136 Sodobna pedagogika/Journal of Contemporary Educational Studies Let./Vol. 70 (136) Št./No. 4/2019 Str./pp. 136–150 ISSN 0038 0474 Anka Jurčević Lozančić, Sanja Basta, and Ivan Šerbetar Teachers’ Attitudes Towards Collaboration with Parents: Development and Evaluation of the Questionnaire Abstract: The interest in research on collaboration with parents is becoming more apparent as family engagement changes. Such engagement is conditioned by the family’s interest in changing its position in general and in becoming involved in the curriculum of the educational institution. The questionnaire was constructed and carried out in order to evaluate teachers’ attitudes regarding collaboration with pa- rents. Exploratory factor analysis was employed on a sample of 545 school and preschool teachers. Factor structure was established using principal component analysis, and a three-factor solution was obtained, explaining 42% of the variance, cumulatively. The extracted components supported previous theoretical underpinnings about three dimensions—parental involvement in education, teacher competences for working with parents and expectations from collaboration—which may shed light on the teacher-parent partner- ship. The Cronbach’s alphas for those dimensions were .833, .854 and .633, respectively. In the second part of the study , a confirmatory factor analysis, using a new sample (N = 268), was administered on the three-factor model of the data. With the exception of significant χ 2 ’s (371.41; p < .01), all other indices of fit of the respecified model indicated a close fit of the data to the model, which was demonstrated by the indices of χ 2 /df = 1.89, while the RMSEA with a 90% confidence interval was 0.057 [CI 0.048 - 0.066], the NNFI reached a value of 0.94 and the CFI was 0.95. The results provide guidelines for improving initial teacher education and point to the knowledge and skills needed for empowering preschool and primary school teachers in their collaboration with parents. Keywords: teacher-parent partnerships, teacher-parent collaboration, teacher-parent questionnaire, teacher competences UDC: 37.064.1 Scientific paper Anka Jurčević Lozančić, Phd., full professor, University of Zagreb, Faculty of Teacher Education, Savska cesta 77, 10000 Zagreb, Croatia; e-mail: anka.jlozancic@ufzg.hr Sanja Basta, mag. ed., 22. June Primary school, Franje Lovrića 27, 44000 Sisak, Croatia; e-mail: sanja. basta2@gmail.com Ivan Šerbetar, PhD., assistant professor, University of Zagreb, Faculty of Teacher Education, Savska cesta 77, 10000 Zagreb, Croatia; e-mail: ivan.serbetar@ufzg.hr Jurčević Lozančić, Basta, Šerbetar 137 Introduction Collaboration with parents encourages the establishment of a relationship between preschool teachers, teachers and parents that is marked by quality and respect in a harmonious and collaborative activity (Epstein 2001; Gestwicki 2016; Ward 2013). In addition to mutual involvement and equality , the responsibility for reaching common goals is recognised (Cox-Petersen 2011). Collaboration is not the starting point in this relationship, rather it is a process with a structure and agenda that results in support as the outcome. This, implies finding ways or means of achieving more intensive collaboration between the family and the educational institution, which helps parents to more successfully realise their parental role, develops preschool teachers’ and teachers’ competences and helps the overall per- ception of the child and the family . The partnership between parents and preschool teachers and teachers is an important factor for a child’s development in an insti- tutional context (Epstein 2001). It enables a child to feel accepted, safe, satisfied and happy in an environment that positively affects the development of abilities and which is in agreement with the child’s needs, abilities and specificities. Hornby (2011), Cox-Petersen (2011) and Ward (2013) stated that two-way communication, mutual support, joint decision-making and mutual improvement of learning are the most important characteristics of this collaboration. According to Ljubetić (2014), partnership implies a continuous awareness of the importance of better understanding the partners in a relationship, the communica- tion processes that occur between them, the often unexploited potential they have and, ultimately , wellbeing, which quality partnerships between preschool teachers, parents and teachers bring. Ljubetić (2014) also pointed out the positive correlation between a child’s success and the frequency of parents’ active participation and, accordingly , concludes that a child’s progress is better if a parents’ involvement is early on and for as long as possible. Good collaboration aims to establish continuity in education, and it ensures effective relationships between all collaborators, which is key for their further development and for the development of a child-oriented curriculum and institutional curriculum. Cankar, Deutsch and Sentocnik (2012) confirmed that both teachers and parents recognise the importance of collaboration, 138 Sodobna pedagogika/Journal of Contemporary Educational Studies Jurčević Lozančić, Basta, Šerbetar and, while teachers’ attitudes towards it are more optimistic than parental ones, teachers express discomfort in communication with parents due to the constraints of the institutional framework of the school system. Therefore, within their curricular documents, educational institutions should foster balanced views of a child and his/her development and education, as different attitudes form obstacles and impediments in creating mutual understanding and collaboration. Previous research showed that the crucial importance of attitudes for explaining social behaviour is generally based on the prognostic validity of at- titudes on behaviour, i.e. a person will behave in agreement with his/her attitude. Because of that, establishing specific attitudes of preschool and primary school teachers towards collaboration with parents is particularly important as they are direct creators, organisers and implementers of their programme of work, which they assess, change and improve (Cox-Petersen 2011; Jurčević Lozančić 2016). Kušević (2016) pointed out the lack of empirical data that is relevant for the Croatian edu- cational context and stated the need to pursue the research on specific aspects of collaboration and educational institutions. Since, to the best of our knowledge, there is no instrument in the Croatian context that quantitatively assesses attitudes of preschool and primary school teachers regarding teacher-parent collaboration and the self-perception of teacher competences for implementing such collaboration, the design and evaluation of an adequate instrument is the main aim of this study . Method Instrument and procedure The Teachers’ attitudes towards collaboration with parents questionnaire is meant to contribute to the understanding of teachers’ attitudes towards working with parents. The introductory part of the questionnaire provides participants’ demographic data, including their gender , profession, former education, professional development and time spent in service. The main part of the questionnaire assesses teachers’ attitudes towards partnerships with parents. The items were rated on a five-point Likert scale (1 “strongly disagree” to 5 “strongly agree”). The questionnaire was designed based on the theoretical model and perception of the parental role and school’s role in the education of a child (e.g. Sheridan & Kratochwill 2007). The traditional orientation assumes a relationship where parents hand over the responsibility of a child’s education to the school, which, in turn, supports such an attitude and does not expect parents’ active involvement in school life. Communication with parents is occasional, often focusing on problems, and, in general, it is initiated by the school. The concept of a partnership with parents assumes that there is a community and is based on the development of collaborative relationships. In a partnership, roles are clear and supportive, educational goals are outlined jointly and the active role of all stakeholders is mutually defined, agreed on and accepted. Referring to the concept of a collaborative partnership, Ward (2013) concludes that collaboration skills are preconditions for its development. According Teachers’ Attitudes Towards Collaboration with Parents: Development and Evaluation ... 139 to the mentioned author , respecting parents as equal partners is the foundation for developing positive attitudes regarding the involvement of the parents. The questionnaire was developed through several steps. The pilot ver- sion, which contained 33 items, was primarily applied to establish its appro- priateness and clarity of wording. It was carried out on a pilot sample of 36 preschool teachers and 13 primary school teachers. After discussion with the respondents followed by screening of the data, some items were removed and others were rephrased. Data for the final evaluation of the questionnaire were gathered during several instances of administering the questionnaire to ele- mentary and preschool teachers during their professional development meetings. In the second part of the study , the finalised form of the questionnaire was applied to the new sample of respondents with the purpose of establishing the structural validity of the instrument. Participants A total of 545 participants took part in the first part of the study. The vast majority of them were women 96.2% (n = 525), and only 3.8% (n = 21) were men. The youngest participant was 24 years old and the oldest was 65, while the mean age was 41.87 (±9.85) years. Approximately , one half of them were primary school teachers (n = 290; 53.1%), while the other half were preschool teachers (n = 255, 46.7%), and one participant marked “other” for his/her vocation. Most of the participants had a college (n = 290; 53.1%) or university (n = 236; 43.2%) degree, while just a small portion of them only had a high school degree (n = 13; 4%) or had finished postgraduate studies (n = 5; 0.9%). Demographic data from two respondents were missing. The second part of the study included 262 subjects, of which 89 of them were primary school teachers and 173 were preschool teachers. Regarding gender , 261 of them were female and only 1 was male. The average age was 43.70 (±10.16) years. Results Exploratory factor analysis Exploratory factor analysis was performed to establish the internal dimensions of the questionnaire.Initially , the usual assumptions for factor analysis were checked. Firstly, we examined the correlations for multicollinearity, but highly correlated (r > .9) variables were not found. However, three items that correlated by less than .3 with at least one other item were found. These were excluded. Secondly, we checked the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and found a KMO of .90, which is well above the recommended value of .6 (Tabachnick & Fidell 2007), and Bartlett’s test of sphericity was significant (χ 2 (435) = 5478.20, 140 Sodobna pedagogika/Journal of Contemporary Educational Studies Jurčević Lozančić, Basta, Šerbetar p = .000). In the correlations of the anti-image correlations matrix, we found no items with correlations less than the recommended .5. Principal component analysis (PCA) was used to establish the factor structure of the questionnaire. Since the components were correlated (Table 1), promax factor rotation with Kaiser normalisation was chosen (Tabachnik & Fidell 2007). Five factors in the initial solution had eigenvalues greater than 1. The first, second and third factors explained 25%, 11% and 6% of the variance, respectively . Although the fourth and fifth factors had eigenvalues just above 1, a three-factor solution, with 42% of the overall variance explained, was preferred, based primarily on previous theoretical underpinnings but also because the number of primary loadings was insufficient, and there were difficulties in interpreting those factors. Component 1 2 3 1 1.00 2 .325 1.00 3 .397 .216 1.00 Table 1: Component correlation matrix During several iterations, a total of four items were excluded due to having a primary factor loading of less than .4, while another three items were eliminated because of cross-loadings greater than .3 that were accompanied by weak primary loadings. Finally, 23 items remained in the questionnaire, and none of the com- munalities were less than .30, while the majority of them were above .4, and five of them were above .5. In the final solution, ten items remained in forming the first component, i.e. parental involvement in education; nine items remained in forming the second component, i.e. teacher competences for working with parents; and four items remained in forming the third component, i.e. expectations from collabora- tion. The component loading matrix and communalities for the final solution are shown in Table 2. Components 1 2 3 Comm F/C % 19. Parents should be allowed to join the classroom. .727 -.209 .470 8/23 24. Parents should participate in developing the curriculum of the educational institution. .718 -.235 .499 9/11 36. Parents should participate in decision-making regarding life and work in educational institutions and, consequently, regarding their children’s progress. .673 .457 3/23 35. It is recommended that parents are invited to participate in preschool/school activities, field trips and events. .628 .427 1/41 41. Parents should be able to attend meetings where they will have a chance to talk to other parents regarding education, upbringing and developmental issues they come across, and to exchange experiences. .615 .401 2/37 Teachers’ Attitudes Towards Collaboration with Parents: Development and Evaluation ... 141 26. At least once per year, mutual creative workshops for children and parents should be organised. .609 .406 0.5/51 45. Pedagogical workshops for parents should be organised to help them solve educational issues regarding their children. .607 .245 .497 0.5/47 30. Teachers should help the parents regarding a child’s development with advice, education and referrals to professional literature and external professional help. .556 .208 .469 0.5/43 33. Planning personal professional growth and development for quality work with parents is necessary. .547 .356 0.2/47 29. At the end of the year, with the help of children, parents should be presented with the achievements of the kindergarten group or class and the activities in which it participated. .477 .332 1/49 49. Teachers should be trained to find constructive solutions to issues they come across regarding collaboration with parents. .780 .576 2/15 28. Teachers have the necessary knowledge for advisory work with parents. .771 .585 2/14 34. Teachers are trained in holding workshops for improving parents’ pedagogical competences. .726 .589 3/11 22. Throughout formal education, teachers are sufficiently trained for collaboration with parents. .682 .445 6/8 25. Teachers are trained to offer support to parents regarding their parenting. .676 .546 2/11 31. Teachers are trained in involving parents in activities in an educational institution. .223 .662 .562 2/20 37. Teachers are trained in leading parent-teacher meetings. .620 .407 1/43 46. Teachers are trained to conduct written communication in collaboration with parents. .602 .319 .494 1/27 43. Teachers are trained in using information technology for the purpose of collaboration with parents. -.206 .542 .260 .352 2/23 42. At the onset of collaboration, parents should be informed about what is expected of them and their child. .698 .490 0.4/67 17. Collaboration is important for the growth and development of a child. .640 .434 0.4/86 18. Parents should show greater interest in establishing partnerships with preschool/school teachers. .614 .386 0.4/69 20. Parents should be informed about the kindergarten and school activities in which their child is participating. .251 .541 .442 0.4/77.5 Note. Factor loadings < .2 are suppressed; Comm – communality; F/C – floor/ceiling effect – percentage of scores in the lowest/highest category Table 2: Component loading matrix and communalities of the components (23 items) and the floor/ ceiling effect 142 Sodobna pedagogika/Journal of Contemporary Educational Studies Jurčević Lozančić, Basta, Šerbetar Internal consistency The internal consistency of the scales was investigated using Cronbach’s alpha. The coefficients for the parental involvement, teacher competences for working with parents and the expectations from collaboration subscales were .83 (10 items), .85 (9 items) and .63 (4 items), respectively (Table 3). The Cronbach’s alpha for the whole test was .867. No substantial improvement was achieved by removing any of the remaining items. Item level analysis – Floor/ceiling effects Although the scale-level analysis already provided a good understanding of the subjects’ attitudes towards teacher-parent collaboration, one very interesting aspect of the data should be reported at the item level. Namely, in several items, we observed clustering of the scores towards the high end, which is characteristic of the effect known as the ceiling effect (CE), which is characterised by a grouping of the scores at or near the upper limit (Everitt 2002). Grouping near the bottom is the opposite phenomenon, known as the floor effect (FE), which was not observed in the current data. The CE occurred in 7 items, where the percentage of the subjects who responded by choosing the highest category was between 40% and 86% (Table 2). This effect was particularly evident in items of the last subscale—expectations from collaboration: – At the onset of collaboration, parents should be informed about what is expected of them and their child (CE = 67%), – Collaboration is important for the growth and development of a child (CE = 86%), – Parents should show greater interest in establishing partnerships with preschool/ school teachers (CE = 69%), – Parents should be informed about the kindergarten and school activities in which their child is participating (CE = 76%). The range of scores in those items was 3, which means that not one of the subjects answered with 1 (“strongly disagree”). Perhaps the most conspicuous example is item 17, where none of the subjects reported 1, and only two subjects (0.4%) reported 2 (“disagree”), while 471 (86.3%) respondents “strongly agreed”. Structural validity Model χ 2 df χ 2 /df NNFI CFI RMSEA [CI] Initial 512.80** 206 2.48 0.91 0.92 .077 [0.069 - 0.084] Respecified 371.41** 196 1.89 0.94 0.95 .057 [0.048 - 0.066] **p < .01 (N = 262) Table 3: Goodness-of-fit parameters for the Teachers’ Attitudes Towards Collaboration with Parents questionnaire Teachers’ Attitudes Towards Collaboration with Parents: Development and Evaluation ... 143 A confirmatory factor analysis (CFA) using LISREL 8.80 (Jöreskog & Sorbom 2001) was administered on the three-factor model of the data. Although the as- sessment of univariate and multivariate normality was found to be significant, the maximum likelihood method of estimation was applied because in other methods of estimation a much larger sample size is required. The goodness of fit was assessed using the following: chi-square, chi-square/ df ratio (recommended to be less than 2.0), non-normed fit index (NNFI, recom- mended to be .90 or greater), comparative fit index (CFI; recommended to be .90 or greater) and root-mean-square error of approximation (RMSEA; recommended to be .05 or less). Initial model estimation indices demonstrated a fairly close fit to the data (Table 3), and all t-values were above 1.96, ranging from 5.85 to 11.17, and the R 2 ’s were above .25. However, several modification indices suggested that the observables were closely related, indicating a possible better fit of the data hence, the model was respecified by adding error covariance between the proposed items based on modification indices. New estimations yielded improved goodness-of-fit indices, and the initial chi-square value decreased, while remaining significant, and all the other indices indicated a close fit to the data. Correlations between the factors were as follows: parental involvement and teacher competences, r = .43; parental involvement and expectations from collaboration, r = .63; and teacher competences and expectations from collaboration, r = .40. Discussion The primary goal of this study was to develop and test an instrument which evaluates teachers’ attitudes towards collaboration with parents. The evaluation of the assumed factor structure of the questionnaire indicated that three distinct dimensions underlie teachers’ responses to the questionnaire. All but three factor loadings for the items were above .6, which may be considered as very good ac- cording to Comrey and Lee (1992). Although a modest amount of overall variance was explained by those factors, gathering of the items still occurred according to our previous theoretical and empirical conceptualisation which represents teach- er-parent collaboration. This approach is supported by Streiner and Norman (2008) who argued that content validation should also be based on an expert’s judgement grounded in the item contents and substance. As shown by PCA, the dimensions which underline our theoretical model of partnerships were the following: parental involvement in education, teacher com- petences for working with parents and expectations from collaboration. Internal consistency was assessed, and it was established that the first and second components were convincingly consistent, while the third component showed somewhat lower consistency . However, since the coefficient alpha is partly a function of the number of items in the scale (Cortina 1993; Streiner 2003) and the scale consisted of only four items, internal consistency may be improved in the further development of the questionnaire by adding new items to that scale (Šerbetar and Sedlar 2016). 144 Sodobna pedagogika/Journal of Contemporary Educational Studies Jurčević Lozančić, Basta, Šerbetar Several aspects of the current data need further elaboration. For example, it is very indicative that the expectations from collaboration subscale was evaluated as exceptional and that all four items on that scale showed the ceiling effect (CE). The technical explanation for the CE is that the scales may be too narrow (Terwee et al. 2007), which may impose a threat to content validity and reliability. However, we do not perceive the CE as a problem of the item wording or the insufficient range of the scale but, simply , as a profoundly positive reflection of participants’ attitudes towards the items and the overall approval of the teacher-parent partnerships. Regarding the structural validity of the scale, the CFA showed that 3 latent variables, which emerged from the exploratory part of the analysis, were well summarised by a set of 23 observed variables. All indices of fit suggested that the model closely matched the observed data. The only exception was chi-square. Nevertheless, it is known that a large chi-square usually indicates that the model estimation does not reproduce covariance of the sample and, therefore, does not fit the data well. However , chi-square has long been known to be problematic (Jöreskog 1969) due to its sensitivity to large sample sizes and also to non-normal distribu- tions of the data, which disqualify chi-square (Brown 2006) as a relevant index of fit. All the other indices of fit in the current study , and also the factor loadings and factor correlations, suggest that the three-factor scale of the questionnaire was well constructed by the items. Most of the previous work in the area of teacher-parent collaboration is theoretical, grounded in studies based on authors’ insights from different per- spectives of studying the collaboration. The current study explored the concept of teacher-parent collaboration empirically on the basis of preschool and school teachers’ self-perceptions. Three distinct dimensions emerged from the developed scale, which have been recognised in many studies, but, perhaps for the first time, they have now been quantitatively established by factor analysis. Parental involvement In general, the literature relating to areas of collaboration offers results which emphasise that the higher the parental involvement, the higher the child’s possible wellbeing (Bleach 2015; Gonzalez-Mena 2016; Hornby 2011) or the more positive outcomes there are for the child, expressed in terms of better academic and socio-emotional development (McWayne and Owsianik 2004). According to Hoover-Dempsey et al. (2005), parental involvement may be construed as home-based behaviours (e.g. helping with homework), school-based activities (e.g. attending school events) or parent-teacher communication (e.g. talking with the teacher about homework). In their previous work, they developed a model (Hoover-Dempsey and Sand- ler’s model) of parental involvement which includes (a) an active role construction for involvement (i.e. parents believe that they should be involved) and a positive sense of efficacy for helping the child learn; (b) the perception of invitations for involvement from the school, teacher and student; and (c) important elements of Teachers’ Attitudes Towards Collaboration with Parents: Development and Evaluation ... 145 parents’ life context that allow or encourage involvement (Hoover-Dempsey et al. 2005, p. 107). Although the described model was based on parents’ motivation, sev- eral points of congruency between the model and our results, for instance, parental participation in the development of the curriculum and joint decision-making or an invitation for parents to participate in school activities, all substantially loaded on our first component—parental involvement. Most of the items on the parental involvement subscale are related to the invitation for parents to become involved in different school events and activities. According to the above described model by Hoover-Dempsey et al. (2005), those invitations may be a significant contributor to more active parental beliefs about their personal role and may increase positive beliefs about the effect of one’s actions. Accordingly, studies investigating parents’ perspectives have shown that the parents evaluated more highly those forms of participation in which they were in- volved more often (Cankar et al. 2012; Cox Petersen 2011; Mandarić Vukušić 2018). Teacher competences for working with parents The second dimension which emerged from teachers’ attitudes towards col- laboration with parents was their perceived competences for handling partnerships with parents. Interestingly, those competences were perceived as the lowest. Ljubetić and Mandarić Vukušić (2012) also found that preschool teachers evaluate themselves as being inadequately competent for pedagogical work with parents. They stated that preschool teachers are simply not prepared well enough to build partnerships with parents. That finding is in accordance with several items in our scale, including teacher training statements, for whose items the par- ticipants also expressed some reservations. Epstein (1995) also noted that “most educators enter schools without an understanding of family backgrounds, concepts of caring, the framework of partnerships ... and most teachers and administrators are not prepared to understand, design, implement, or evaluate good practices of partnership with the families of their students” (pp. 92-93). Bergeron and Deslandes (2011) and Westergård (2013) confirmed that teachers’ professional and personal competences are very important for effective collaboration with parents. Most of the items from the subscale in the present research are related to the professional competences which, according to Westergård (2013), refer to the capacity of a teacher to identify students’ learning needs, arrange interventions based on didactic and pedagogic knowledge, give parents adequate instructions, etc. Equally important are personal competences, which include a positive attitude towards students, a passion for teaching and the ability to establish positive rela- tionships with parents. (Westergård 2013). However, according to Gonzalez-Mena (2016), partnerships between parents and teachers are conditioned by teachers’ skills, proficiency and beliefs gained during their education, something which is also emphasised by Gestwicki (2016). Two items in our competence scale are related to communication as an im- portant element of general teacher competences. 146 Sodobna pedagogika/Journal of Contemporary Educational Studies Jurčević Lozančić, Basta, Šerbetar Results of a study conducted in China (Xiaoyang et al. 2018) suggest that regular two-way communication between parents and teachers (and schools) and parental involvement in school activities have a significant impact on the child’s school achievement. Moreover, parents emphasised the need for more intense communication in cases of behavioural or learning problems. On the basis of the interview data for 16 parent-teacher pairs, Westergård (2013) concluded that there is a need to develop several teacher competences—rela- tional, context and communication competences. The last one was explained as the “teacher’s ability to communicate respectfully, openly, positively and reciprocally with parents” (p. 95). Competences for collaboration with parents present one segment of preschool teachers’ competences and should, and can, be developed along with all other professional and personal competences. Further professional training and development should be a necessary prerequisite for raising the present level of the quality of partnerships between families and schools. Expectations from collaboration The third and last subscale was organised around four items that reflected teachers’ beliefs about mutual expectations from collaboration, i.e. what parents should expect from teachers and what teachers expect from parents. In the first case, parents may expect to be informed about what is expected of them and their child, but also about kindergarten and school activities in which their child is par- ticipating. On the other hand, teachers expect parents to be motivated in terms of involvement and collaboration, e.g. parents should exhibit greater interest in establishing a partnership with teachers. With respect to that scale, Epstein (2011) explained the school-family partnership as family responsibility, on the one hand, and the role of the school in updating the parental involvement in school work, on the other. Findings confirming greater motivation and interest of parents were also derived in studies conducted by Ljubetić and Mandarić Vukušić (2012) and Maleš et al. (2014). Sapungan and Sapungan (2014) also stated many benefits of the voluntary involvement of parents in a child’s education, while under-motivated parents were emphasised as the most common obstacle. Conclusion Generally speaking, a partnership relationship is a harmonious relationship of those with a common task, i.e. a child’s education, and, simultaneously, a pro- cess in which parents, preschool teachers and primary school teachers have equal possibilities to learn and develop their professional and personal competences. In the current study, an instrument for evaluating teachers’ perspectives on collaboration with parents is developed and validated. The psychometric charac- teristics of the instrument are more than acceptable although some possibilities Teachers’ Attitudes Towards Collaboration with Parents: Development and Evaluation ... 147 for improvement were also observed. The proposed factor structure of the questionnaire is relatively well sum- marised by the covariation of 23 variables, which was supported by the results of confirmatory factor analysis. However, the three-factor structure may be updated in further development of the questionnaire. The internal consistency of the third scale may be enhanced by adding new items. Nonetheless, it could be beneficial to consider the new, but potentially relevant, factor that is possibly oversighted in the current study. The results of the present study and the developed instrument can enable further research on teacher-parent collaboration in order to enhance the effectiveness of the teacher-parent relationship as a very important aspect of education. Further research in the area of teacher-parent collaboration should focus on developing a congruent instrument for measuring parental attitudes towards such collaboration. Authors’ contributions: AJL, SB and IŠ have contributed to the study design and the revision of the questionnaire. AJL and SB carried out the data collection. AJL, SB and IŠ drafted the manuscript. IŠ conducted the data analysis. All authors contributed to the interpretation of the results, to the critical review and to the final approval of the manuscript. Ethical statement: All subjects gave their informed consent for inclusion in the study. Conflict of interest statement: The authors declare no conflict of interest. References Bergeron, I. and Deslandes, R. (2011). Teacher’s first meeting with groups of parents of elementary students at the beginning of the school year: parent’s expectations and responses to these expectations. International Journal about Parents in Education, 5, issue 2, pp. 43–55. Bleach, J. (2015). Supporting Parents. In: M. Reed and R. Walker (eds.). A Critical Com- panion to Early Childhood. Los Angeles, London, New Delhi, Singapore, Washington DC: SAGE, pp. 228–239. Brown, T . A. (2006). Confirmatory factor analysis for applied research. New York, NY, US: The Guilford Press. Cankar, F ., Deutsch, T. and Sentocnik, S. (2012). Approaches to building teacher-parent cooperation. CEPS Journal 2, issue 1, pp. 35–55. Comrey, A. L. and Lee, H. B. (1992). A First Course in Factor Analysis (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum. Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, issue 1, pp. 98–104. Cox Petersen, A. (2011). Educational Partnerships: Connecting Schools, Families, and the Community. Thousand Oaks, London, New Delhi, Singapore: SAGE. 148 Sodobna pedagogika/Journal of Contemporary Educational Studies Jurčević Lozančić, Basta, Šerbetar Deslandes, R., Fournier , H. and Rousseau, N. (2005). Relations of trust between parents and teachers of children in elementary school. In: R. A. Martinez Gondalez, M. H. Perez Herrero and B. Rodriguez Ruiz (eds.). Family-school-community partnerships, merging into social development. Oviedo, Spain: Grupo SM, pp. 213–232. Epstein, J. L. (1995). School/family/community partnerships: Caring for the children we share. Phi Delta Kappan, 76, issue 9, pp. 701–72. Epstein, J. L. (2001). Toward a Theory of Family-School Connections: Teacher Practices and Parent Involvement. In: J. L. Epstein (ed.). School, Family and Community Partnerships. Preparing Educators and Improving Schools. Boulder, Colorado: West- view Press, pp. 21–38. Epstein, J. L. (2011). School, family, and community partnerships: preparing educators and improving schools. Boulder, CO: Westview Press. Everitt, B. S. (2002). The Cambridge dictionary of statistics (2nd ed.). Cambridge, UK: CUP . Gestwicki, C. (2016). Home, School and Community Relations. Boston: Cengage Learning. Gonzalez Mena, J. (2016). Child, Family, and Community: Family-Centered Early Care and Education. Upper Saddle River, NJ: Pearson/Prentice Hall. Hornby , G. (2011). Parental Involvement in Childhood Education: Building Effective School Family Partnerships. New York: Springer. Hoover Dempsey, K. V ., Walker, J. M. T ., Sandler, H. M., Whetsel, D., Green, C. L., Wilkins, A. S. and Closson, K. (2005). Why do parents become involved? Research Findings and Implications. The Elementary School Journal, 106, issue 2, pp. 105–130. Jöreskog, K. G. (1969). A general approach to confirmatory maximum likelihood factor analysis. Psychometrika, 34, pp. 183–202. Jöreskog, K. G. and Sörbom, D. (2001). LISREL 8.8: User’s reference guide. Lincolnwood, IL: Scientific Software International. Jurčević Lozančić, A. (2016). Socijalne kompetencije u ranome djetinjstvu. Zagreb: Učiteljski fakultet Sveučilišta u Zagrebu. Kušević, B. (2016). Zainteresirani roditelji uvijek nađu vremena za dolazak u školu! Suradnja obitelji i škole u kontekstu reprodukcije društvenih nejednakosti. Društvena istraživanja: časopis za opća društvena pitanja, 25, issue 2, pp. 179–198. Ljubetić, M. (2014). Od suradnje do partnerstva obitelji, odgojno-obrazovne ustanove i za- jednice. Zagreb: Element. Ljubetić, M. and Mandarić Vukušić, A. (2012). Razumijevanje partnerstva i osposobljenost odgojitelja za građenje partnerskih odnosa s roditeljima. In: I. Pehlić, V . Edina and A. Hasanagić (eds.). Suvremeni tokovi u ranom odgoju. Zenica: Islamski pedagoški fakultet Univerziteta u Zenici, pp. 591–609. Maleš, D., Kušević, B. and Širanović, A. (2014). Child Participation in Family-School Coop- eration. Center for Educational Policy Studies Journal, 4, issue 1, pp. 121–136. Mandarić Vukušić, A. (2018). Professional Development of Kindergarten and Elementary School Teachers for Collaboration with Parents. Croatian Journal of Education, 20, issue 1, pp. 73–94. McWayne, C. and Owsianik, M. (2004). Parent Involvement and the social and academic competencies of urban kindergarten children in Harvard Family Research Project. Retrieved from www .hfrp.org/publications-resources/publications-series/familyinvolve- ment-research-digests/parent-involvement-and-the-social-and-academiccompetencies- of-urban-kindergarten-children (accessed on 10. 4. 2019). Teachers’ Attitudes Towards Collaboration with Parents: Development and Evaluation ... 149 Sapungan, G. and Sapungan, R. (2014). Parental Involvement in Child’s Education: Impor- tance, Barriers and Benefits. Asian Journal of Management Sciences & Education, 3, issue 2, pp. 42–48. Sheridan, S. M. and Kratochwill, T . R. (2007). Conjoint Behavioral Consultation: Promoting Family-School Connections and Interventions. London: Springer. Streiner, D. L. (2003). Starting at the beginning: An introduction to coefficient alpha and internal consistency. Journal of Personality Assessment, 80, issue 1, pp. 99–103. Streiner, D. L. and Norman, G. R. (2008). Health Measurement Scales, a practical guide to their development and use. 4th edition. Oxford: Oxford University Press. Šerbetar, I. and Sedlar, I. (2016). Assessing Reliability of a Multidimensional Scale by Coef- ficient Alpha. The Journal of Elementary Education, 9, issue 1–2, pp. 189–197. Tabachnick, B. G. and Fidell, L. S. (2007). Using Multivariate Statistics (5th ed.). New Y ork: Allyn and Bacon. Terwee, C. B., Bot, S. D. M., de Boer, M. R., van der Windt D. A. W . M., Knol, D. L., Dekker, J., Bouter, L. M. and de Vet, H. C. W . (2007). Quality criteria were proposed for mea- surement properties of health status questionnaires. Journal of Clinical Epidemiology, 60, pp. 34–42. Ward, U. (2013). Working with Parents in the Early Years. London: SAGE. Westergård, E. (2013). Teacher Competencies and Parental Cooperation. International Journal about Parents in Education, 7, issue 2, pp. 91–99. Xiaoyang, Y., Kaur Sidhu, G. and Veloo, P . K. (2018). Parents’ perspectives on school and family cooperation in primary schools: a case study in China. The Turkish Online Journal of Design, Art and Communicatio, special edition, pp. 1263–1269. Anka JURČEVIĆ LOZANČIĆ (Univerza v Zagrebu, Hrvaška) Sanja BASTA (Osnovna šola 22. junija, Hrvaška) Ivan ŠERBETAR (Univerza v Zagrebu, Hrvaška) ODNOS UČITELJEV DO SODELOVANJA S STARŠI: ZASNOVA IN EVALVACIJA VPRAŠALNIKA Povzetek: Zanimanje za raziskovanje sodelovanja med šolo in starši postaja vse večje, odkar lahko zaznavamo večje spremembe na področju angažiranosti staršev za takšno sodelovanje. Na slednje vplivajo zlasti splošne težnje po spremembi položaja in vloge družine, ob tem pa tudi težnje po vse večji vključenosti v kurikularne dejavnosti šolske institucije. Vprašalnik, ki smo ga zasnovali in preizkusili, je namenjen evalvaciji odnosa učiteljev do sodelovanja s starši. Na vzorcu 545 vzgojiteljev in učiteljev smo opravili eksploratorno faktorsko analizo. Faktorsko strukturo smo preverili s pomočjo analize glavnih komponent, ki je pokazala model, s katerim smo lahko pojasnili 42 % kumulativne variance. Izločene komponente so potrdile teoretska predvidevanja o obstoju treh dimenzij – (i) vpletenost staršev v izo- braževanje, (ii) kompetence učiteljev za delo s starši in (iii) pričakovanja, ki jih gojijo do sodelovanja s starši – s čimer je mogoče osvetliti področje partnerstva med učitelji in starši. Cronbachovi koeficienti alfa so za omenjene tri dimenzije znašali 0.883, 0.854 in 0.633. 150 Sodobna pedagogika/Journal of Contemporary Educational Studies Jurčević Lozančić, Basta, Šerbetar V drugem delu raziskave smo na novem vzorcu (N = 268) opravili konfirmatorno faktorsko analizo. Z izjemo statistično pomembnega χ 2 (371.41; p < .01) so vsi ostali indeksi prileganja pokazali ustrezno prileganje modela podatkov: χ 2 /df = 1.89, RMSEA je ob 90 % intervalu zaupanja znašal 0.057 [CI 0.048 - 0.066], NNFI je dosegel vrednost 0.94, CFI pa vrednost 0.95. Na podlagi rezultatov smo lahko oblikovali smernice za izboljšanje začetnega izobraževanja učiteljev ter pokazali na znanje in spretnosti, ki jih potrebujejo vzgojitelji in učitelji osnovnih šol za uspešno sodelovanje s starši. Ključne besede: partnerstvo med učitelji in starši, sodelovanje med učitelji in starši, vprašalnik, kompetence učiteljev E-naslov: anka.jlozancic@ufzg.hr