents are able to perform poorly formalised tasks. To determine the point at which both objective func­tions are minimal, we use the gradient descent method with step splitting. The geometric interpretation of the method is that at each step we move along the anti-gradi­ent vector, reduced by . times, determining the direction of optimisation. Based on this data, a balance graph is con­structed, which displays the intersection point of two func­tions, which corresponds to the optimal percentage ratio of agents and actors. Compliance analysis is performed to improve the qual­ity of the matching of the incoming tasks to the most ap­propriate executors. In order to provide on-time processing of incoming tasks, agents and actors need to be assigned correspondingly. Considering their autonomous behaviour and self-organisation, this should be done indirectly by at­tracting their attention at required moments. The result is described by the following compliance indicator: (8) where – focus time (time of actor changing his/her interest represented by semantic descriptor); – assignment time, . Compliance indicator (5) allows the main correlations between the event flows of tasks and assignment to agents or actors to be determined. Compared to the high dynamics of being forced by changes of the actors’ interest, the focus of each agent does not change with time. The compliance indicator for agents represents their impact: (9) Considering the limited number of tasks, statements (8) and (9) are in contradiction. To maximise the number of agents, the statement (8) should be targeted first, then the number of skipped tasks is reduced and assigned to corresponding human actors by attracting their interest. Such an approach provides an efficient proportion of computer agents and human actors in the mixed intelli­gence. In general, human actors turn out to be preferable for processing unpredictable events in real time, while AI agents are better at simulating multiple versions of possi­ble situations to predict, analyse, compare and optimise. Assignment is the last stage of the method that gener­ates the planned flows of . The tasks are not assigned declaratively, but proactively taken out from the common queue by agents and actors according to their preferences. Actors pick out the tasks for execution them­selves, while the agents are assigned automatically. Based on the execution history, the knowledge base is updated in automatic mode. The staff knowledge base is used to simulate the agents of new types and data process­ing at the optimisation stage. Considering the features of agents and actors, as well as the difference in the criteria for interaction and task pro­cessing, the staff knowledge base is combined from the on­tologies of the task (see Fig. 2), the actor (see Fig. 3) and the agent (see Fig. 4). They present the main concepts in the form of a semantic model of RDF data and are visual­ised by OntoGraf – a conceptual map, which can be equat­ed to a semantic network, since it has a network structure, the ability to specify the type of connection, uses words at the vertices, and has multiple connections between the concepts. The model visualisation is presented using Pro­tégé: a local, freely distributed Java program designed to build (create, edit and view) ontologies of the application area. The developed ontologies are necessary to introduce the hierarchical structure of the criteria for fulfilling tasks, which makes it possible to determine the relationships, dif­ferences and similarities between the ontologies of tasks and their executors. Figure 2: Task ontoogy Figure 3: Actor ontology Figure 4: Agent ontology 5 Service desk case studies The problem of AI and HR balancing was studied by an example of service desk automation. We have taken a dataset of 648 real calls that require service tasks gener­ation and processing during a fixed period of time. The data was taken from the real service desk centre of an IT company and contains support requests. The call flow was divided into three types: Information request, Providing access, and Closing access. According to the model and problem statement intro­duced above, the system goal is to minimise the number of execution failures and the task processing timeout. For the initial calculations, the averaged parameters of the speed of processing calls, the number and time of downtime for agents and actors were determined. These parameters were analysed using the statistical analysis software package STATISTICA by StatSoft. More details about the analysis are presented in the appendix. According to the method introduced, for = 1 the failure rate was zero, but the downtime increased by 1.9 times compared to the initial values obtained. For = 0 the waiting time for tasks in the queue was reduced by 1.4 times, but the number of failures reached 90 tasks due to renouncements and showed a tendency to increase, since the agents were able to perform poorly formalised tasks. An option was also considered in which the number of agents and actors was equally divided. Based on the cal­culations made, a graph was built (see Fig. 5). The graph shows how the performance indicator changes depending on the percentage of agents and actors for the objective functions introduced. Based on the data obtained by the gradient descent method for two objective functions, a balance graph was constructed (see Fig. 6), which displays the intersection point of two functions, which corresponds to the optimal percentage ratio of agents and actors. 6 Conclusion Analysis and testing of AI and HR balancing in theory and practice gives a new idea for organisational manage­ment to look for the optimal percentage ratio of comput­er agents and human actors. Herewith, while solving this problem, the main objective should be changed from the total or maximum possible replacement of executors by software and robots to a rational combination of artificial and human executors, targeting their maximum coopera­tion. Therefore in a modern digital enterprise there exists and can be estimated a rational proportion of computer agents and human actors, which becomes a significant in­dicator of their performance. This parameter varies with time depending on the business requirements. In general, human actors are preferable for processing unpredictable events in real time, while agents are better at the modelling and simulation of multiple possible situations for forecast­ing, analysis, comparison, and optimisation. The results of the performed analysis of service desk automation applying a rational proportion of intelligent technologies helped to formulate the main differences and features of the subjects of mixed intelligence, presented in Table 1. This conclusion may be of interest to the develop­ers of modern AI solutions based on multi-agent technolo­gies and network organisational structures. The actors have the prospect of a deeper and more cre­ative execution of tasks. At the same time, in mixed teams, dealing with a rapid change in context, the advantage will be on the side of the agents. Considering the theory of or­ganisational management and the influence of the human factor, the approaches to stimulating agents and actors should be different. The actors need to understand the ben­efits of their involvement and the impossibility of their to­tal replacement by agents. The agents should provide sta­ble and reliable services and be able to cooperate with the actors to develop new business opportunities. To reduce the amount of waiting time specifically for the actors, it is necessary to attract their attention to urgent tasks in the necessary steps of the business processes. As presented in this paper, the complete replacement of personnel leads to a lack of adaptability, while the in­troduction of additional artificial intelligence gives new opportunities and advantages. Besides, the employees of modern companies need to decide nowadays what area of their impact will be the most beneficial and adopt training paths and development strategies accordingly. Literature Figure 5: Dependency of tasks processing metrics for different agents/actors ratio Figure 6: Dependency of failures on waiting time Table 1: Features of mixed intelligence subjects Criterion Computer agent Human actor Time Model time Real time Reality Able to simulate multiple versions of scene Capable of acting in real scene or a few forecast scenes Focus Rigid and predefined Flexible, changing under the influence of interest Role Reliable and predictable Highly qualified and motivated Context influence No Considerable Problem statement Formalised Lacking Limitations Unable to solve unique and poorly for­malised tasks Low productivity Strengths High performance High adaptability Switching between tasks Needs study time Low study time Axelos (2019). ITIL Foundation, ITIL 4th ed. UK: TSO (The Stationery Office), 222. Baratta, D. (2015). Industrial collaborative robot design. A guideline for future design activity. CEUR Workshop Proceedings, Vol 1473. Artificial Intelligence and De­sign AIDE 2015. http://ceur-ws.org/Vol-1473, 7 Barman, A., Ahmed, H. (2015). Big Data in human resou­rce management – developing research context. http://doi.org/10.13140/RG.2.1.3113.6166 Bentley, P.J., Brundage, M., Haggstrom, O., Metzinger, T., Gutenberg, J. (2018). Should we fear artificial intelli­gence? STOA - Science and Technology Options Asse­ssment, 40. http://dx.doi.org/10.2861/412165 De Nicola, A., Missikoff, M., Navigli, R. (2009). A soft­ware engineering approach to Ontology building. In­formation Systems, 34(2), 258-275. http://dx.doi.or­ g/10.1016/j.is.2008.07.002 Djuric, A., Urbanic, R.J., Rickli, J. (2016) A framework for collaborative robot (CoBot) integration in advan­ced manufacturing systems. SAE International Jour­nal of Materials and Manufacturing, 9. http://doi.org/10.4271/2016-01-0337 England, R. (2008). Introduction to real ITSM. CreateSpa­ce Independent Publishing Platform. Fleischmann, A., Kannengiesser, U., Schmidt, W., Stary, C. (2013). Subject-oriented modeling and execution of multi-agent business processes. Web Intelligence (WI) and Intelligent Agent Technologies (IAT), 2013 IEEE/WIC/ACM International Joint Conferences, 138-145. https://doi.org/10.1109/WI-IAT.2013.102 Fleischmann, A., Schmidt, W., Stary, C. (2015). S-BPM in the wild. Springer. http://doi.org/10.1007/978-3-319-17542-3 Gorodetskii, V.I. (2012). Self-organization and mul­tiagent systems: I. Models of multiagent self-orga­nization. Journal of Computer and Systems Sciences International, 51(2), 256-281. https://doi.org/10.1134/S106423071201008X Grace, K., Salvatier, J., Dafoe, A., Zhang, B., Evans, O. (2018). When will AI exceed human performan­ce? Evidence from AI experts, https://arxiv.org/pdf/1705.08807.pdf, 21 Grzeszczyk, T.A. (2018). Mixed intelligent systems. Deve­loping models for project management and evaluati­on. Palgrave Pivot https://doi.org/10.1007/978-3-319-91158-8 Hagendorff, T. (2020). The ethics of AI ethics: an evalua­tion of guidelines. Minds & Machines, 30, 99-120. htt­ps://doi.org/10.1007/s11023-020-09517-8 Halaska, M., Sperka, R. (2018). Is there a need for agent­-based modelling and simulation in business process management? Organizacija, 51(4), 255-270. https://doi.org/10.2478/orga-2018-0019 Holzinger, A. (2014). Extravaganza tutorial on hot ideas for interactive knowledge discovery and data mining in biomedical informatics. Lecture Notes in Artificial In­telligence, 8609, 502–515 https://doi.org/10.1007/978-3-319-09891-3_46 Holzinger, A. (2016). Interactive machine learning for he­alth informatics: when do we need the human-in-the­-loop? Brain Informatics, 3(2) 119–131. https://doi.org/10.1007/s40708-016-0042-6 Ivaschenko, A., Khorina, A., Sitnikov, P. (2019). On­line creativity modeling and analysis based on Big Data of social networks. Advances in Intelligent Systems and Computing, 858, 329–337. https://doi.org/10.1007/978-3-030-01174-1_25 Ivaschenko, A., Korchivoy, S., Spodobaev, M. (2020). Infrastructural models of intermediary service provi­ders in digital economy. Advances in Intelligent Sys­tems and Computing, 1038, 594 – 605. https://doi.org/10.1007/978-3-030-29513-4_44 Ivaschenko, A., Milutkin, M. (2019). HR decision-making support based on natural language processing. Com­munications in Computer and Information Science, 1083, 152–161. https://doi.org/10.1007/978-3-030-29743-5_12 Ivaschenko, A.V., Barabanov, V.F., Podvalny, E.S. (2015). Conditional management technology for multiagent interaction. Automation and Remote Con­trol, 76(6), 1081– 1087. https://doi.org/10.1134/S0005117915060119 Ivaschenko, A.V., Simonova, M.V., Sitnikov, P.V., Shorni­kova, O.V. (2019). Big Data analysis for HR manage­ment at production enterprises. The European Procee­dings of Social & Behavioural Sciences EpSBS. Vol. LVII - GCPMED 2018. Future Academy. 463–471. https://dx.doi.org/10.15405/epsbs.2019.03.46 Khandelwal, P., Zhang, S., Sinapov, J., Leonetti, M., Tho­mason, J., Yang, F., Gori, I., Svetlik, M., Khante, P., Li­fschitz, V., Aggarwal, J., Mooney, R., Stone, P. (2017). BWIBots: A platform for bridging the gap between AI and human–robot interaction research. The Internati­onal Journal of Robotics Research, 36, 1–25. https://dx.doi.org/10.1177/0278364916688949 Kelly III, J. (2015). Computing, cognition and the future of knowing. IBM Research: Cognitive Computing. IBM Corporation. Kragic, D., Gustafson, J., Karaoguz, H., Jensfelt, P., Krug, R. (2018). Interactive, collaborative robots: challen­ges and opportunities. Proceedings of the 27th Inter­national Joint Conference on Artificial Intelligence (IJCAI-18), 2018, 18–25 https://doi.org/10.24963/ijcai.2018/3 Letouze, E., Pentland, A. (2018). Towards a human artifi­cial intelligence for human development. ITU Journal: ICT Discoveries, Special Issue 2 One Internet (2016). Global Commission on Internet Go­vernance. https://www.cigionline.org/initiatives/glo­ bal-commission-internet-governance Patel, K., McCarthy, M.P. (2000). Digital transformation: the essentials of e-business leadership. KPMG/Mc­Graw-Hill Pouchard, L., Ivezic, N., Schlenoff, C. (2000). Ontology engineering for distributed collaboration in manufactu­ring. Proceedings of the AIS2000 Conference. Rosen, Y. (2018). Computer agent technologies in collabo­rative learning and assessment. Encyclopedia of Infor­mation Science and Technology (4th edn.). IGI Global, 2402-2410 https://doi.org/10.4018/978-1-5225-2255-3.ch209 Simek, D., Sperka, R. (2019). How robot/human orchestra­tion can help in an HR department: A case study from a pilot implementation. Organizacija, 52(3), 204-217. https://doi.org/10.2478/orga-2019-0013 Simonova, M., Ilyukhina, L., Bogatyreva, I., Vagin, S., Ni­kolaeva, K. (2016). Conceptual approaches to forecast recruitment needs at the regional level. International Review of Management and Marketing, 6(5), 265-273. Teixeira, J.V.S., Reis, A.M., Mendes, F.B., Vergara, L.G.L. (2019). Collaborative Robots. Occupational and En­vironmental Safety and Health. Studies in Systems, Decision and Control, 202. Springer, Cham, 791-796. https://doi.org/10.1007/978-3-030-14730-3_83 Wilson, H.J., Daugherty, P.R. (2018). Collaborative intel­ligence: humans and AI are joining forces. Harvard Business Review, July–August 2018, 11. Wooldridge, M. (2002). An introduction to multi-agent systems. John Wiley and Sons, Chichester. Zang, S., Ye, M. (2015). Human resource management in the era of Big Data. Journal of Human Resource and Sustainability Studies, 3, 41-45. https://doi.org/10.4236/jhrss.2015.31006 Anton Ivaschenko, professor, PhD (2004) in mathematical modeling, DSc (2012) in management in social and economic systems is a Head of the Computer Science Department of Samara State Technical University. His areas of research include modeling of complex systems in digital economy and social media, intelligent and multi-agent systems. Alfiya R. Diyazitdinova, PhD (2004) in management in social and economic systems, associate professor at the Applied Informatics Department of Povolzhskiy State University of Telecommunications and Informatics. Her areas of research include IT design and management and automated decision-making support. Tatiyana Nikiforova is a PhD student at Samara State Technical University, earned her Engineer of Information Systems and Technologies degree in 2015. Her research interests include modeling, business analysis, digital economy and data analysis. Optimizacija racionalnega deleža inteligentnih tehnologij: Uporaba v storitvenih organizacijah Ozadje in namen: Narašcajoca vloga in vkljucenost umetne inteligence v sodobna digitalizirana podjetja vodi do znatnega zmanjšanja števila zaposlenih in preusmeritve dela osebja na nove zadolžitve. Vendar pa na številnih podrocjih, kot so storitve in podpora, popolna odprava zaposlenih cloveških virov še vedno ni mogoca. Predlaga se proucitev organizacijskega problema iskanja optimalnega deleža racunalniških agentov in cloveških akterjev v mešanem sodelovalnem okolju. Nacrtovanje / metodologija / pristop: Z uporabo tehnike semanticne in statisticne analize smo razvili izvirni model sodelovalne interakcije racunalniških agentov in cloveških akterjev ter metodo optimizacije, ki je nova pri prouceva­nju osredotocenosti izvajalcev pri izracunu kazalnikov skladnosti . Rezultati: Problem smo preucili na primeru avtomatizacije servisne službe. Upoštevanje semantike problemske domene v obliki ontologije uvaja logiko za boljšo porazdelitev in avtomatizacijo nalog. Zakljucki: V sodobnem digitaliziranem podjetju obstaja in je mogoce oceniti racionalno ravnovesje med racunal­niškimi agenti in cloveškimi akterji, kar postane pomemben pokazatelj njegove uspešnosti. Na splošno so cloveški igralci boljši za obdelavo nepredvidljivih dogodkov v realnem casu, medtem ko so agenti boljši pri modeliranju in simulaciji. Kljucne besede: Avtomatizacija storitvenih organizacij, Umetna inteligenca, Cloveški viri, Podpora odlocanju Appendix This appendix contains additional information describ­ing the data studied as an example of service desk auto­mation. The dataset contains 648 real calls that require the generation and processing of service tasks during a fixed period of time. From the total number of requests, the ratio processed by agents and actors is presented in Fig. 7. 74% of 648 calls were performed by agents, the remaining 26% by actors. Fig. 8 describes the distribution of call categories. Below are presented the main metrics for processing the requests, depending on the executor. The waiting time for tasks execution is presented in the “Boxplot by Group” chart in Fig. 9. The graph shows that the actors (group 1) have a higher idle time than the agents (group 2). The stan­dard deviation is in the range from half an hour to 3 hours. This is due to the fact that the queue of tasks is analysed unevenly by the actors, confirming that the focus of the actor is discrete and constantly changing. For group 2, the standard deviation does not exceed a couple of minutes, which suggests that the tasks for the agents are distributed at a speed that does not allow queuing. Fig. 10 shows that the processing time of tasks is sig­nificantly higher for the group 2 agents than for the group 1 actors. For approximately 75% of the tasks performed by group 2, the processing time does not exceed an hour, whi­le for group 1 this time is up to 3.5 hours. The maximum values for both groups are at a considerable distance from the median, which indicates the emergence of tasks that require significant time for both the agents and the actors. Based on the average processing time of calls in rela­tion to the number, the average speed of processing calls for the agents and the actors was determined. For an agent it was approximately 3 calls per hour, for an actor approx­imately 0.3 calls per hour. At the same time, the number of responses from the agent was 67, which corresponds to 14% of the total number of processed requests for the period. Let us evaluate the difference between the two in­dependent samples by the level of the sign of automatic execution (dependence on the executor) using the Mann-Whitney U-test (see Table 2). The differences are statistically significant at a high lev­el, the indicator p <0.05, i.e. the downtime (U = 22088.0) and duration of work (U = 19692.0) of the actors is higher than the agents. These results can be logically explained, since the automation and digitalisation were initially de­signed to optimise and intensify human labour. With the current distribution of tasks, one agent becomes equal to three actors in terms of performance. As a result of one-factor analysis for the variable dura­tion of tasks, a dependence on the category was revealed, illustrated by Table 3. ANOVA analysis of variance was used to determine the differences of execution time spent by the actors for each category of tasks (see Fig. 11). The actors have no failures as they are able to process any task. The tasks of category 1 take longer for successful processing than other categories. Agents have no differences in the average processing time for the tasks of various categories (see Fig. 12). For computer agents, statistically significant differenc­es were established between the number of failures and the number of successfully completed tasks by calculating the Mann-Whitney U-test (see Table 4). Despite the high failure rate (14%) in the amount of 67 requests in group 2, successfully completed tasks still prevail (413 in group 1). At the same time, the efficiency of agents in the direction of reducing the number of failures should be improved. The results of the statistical analysis showed that, de­spite the fact that the majority of tasks are processed by agents, and the processing time and waiting time for the agents are less than for the actors, the number of refusals for the agents is greater. This is confirmed by the “Boxplot by Group” plots. The agents are only effective in certain categories of tasks. The actors work less efficiently, but are more stable in solving any type of problem. This con­clusion was obtained using the Mann-Whitney U test and ANOVA analysis. Figure 7: Tasks processed by actors and agents Figure 8: Processed tasks by category Figure 9: Task waiting time depending on the type of executor Figure 10: Task duration depending on the executor Table 2: Mann-Whitney U test for the groups of actors and agents Variable Rank Sum for Actors Rank Sum for Agents U Z Z adjusted Valid N Actors Valid N Agents Downtime 71 785 138 491 22 088 8.613 9.479 166 482 Work dura­tion 61 319 106 012 19 692 7.821 7.969 163 415 Table 3: Univariate tests of significance for tasks duration Effect SS Degree of freedom MS F Intercept 0.295 1 0.295 80.461 Category 0.495 2 0.248 67.594 Error 2.106 575 0.0037 Figure 11: Dependency of tasks processing time by actors on the category Figure 12: Dependency of tasks processing time by agents on the category Table 4: Mann-Whitney U test for agents by sucessfully processed tasks and failures Variable Rank Sum Group 1 Processed Rank Sum Group 2 Failures U Z p-value Z adjusted p-value Valid N Group 1 Processed Valid N Group 2 Failures Category 100 633 14 807 12 529 1.24 0.215 2.62 0.009 413 67 AUTHOR GUIDELINES / NAVODILA AVTORJEM Manuscripts considered for publication in Organ­izacija (organizacija@um.si) are those which: • Contain original work – which is not published elsewhere in any medium by the authors or anyone else and is not under consideration for publication in any other medium. The author(s) is/are also responsible for any violations of the copyright regulations. • Are focused on the core aims and scope of the journal: Organizacija is an interdisciplinary peer reviewed journal that seeks both theoretically and practically oriented research papers from the area of organizational science, business information systems and human resources management. • Are clearly and correctly written – should contain all essential features of a complete scientific paper, should be written in a clear, easy to understand manner and be readable for a wide audience. • Are written in English – should be clearly and grammatically written, in an easily readable style. Attention to detail of the language will avoid severe misunderstandings which might lead to rejection of the paper. Correct language is the responsibility of the authors. Unless the author is an English native speaker, the paper must be proofread by a language editor, English native speaker All parts of the manuscript should be type-writ­ten (font size 12), with margins of 2.5 cm. Pages should be numbered consecutively throughout the manuscript. The text should be subdivided into numbered chapters. Figures and tables, consecu­tively numbered (Figure 1, Figure 2, ...; Table 1, Table 2, …) can be included in electronic form in the text. All lettering and figure elements must be large enough to be readable when figure or table has been reduced to fit journal page or column. All figures and tables must be specifically referred in the text. Colour pictures cannot be published in the printed version of the journal; colours appear only in the internet version. The paper should start with a cover page with names and mailing and electronic addresses of the authors. To assure the anonymity of the refereeing procedure the names of the authors should not appear in the text. Detailed Guidelines for Authors are available at https://sciendo.com/journal/orga - for Authors. All the papers will be reviewed by at least two ref­erees. Based on the opinions and suggestions of the reviewers, the editors accept the paper, demand minor or major enhancements, or reject the paper. If major enhancements are required the upgraded paper is reviewed again. Manuscripts can be submitted via journal web site (http://organizacija.fov.uni-mb.si). For fur­ther information and clarifications contact Organ­izacija’s editorial office (organizacija@um.si or joze.zupancic@um.si). Address of the Editorial office: University of Maribor, Faculty of Organizational Science Kidriceva cesta 55a 4000 Kranj, Slovenia Fax: +386-4-2374-299 Phone: +386-4-2374-245 V reviji Organizacija objavljamo znanstvene clan­ke, rezultate raziskovalnega dela avtorjev. Predlože­ni prispevki naj bodo napisani v angleškem jeziku. Imeti morajo strukturo IMRAD, ki je obicajna za znanstvena in strokovna besedila. Objavljamo de- la s predmetnega podrocja revije, ki še niso bila objavljena in niso bila poslana v objavo v kakšni drugi reviji ali zborniku. Avtorji so odgovorni za vse morebitne kršitve avtorskih pravic. Besedilo naj bo oblikovano za tiskanje na papirju in levo poravnano. Na zacetku prispevka, takoj za naslovom, naj bo povzetek (izvlecek) dolžine naj­vec 250 besed, kljucne besede, v koncni – sprejeti verziji clanka pa na koncu prispevka tudi kratek strokovni življenjepis vsakega od avtorjev (do 10 vrstic) in letnica rojstva (zaradi vnosa podatkov v knjižnicni informacijski sistem COBISS, v reviji letnica ne bo objavljena). Na prvi strani besedila naj bodo napisani le naslov prispevka, imena in (poštni in elektronski) naslovi avtorjev clanka, po možnosti tudi telefonska številka enega od avtorjev. Da bi za­gotovili anonimnost recenziranja, naj se imena av­torjev ne pojavljajo v besedilu prispevka. Na koncu clanka, za življenjepisi, naj bo slovenski prevod naslova, povzetka in kljucnih besed. Clanek naj bo razclenjen v oštevilcena poglavja. Naslovi clanka, poglavij in podpoglavij naj bodo napisani z malimi crkami, da so razvidne kratice. Slike in tabele v elektronski obliki vkljucite kar v besedilo. Besedilu so lahko priložene slike in/ali ta­bele na papirju v obliki pripravljeni za preslikavo. V tem primeru naj bo vsaka slika na posebnem listu, oštevilcene naj bodo z arabskimi številkami, v bese­dilu naj bo oznaceno, kam približno je treba uvrstiti sliko: na tem mestu naj bo številka slike/tabele in njen podnapis. Slike bomo praviloma pomanjšali in jih vstavili v clanek. Upoštevajte, da morajo biti oznake in besedila na vseh slikah dovolj velika, da bodo citljiva tudi pri velikosti slike, kot bo obja­vljena v reviji. Vse slike naj bodo crno-bele z be­lim ozadjem; barvnih slik v tiskani verziji revije ne moremo objaviti, barve so vidne le v spletni verziji. Clanki morajo biti pred objavo v Organizaciji lekto­rirani. Koncno verzijo mora lektorirati naravni go­vorec oz. lektor s primerljivim znanjem anglešcine. Podrobna navodila avtorjem za pisanje in oblikova­nje clankov so na https://sciendo.com/journal/orga - for Authors. Predložene prispevke pregledata in ocenita najmanj dva recenzenta. Na osnovi mnenj in predlogov re­cenzentov uredniški odbor ali urednik sprejmejo prispevek, zahtevajo manjše ali vecje popravke in dopolnitve ali ga zavrnejo. Ce urednik oziroma re­cenzenti predlagajo vecje popravke, se dopolnjeni prispevek praviloma pošlje v ponovno recenzijo. Clanke za objavo lahko predložite preko spletnega mesta http://organizacija.fov.uni-mb.si. Za nadalj­nje informacije in pojasnila se lahko obrnete na ure­dništvo Organizacije (organizacija@um.si ali joze.zupancic@um.si). Naslov uredništva: Univerza v Mariboru, Fakulteta za organizacijske vede Kidriceva cesta 55a 4000 Kranj Faks: 04-2374-299 Tel.: 04-2374-245 Prva slovenska revija za organizacijska in kadrovska raziskovanja in prakso. Revijo sofinancira Javna agencija za raziskovalno dejavnost Republike Slovenije. Ponatis in razmnoževanje deloma ali v celoti brez pisnega dovoljenja nista dovoljena. Izdajatelj: Univerza v Mariboru, Fakulteta za organizacijske vede Kranj, Založba MODERNA ORGANIZACIJA, Kidriceva cesta 55a, KRANJ, telefon: 04 23 74 200, telefax: 04 23 74 299, E-pošta: organizaija@um.si. Uredništvo revije: Kidriceva cesta 55a, 4000 Kranj, narocniški oddelek: 04 23 74 295. Letna narocnina: za pravne osebe za prvi naroceni izvod 51,47 EUR, drugi naroceni izvod 41,38 EUR, vsak nadaljnji 36,33 EUR, za posameznike 25,23 EUR. Cena posamezne številke je 9,08 EUR. Na leto izidejo 4 številke. Tisk: Tiskarna Koštomaj d.o.o. Naklada 200 izvodov. Organizacija is covered by the following services: Cabell’s Directory, CEJSH (The Central European Journal of Social Sciences and Humanities), Celdes, Clarivate Analytics - Emerging Sources Citation Index (ESCI), CNPIEC, Die Elektronische Zeitschriftenbibliothek, DOAJ, EBSCO - TOC Premier, EBSCO Discovery Service, ECONIS, Ergonomics Abstracts, ERIH PLUS, Google Scholar, Inspec, International Abstracts in Operations Research, J-Gate, Microsoft Academic Search, Naviga (Softweco), Primo Central (ExLibris), ProQuest - Advanced Polymers Abstracts, ProQuest - Aluminium Industry Abstracts, ProQuest - Ceramic Abstracts/World Ceramics Abstracts, ProQuest - Composites Industry Abstracts, ProQuest - Computer and Information Systems Abstracts, ProQuest - Corrosion Abstracts, ProQuest - Electronics and Communications Abstracts, ProQuest - Engineered Materials Abstracts, ProQuest - Mechanical & Transportation Engineering Abstracts, ProQuest - METADEX (Metals Abstracts), ProQuest - Sociological Abstracts, ProQuest - Solid State and Superconductivity Abstracts, Research Papers in Economics (RePEc), SCOPUS, Summon (Serials Solutions/ProQuest), TDOne (TDNet), TEMA Technik und Management, WorldCat (OCLC) CONTENTS - 2/2021 98 112 131 147 162 Jin-Nan WU, Mengmeng SONG, Joseph C. UGRIN, Lin LIU, Tingting ZHU Cyberloafing Research 1997-2019: A Citation-based Literature Review Mehmet ÇETIN, Gulmira SAMENOVA, Filiz TÜRKKAN, Ceylan KARATAS The Role of Daily Affect in Leader-Member Exchange: A Multilevel Investigation in Public Health Administration Phuong TRAN HUY, Hong Chuong PHAM Effect of Management Commitment to Service Quality on Car-hailing Drivers’ Service Behaviors: The Case of GrabCar in Vietnam Erik RUŽIC, Dragan BENAZIC The Impact of Internal Knowledge Sharing on Sales Department’s Innovativeness and New Product Commercialization Anton IVASCHENKO, Alfiya R. DIYAZITDINOVA, Tatiyana NIKIFOROVA Optimisation of the Rational Proportion of Intelligent Technologies Application in Service Organisations