Organizacija, Volume 49 Research Papers Number 1, February 2016 DOI: 10.1515/orga-2016-0001 Awareness and Attitude Towards Green IS in Slovenian Enterprises Alenka Baggia1, Alenka Brezavšček1, Matjaž Maletič1, Petra Šparl1, Hendry Raharjo2, Anja Žnidaršič1 1University of Maribor, Faculty of Organizational Sciences, Kidričeva cesta 55a, 4000 Kranj, Slovenia alenka.baggia@fov.uni-mb.si 2Chalmers University of Technology, Department of Technology Management and Economics, SE-412 96 Gothenburg, Sweden, hendry.raharjo@chalmers.se Background: This study draws upon the use of Information Systems in support of achieving sustainability, known as Green IS. Furthermore, this study builds on the premise that Green IS offers the opportunity for organizations to act proactively in terms of environmental preservation as well as to mitigate the effects of global climate change and other environmental problems. Aim: In particular, this study aims to assess the extent of awareness among managers regarding the use and the acceptance of Green IS in Slovenian enterprises. Method: Using empirical data based on a large-scale survey among senior managers within Slovenian enterprises this study utilized several statistical methods (such as t-test, analysis of variance and multiple linear regression) to analyse the research questions. Results: In general, findings seem to suggest that institutional mechanisms might be a plausible explanation for differences regarding the attitude towards Green IS adoption. For instance, enterprises with at least one implemented sus-tainability related certificate expressed higher levels of willingness to use Green IS in order to facilitate the achievement of sustainable development. Moreover, the results of the regression analysis revealed that both Institutional Mimetic pressure and Internal Environment Impact has positive impact on Green IS adoption. Conclusion: The main conclusion is that the internal environmental impact is considered the most influential factor of the attitude towards Green IS adoption. The culture or individual perception of managers and employees play an important role in the Green IS adoption. Indeed, enterprises that have no intention of improving their environmental performance, but adopt Green IS by the means of seeking legitimacy among external stakeholders, cannot provide a sustainable improvement in environmental management. Keywords: Green Information System, Sustainability, Environment, Information System 1 Introduction Global warming and other environmental threats have lately increased the level of environmental awareness (Dedrick, 2010). Individuals, organizations and governments are becoming aware of the necessity of sustainability for performing and managing work (Siedel, Recker, Pimmer & vom Brocke, 2010). Since information technologies and information systems represent 2% of CO2 emissions (WWF, Gartner Group, 2008) and this share is supposed to increase to 3% until 2020 (McKinsey Global Institute, 2008), the shift towards green or sustainable economy seems inevitable. Environmental issues have forced the information technologies and consequently information systems to become more environmentally friendly, enabling energy efficient business processes (Brooks, Wang & Sarker, 2012). It is also evident, that business process management (Siedel et al., 2011) and the shift towards the sustainable business practices can benefit substantially from Green information systems (Green IS) practices with Green Received: 19th September, 2015; revised: 7th November 2015; accepted; 6th January 2016 15 Organizacija, Volume 49 Research Papers Number 1, February 2016 information technology (Green IT) as an enabler (Esfahani, Rahman and Zakaria, 2015). Watson, Boudreau, Chen and Huber (2008) were the first to carefully distinguish between Green IS and Green IT. The concept of Green IS can be defined as the usage of information systems to enable sustainable development in economy (Boudreau, Chen & Huber, 2007; Watson et al., 2008). Green IS are enabled by a unique structure of people, processes and IT with a goal to achieve eco-capacity, eco- efficiency, eco-effectiveness and eco-collaboration (Butler, 2011). Whereas Brooks et al. (2012) sees Green IS as an incentive of using IT infrastructure to change organizational processes and/or practices in order to improve energy efficiency and decrease the level of environmental impact, while introduce environmental friendly products and services at the same time. To be even more precise, Green IS represent a wide spectre of solutions, where an information system (a Table 1: Operationalization of constructs Item GPP Green IS Adoption - Pollution Prevention GPP1 Our company encourages the usage of software for reduction of emissions GPP2 Our company encourages the usage of software for reduction of waste GPP3 Our company encourages the usage of software for reduction of hazardous and toxic materials GPL Green IS Adoption - Product Lifecycle GPL1 Our company encourages the usage of software for product life cycle management, which enables environmentally friendly sourcing and acquisition GPL2 Our company encourages the usage of software for product life cycle management, which enables environmentally friendly product/service development GPL3 Our company encourages the usage of software for product life cycle management, which enables environmentally friendly planning of production/service processes GPL4 Our company encourages the usage of software for product life cycle management, which enables environmentally friendly distribution and delivery GSD Green IS Adoption - Sustainable Development GSD1 Our company encourages the usage of software for online collaboration GSD2 Our company encourages the usage of software for teleworking GSD3 Our company encourages the usage of software for paperless business processes GSD4 Our company encourages the usage of software for measuring and monitoring of organizational environmental performance ICP Institutional Pressure - Coercive Pressure ICP1 The regulations are forcing our company to use green IS ICP2 Suppliers are forcing our company to use green IS ICP3 Important customers are forcing our company to use green IS IMP Institutional Pressure - Mimetic Pressure IMP1 Our company's main competitors, who adopted green IS, have benefited greatly financially IMP2 Our company's supply chain members, who adopted green IS, are perceived favourably by their customers IMP3 Our company's supply chain members, who adopted green IS, have benefited greatly financially IEI Internal Environment Impact IEI1 The management of our company would like to implement green IS IEI2 Employees in our company have proposed the implementation of green IS IEI3 Green IS are a frequent topic of conversation in our company 16 Organizacija, Volume 49 Research Papers Number 1, February 2016 combination of people, software and IT) has an additional functionality that enables a more sustainable performance of business process. Some of often used Green IS are information systems that enable energy consumption management, optimization of production processes, consolidation of data centres and operations, reuse and care for appropriate disposal of IT, micro and macro economical aspects, system performance, efficient usage of a system, as well as social and ethical aspects aligned with purchase. The Green IS research has mainly been focused to energy efficiency (Watson, Boudreau and Chen, 2010) while other forms of Green IS have mainly been neglected in the literature (e.g. information systems for waste reduction management). For example, Energy informatics, a special research field presented by Watson and Boudreau (2011), discusses the importance of IS for the reduction of energy consumption. According to several academic discussions (Melville, 2010; Watson, Boudreau & Chen, 2010, vom Brocke, Watson, Dwyer, Elliot & Melville, 2013), the research and development in the area of Green IS should be encouraged. Zheng (2014) presents organizational, technological and environmental mechanisms as Green IT/IS adoption motivators. Wati and Koo (2012) researched the motivational perspective of Green IS behaviour intention and actual use of Green IS and concluded that several self-determined motivational factors influence the behaviour intention to use Green IS, while there is no significant influence of behaviour intention on actual use. The actual use of Green IS is according to Wati and Koo (2012) influenced by external pressure, not only by motivational factors. Since all the aspects of sustainability have not been equally researched, and the focus is mainly on environmental aspect of sustainability (Farzad and Junker, 2015), there is a lack of literature on general awareness of social and economic aspect of sustainability related to IS. Therefore, we decided to follow the research on senior managers' attitude to Green IS in Malaysian companies presented by Gholami, Sulaiman, Ramayah and Molla (2013) and divide the concept of Green IS into three basic forms: pollution prevention, product stewardship and sustainable development. Accordingly, the aim of our research was to evaluate the general awareness and the attitude towards Green IS in Slovenian enterprises. 2 Methods A survey was developed based on the previous study presented by Gholami et al. (2013) and Mishra, Akman & Mishra (2014). Several items were withdrawn based on the issues of translation and importance in regional circumstances, while several items were added based on previous research (Baggia and Brezavscek, 2015). Items describing the Green IS adoption were obtained from Gholami et al. (2013). Two additional items (Envi- ronmentally friendly product/service development and Planning of production/service processes) were added for the description of Green IS adoption regarding the enterprise's attitude toward the software product lifecycle management, while one item (Measuring and monitoring of organizational environmental performance) was added for the description of enterprise's attitude towards sustainable development. Items discussing the institutional pressure were all adopted from Gholami et al. (2013), while items discussing the internal environment impact were adopted from Mishra, Akman & Mishra (2014). Items discussed in this paper are presented in Table 1. Five point Likert type scale was used to measure the items. 2.1 Sampling, data and statistical methods used The presented survey was performed as a web survey used to collect data about attitude toward Green IS among managers of Slovenian enterprises. Invitations to the web survey were sent via email to 3623 randomly chosen enterprises in Slovenia, where the sample matched the demographi-cal structure of Slovenian companies according to region and main activity of the enterprise, representing 2% of all Slovenian enterprises. Companies with less than two employees were excluded from the sample, based on the assumption that their usage of IS in general is limited. In the period from 25th of May 2015 to 25th of August 2015 we received 222 responses in total. The survey was addressed to the director of the enterprise or to the director of the informatics sector. For the analysis of research questions, different statistical methods were used: one sided independent samples /-test, ANOVAs or Robust test of equality of means and multiple linear regression. 2.2 Characteristics of the enterprises in the sample On the question about the main activity of the enterprise (classified according to Statistical Classification of Economic Activities in the European Community, NACE Rev 2, 2008) 185 enterprises provided their response. The largest proportion marked Other Service Activities (13.5%), Information and Communication (13.0%), and Manufacturing (10.3%) (Figure 1). The sample consists of 23% micro enterprises (up to 9 employees), almost half of the enterprises were small sized (46%), one quarter of enterprises were medium sized (50 to 249 employees), while 6% were large enterprises with at least 250 employees. The questionnaire was fulfilled by 58.2% of men and 41.8% of women. The age of the respondents varies from 23 to 65 years, with mean age 43.5 years (s=9.58 years). 17 Organizacija, Volume 49 Research Papers Number 1, February 2016 Figure 1: Structure of enterprises in the sample according to their main economic activity A - Agriculture, Forestry And Fishing B - Mining And Quarrying C - Manufacturing D - Electricity, Gas, Steam And Air.. E - Water Supply: Sewerage. Waste. F - Construction G - Wholesale And Retail Trade: Repair Of.. .1 H - Transportation And Storage I - Accommodation AndFood Service Activities J - Information And Communication K - Financial And Insurance Activities L - Real Estate Activities M - Professional, Scientific And Technical. N - Admini strative And Support Servi ce. O - Publi c Admini stration And Defence:. P - Education Q - Human Health And Social Work Activities R - Arts, Entertainm ent And Recreation S - Other Service Activities 10.3% 13.0% 0.0% 5.0% 10.0% I 13.5% 15.0% 20.0% 3 Results According to the topic of items in particular subsection, we first discuss the organizational circumstances leading the enterprise to shift their viewpoint to sustainability, second we discuss the promotion of Green IS usage in enterprises. In the last subsection, we discuss the correlations between different circumstances and the promotion for Green IS adoption. 3.1 Descriptive statistics There are several external and internal issues leading the enterprise to shift their viewpoint to a more sustainable and further to use Green IS. Descriptive statistics for measured items and constructs are presented in Table 2. Enterprises receive the highest coercive pressure by regulations (3.16), while suppliers (2.64) and important customers (2.62) have lower impact on usage of Green IS. Among three items of mimetic pressure, the highest average grade received the statement that company's main competitors, who adopted Green IS, have benefited greatly financially (3.04). The highest internal environmental impact on implementation of Green IS have managers (3.19), while the implementation of Green IS is not a frequent topic in the conversations (2.91). Regarding the adoption of Green IS, all three measured items of Pollution Prevention due to use of software were on average estimated high; 3.89 for software for reduction of waste, 3.87 for software for reduction of hazardous and toxic materials, and 3.80 for software for reduction of emissions. Among four items of Green IS Adoption for Product lifecycle, the companies encourage the most the usage of software that enables planning of production/service processes (3.74) and product/service development (3.73). Encouragement for adoption of Green IS for Sustainable development is the highest for online collaboration (3.96), followed by paperless business processes (3.67), while the least sympathy received measuring and monitoring of organizational environmental performance (2.90) and telework-ing (2.93). In the questionnaire 10 green certificates were listed, and the enterprises marked if they already had it or not and if they are in the process of implementation: ISO 14001 -Environmental Management System, ISO 22000 - Food safety management, ISO 26000 - Social responsibility, ISO 50001 - Energy management, EMAS - Eco-Management and Audit Scheme, EcoLabel - label promoting environmental excellence, FSC - Forest Stewardship Council, Family friendly certificate - European work and family audit, SA8000 - Social Accountability International, OHSAS 18001 - Occupational Health and Safety Management 18 Organizacija, Volume 49 Research Papers Number 1, February 2016 Table 2: Descriptive statistics of the measured items and constructs Item N Mean SD GPP 210 3.86 0.760 GPP1 211 3.80 0.893 GPP2 211 3.89 0.874 GPP3 211 3.87 0.863 GPL 200 3.72 0.756 GPL1 205 3.70 0.844 GPL2 204 3.73 0.813 GPL3 203 3.74 0.875 GPL4 204 3.69 0.836 GSD 202 3.37 0.771 GSD1 203 3.96 0.872 GSD2 205 2.93 1.165 GSD3 205 3.67 0.942 GSD4 204 2.90 1.036 ICP 205 2.80 0.782 ICP1 205 3.16 0.952 ICP2 205 2.64 0.916 ICP3 205 2.62 0.892 IMP 101 2.98 0.805 IMP1 113 3.04 0.999 IMP2 130 2.99 0.858 IMP3 109 2.97 0.855 IEI 196 3.04 0.816 IEI1 197 3.19 0.869 IEI2 198 3.01 0.934 IEI3 198 2.91 0.965 Systems. Respondents were also allowed to input additional certificates. There are 136 enterprises that have no sustainability related certificates (60.7% out of 224) and 88 enterprises that have from 1 to 10 green certificates, with mean value equal to 2.7 (s=2.1). More precisely, 20% of enterprises have ISO 14001 - Environmental management, 6% have ISO 22000 - Food safety management, 7 % have ISO 26000 - Social responsibility, 8 % have ISO 50001 - Energy management, 4% EMAS - Eco-Management and Audit Scheme, 7% EcoLabel, 11% have Certificate Family Friendly Enterprise, 5% have SA 8000 - Social Accountability International, 17% have OHSAS 18001 - Occupational Health and Safety Management. The certificates which are in the process of implementation in 9 % of enterprises are ISO 14001 and ISO 50001. 3.2 Analysis of the research questions RQ1: Do the enterprises with at least one implemented sustainability related certificate have on average higher estimates of importance of Green IS adoption than the enterprises who do not have any of sustainability related certificates? The first research question was analysed with one sided independent samples t-tests. Since we assumed that the enterprises with at least one sustainability related certificate will have higher Green IS awareness, one sided t-test was performed on all of 11 items expressing the enterprise's attitude towards the implementation of Green IS and other sustainability issues (2-sided p-values obtained in SPSS were recalculated as p/2). For four out of eleven items the research question was answered positively. Detailed results are presented in Table 3 and Figure 2. For two out of three items regarding Pollution Prevention due to Green IS Adoption the research question was answered positively. The statement »Our company encourages the usage of software for reduction of emis-sions.« was evaluated statistically significantly higher by enterprises who implemented at least one sustainability related certificate (3.94) than by enterprises where they do not have any of sustainability related certificates (3.73) at 5% significance level (/=-1.753, p=.042). The statement that company encourages the usage of software for reduction of hazardous and toxic materials was on average evaluated higher by the enterprises with at least one sustainability related certificate (4.06) compared to the other enterprises (3.77) at 5% significance level (t=-2.305, p=.011). None of four items of Product lifecycle due to Green IS Adoption was evaluated statistically significantly higher by the enterprises with implemented sustainability related certificates, while half of the items in Sustainable development were evaluated statistically significantly higher by the enterprises with implemented sustainability related certificates as assumed. The enterprises with implemented sustainability related certificates encourage paperless business processes (3.85) more than the others (3.58) at 5% significance level (t=-1,960, p=.026). The same goes for measuring and monitoring the organizational environmental performance, since the average estimate for enterprises with sustainability related certificates is equal to 3.18 and statistically significantly lower (2.75) for the enterprises with no green certificates (t=-2,886, p=.002). RQ2: Do the enterprises with at least one implemented sustainability related certificate have on average higher estimates of importance of Institutional coercive pressure, Institutional mimetic pressure and Internal environment impact than the enterprises who do not have any of sustainability related certificates? 19 Organizacija, Volume 49 Research Papers Number 1, February 2016 Table 3: Descriptive statistics and results of t-test for all 20 items according to enterprises who have implemented sustainability related certificate or not Item Number of implemented environmental certificates /-test for equality of means None At least one N Mean SD N Mean SD t df p (2-sided) p (1-sided) GPP1 139 3.73 .915 72 3.94 .837 -1.735 155.5 .085 .043 GPP2 139 3.83 .884 72 4.01 .847 -1.474 209 .142 .071 GPP3 139 3.77 .854 72 4.06 .854 -2.305 209 .022 .011 GPL1 134 3.68 .855 71 3.73 .827 -.429 203 .668 .334 GPL2 134 3.67 .839 70 3.84 .754 -1.432 202 .154 .077 GPL3 134 3.69 .869 69 3.84 .885 -1.131 201 .259 .130 GPL4 134 3.68 .800 70 3.70 .906 -.169 202 .866 .433 GSD1 131 3.90 .867 72 4.07 .877 -1.321 201 .188 .094 GSD2 133 2.87 1.202 72 3.04 1.093 -.994 203 .321 .161 GSD3 133 3.58 .955 72 3.85 .899 -1.960 203 .051 .026 GSD4 132 2.75 .984 72 3.18 1.079 -2.886 202 .004 .002 ICP1 133 3.05 .976 72 3.36 .877 -2.291 203 .023 .011 ICP2 133 2.56 .941 72 2.79 .855 -1.764 203 .079 .040 ICP3 133 2.50 .858 72 2.83 .919 -2.559 203 .011 .006 IMP1 66 3.11 .963 47 2.94 1.051 .890 111 .375 .812a IMP2 79 2.92 .844 51 3.10 .878 -1.130 128 .261 .130 IMP3 68 2.90 .794 41 3.10 .944 -1.189 107 .237 .119 IEI1 125 3.07 .863 72 3.39 .848 -2.497 195 .013 .007 IEI2 126 2.90 .902 72 3.21 .963 -2.281 196 .024 .012 IEI3 126 2.78 .954 72 3.15 .944 -2.670 196 .008 .004 a One-sided p-value of the item IMP1 was recalculated as 1-p(2-sided)/2, since averages in the sample are just the opposite according to assumption in the second research question. The second research question was analysed with the one sided independent samples t-tests. Since we assumed that the enterprises with at least one sustainability related certificate will have higher estimate of influence from institutional pressure, one sided t-test was performed on all of 9 items expressing the enterprise's institutional pressure on sustainability issues (2-sided p-values obtained in SPSS were recalculated as p/2). For six out of nine items the research question was answered positively. Detailed results are presented in Table 3 and Figure 2. All three items in Institutional coercive pressure were on average evaluated lower in the enterprises without sustainability related certificate. The enterprises that implemented at least one sustainability related certificate evaluated force of regulations to use Green IS higher (3.36) than the others (3.05) at 5% significance level (/=-2.291, p=.011). Similarly, the impact of suppliers to use Green IS was on average evaluated higher in the enterprises who already have sustainability related certificates (2.29) than in the enterprises with no sustainability related certificates (2.56) at 5% significance level (¿=-0.1764, p=.040). Higher impact of important customers on use of Green IS is detected by the enterprises who already have sustainability related certificates (2.83) than by the others (2.50) at 5% significance level (t=-2.559, p=-.006). None of three items in Institutional Mimetic pressures were evaluated statistically significantly lower by the enterprises with no sustainability related certificate, while the contrary is true for three items of Internal environmental impact. Regarding the Internal Environment Impact, in the enterprises with implemented sustainability related certificates managers have higher desires to implement Green IS (3.39) than in the other enterprises (3.07) at 5% significance level (¿=-2.281, p=.007). Similarly, the initiative of employ- 20 Organizacija, Volume 49 Research Papers Number 1, February 2016 Figure 2: Mean values of 20 items according to enterprises who have implemented sustainability related certificate or not ees for implementation of Green IS is higher in the employees where they already have any of sustainability related certificates (3.21) than in the others (2.91) at 5% significance level (/=-2.281,/>=.012). In the enterprises where they already have implemented at least one sustainability related certificate Green IS are more frequent topic (3.15) than in the others (2.78) at 5% significance level (/=-2.670, p=.008). According to the number of employees, the enterprises are divided into four categories: micro, small, medium, and large. Analyses of variance (ANOVAs) or Robust test of Equality of means in case on unequal variances were used to test whether there exist statistically significant differences in average estimates between enterprises of different sizes of the items. In addition, to investigate groups where the differences exist post hoc tests were used. RQ3: Are there any differences in the estimates of importance of Green IS adoption among categories according to the size of enterprise? Statistically significant differences at 5% significance level were revealed only for one measured item (Table 4 and Figure 3): GSD3 - enterprise encourages use of Green IS for paperless business processes (F=3.766, p=.012). Games-Howell post hoc test reveals that mean values of GSD3 differentiate significantly among micro (3.47) and large enterprises (4.25) (p=.003) and small (3.63) and large enterprises (p=.004) at 5% significance level. RQ4: Are there any differences in the estimates of importance of Institutional coercive pressure, Institutional mimetic pressure and Internal environment impact among categories according to the size of enterprise? Statistically significant differences at 5% significance level were revealed only for two measured items (Table 4 and Figure 3): ICP1 - the regulations are forcing the enterprise to use green IS; (F=4.032, p=.09) and IEI2 - employees in our company have proposed the implementation of Green IS (F=2.747, p=.044). Games-Howell post hoc test reveals that mean values of ICP1 show statistically significant differences only between small (2.99) and large enterprises (3.83) at 5% significance level (p=.024). Gabriel post hoc test reveals that for IEI2 significant differences in average estimates exist between small (2.89) and large (3.19) enterprises at 5% significance level (p=.046). RRobust test (more precisely, Brown-Forsythe test) in the case of unequal variances among groups. 21 Organizacija, Volume 49 Research Papers Number 1, February 2016 Table 4: Descriptive statistics and results of ANOVAs for 20 items according to the size of enterprises Item Number of employees ANOVA/ RRobust test 1-9 10-49 50-249 More than 500 N Mean SD N Mean SD N Mean SD N Mean SD F P GPP1 45 3.82 .984 88 3.72 .909 48 3.83 .859 12 4.08 .900 .658 .579 GPP2 45 4.00 .905 88 3.83 .874 48 3.92 .846 12 4.08 .900 .562 .640 GPP3 45 3.84 .999 88 3.84 .786 48 3.88 .914 12 4.00 .953 .124 .946 GPL1 44 3.77 .912 87 3.56 .872 48 3.85 .772 12 3.67 .651 1.398 .245 GPL2 43 3.65 .948 87 3.71 .761 48 3.85 .799 12 3.92 .669 .706 .550 GPL3 44 3.59 .996 86 3.76 .825 48 3.90 .905 11 3.82 .603 .943 .421 GPL4 44 3.64 .892 86 3.71 .795 48 3.69 .926 12 3.67 .651 .074 .974 GSD1 45 3.93 .751 87 3.87 .950 48 4.06 .885 12 4.42 .515 1.620 .186 GSD2 45 2.96 1.127 88 2.84 1.202 48 3.02 1.158 12 3.58 .900 1.513 .213 GSD3 45 3.47 1.079 88 3.63 .963 48 3.85 .772 12 4.25 .452 3.766 .012R GSD4 45 2.87 1.140 87 2.84 .987 48 2.94 .998 12 3.50 1.000 1.493 .218 ICP1 45 3.11 .910 88 2.99 .851 48 3.38 1.084 12 3.83 .835 4.032 .009R ICP2 45 2.62 .912 88 2.57 .868 48 2.79 1.010 12 2.83 1.115 .765 .515 ICP3 45 2.49 .869 88 2.63 .835 48 2.79 1.031 12 2.67 .985 .880 .452 IMP1 26 3.12 .993 45 2.91 1.125 29 2.97 .823 9 3.67 .707 1.556 .205 IMP2 30 3.10 .885 55 2.93 .879 31 2.90 .831 9 3.11 .782 .412 .744 IMP3 24 3.04 .955 43 2.91 .895 28 2.89 .786 8 3.25 .463 .490 .690 IEI1 44 3.09 .884 88 3.11 .903 47 3.32 .810 12 3.58 .793 1.578 .196 IEI2 45 2.93 .939 88 2.89 .915 47 3.19 .947 12 3.58 .900 2.747 .044 IEI3 45 2.89 1.027 87 2.84 1.010 48 3.00 .875 12 3.17 .835 .577 .631 In the questionnaire, each enterprise has selected one of 19 activities from NACE Rev 2 classification. We reorganized this classification into 7 enterprise categories, as follows: • Energy: B, D, E • Industry: C, F • Wholesale, retail, hotels & restaurants: G, I • Information and communication: J • Education, science and entertainment: M, P, R • Public administration, health and social work: O, Q • Other: A, H, K, L, N, S RQ5: Are there any differences in estimates of importance of Green IS adoption among seven groups of enterprises according to their main activity? Among 11 items ANOVAs revealed that only two items from Sustainable development have statistically significant estimates among seven groups of enterprises according to their activity (Table 5, Figure 4). The first one is online collaboration (F=2.470, p=.027), where Games-Howell post hoc test revealed that statistically significant differences exist between Information and communication sector (4.33) and Wholesale, retail, hotels and restaurants (3.48) (p=.022) at 5% significance level. Second item is teleworking (F=4.667, p=.000), where statistically significant differences at 5% significance level exist between Information and communication (3.92) and one of the following sectors: Wholesale, retail, hotels and restaurants (2.43) (p=.000), Public administration, health and social work (2.67) (p=.004) and Other activities (2.71) (p=.000). RQ6: Are there any differences in estimates of importance of Institutional coercive pressure, Institutional mimetic pressure and Internal environment impact among seven groups of enterprises according to their main activity? ANOVAs revealed (Table 5), that none of the items from the group of organizational circumstances differ statistically significant. 22 Organizacija, Volume 49 Research Papers Number 1, February 2016 Figure 3: Mean values of three items showing statistically significant differences according to four groups of enterprises according to size RQ7: What impact have Institutional coercive pressure (ICP), Institutional mimetic pressure (IMP) and Internal environment impact (IEI) on Green IS adoption? The seventh research question was analysed by multiple linear regression. Three independent variables were calculated as the mean value of measured items in the corresponding construct, while Green IS adoption was calculated as mean value of eleven items included in Pollution Prevention, Sustainable development, and Product lifecycle. Before examining the regression model itself, correlations among included variables were investigated to rule out the possible problem of multicollinearity. All correlation coefficients among variables are statistically significant at 1% significance level. More precisely, correlations among dependent variable and predictors are as follows: the highest positive correlation is between IEI and Green IS adoption (r=0.549), followed by correlation coefficient between IMP and Green IS adoption (r=0.461), and correlation between ICP and Green IS adoption (r=0.412). The highest correlation among predictors is between IEI and ICP (r=0.420), followed by correlation between IEI and IMP (r=0.376), and ICM and ICP (r=0.365). Since none of the correlation coefficients is above 0.8, the multicollinearity is not the problem in our data. Another indicator to rule out the multicollinear-ity problem are variance inflation factors (VIFs) or the tolerance statistics. The VIFs statistics are for all three predictors in range from 1.227 and 1.493 and therefore relatively close to 1, while the lowest tolerance statistic for ICP is 0.668 which is highly above 0.2. The overall model fit was assessed. R2 is equal to 0.41, indicating that 41% of variance of the Green IS adoption could be explained with three predictors. Durbin-Watson statistic is 1.368 indicating that errors of the model are independent. Results of ANOVA suggests that the proposed model fits well (7=21.305, p=.000). Based on unstandardized regression coefficients, the regression model is as follows: Green IS adoption = 1.743 + 0.047 ■ ICP + 0.226 ■ IMP + 0.365 ■ IEI Coefficients for IMP and IE differ statistically significantly from zero (p=.003 and p=.000, respectively) at 5% significance level. Therefore, we can conclude that both Institutional mimetic pressure (IMP) and Internal Environment Impact (IEI) have a positive impact on Green IS adoption. 4 Conclusions According to Melville (2010), information systems are one of the inadequately understood enablers of sustainable business practices. It was therefore not expected that only 15.3% 23 Organizacija, Volume 49 Research Papers Number 1, February 2016 .g o a .15 S 'K £ ■s << S § S 3 -o £ 24 v £ - C1 ri ® .277 .501 .615 .546 MD .6 .572 .985 .0 .000 .092 .098 t-- .493 .251 .668 .974 .780 .125 .167 .266 k 1.263 .893 .744 .833 .681 .798 .169 2.470 4.667 1.848 1.815 1.275 .905 1.319 .677 7 o