For the past two decades, scholars have extensively studied teacher collaboration and collective teacher efficacy (CTE), investigating their associations with various outcomes at the teacher level (e.g., job satisfaction, teacher commitment; Klassen et al., 2010), the classroom level (e.g., instructional quality; Goddard & Kim, 2018), and the student level (e.g., reading or maths achievement; Goddard et al., 2015). Teacher collaboration involves interaction within a group with the shared goal of accomplishing a task (Vangrieken et al., 2015). This collaboration manifests in various forms of joint work, such as subject or grade-level teacher teams, co-teaching, or professional learning communities (Vangrieken et al., 2015). CTE refers to the belief individual teachers hold about the collective capability of the entire teaching faculty to make an educational difference for their students, surpassing the educational impact of homes and communities (Tschannen-Moran & Barr, 2004).

Concerning the causal relationship between teacher collaboration and CTE, previous empirical investigations followed two distinct theoretical approaches: (1) Founded in social cognitive theory, CTE can emerge from mastery and vicarious experiences gained through interactions with colleagues and observing their professional behavior (Bandura, 1997). That is, teachers who engage in collaboration are more likely to feel efficacious as a group (Moolenaar et al., 2012). (2) According to expectancy-value theory, however, the anticipation of success plays a pivotal role in predicting an individual's motivation to undertake a specific activity (Wigfield & Eccles, 2000). From this standpoint, teachers are more inclined to collaborate with their colleagues if they anticipate successful joint efforts (Authors, 2020). However, existing studies used cross-sectional data to examine the relationship between teacher collaboration and CTE (Authors, 2020) or missed to investigate bidirectional temporal associations to find evidence for causality (Goddard et al., 2015). Further, the role of principals in shaping the possible interrelations has not been taken into consideration. 

In the present study, we suggest that the relationship between teacher collaboration and CTE is likely to be both reciprocal and mutually reinforcing. We further assume that principals play a crucial role in shaping teachers' collaborative efforts because they can guide and support these processes (Honingh & Hooge, 2014). In particular, principals can foster joint work of teachers by creating the necessary conditions for collaboration at the school-level (Authors, 2023; Honingh & Hooge, 2014). That is, principals can establish teacher teams by moderating teacher conferences and allocate time slots for teachers to collaborate when scheduling teachers’ class times (Authors, 2023). Based on our assumption on the reciprocal relationship between teacher collaboration and CTE, we finally hypothesize an indirect relationship between PLP and CTE. Therefore, our study aims to explore the causal relationship between teacher collaboration, collective teacher efficacy, and principal leadership practices for collaboration. Using a random intercept cross-lagged panel model (RI-CLPM), we address the following research questions:

1. Is there a causal relationship between teacher collaboration and collective teacher efficacy?

2. What is the effect of principals’ leadership practices on teacher collaboration?

3. What is the effect of principals’ leadership practices on collective teacher efficacy?

We address our research questions by analyzing survey data obtained from a project evaluating the impact of an 18-month professional development program on enhancing principals' leadership practices for school improvement in Germany (Authors, under review). Throughout the project, we collected data from principals and teachers across three measurement occasions. This study utilizes panel data from 1112 teachers (69% female, 31% male) in 29 schools, with 619 teachers at T1, 674 at T2, and 263 at T3. Response rates averaged 60% at T1, 67% at T2, and 54% at T3. Teachers assessed their principals' leadership practices in terms of providing structures for school-based teacher collaboration (PLP, 4 items, four-point Likert scale; e.g., The principal at our school makes sure that teachers have time for collaboration), the frequency of collaboration (3 items, six-point scale; e.g., exchanging instructional materials with colleagues), and teachers' collective efficacy (CTE, 3 items, four-point Likert scale; e.g., We can make progress in our school as we are pursuing the same goals as faculty staff). Reliability estimates for all scales were satisfactory (0.67 < ω < 0.84; Nájera Catalán, 2019). Intra-class correlation coefficients indicated substantial variance between groups (0.18 < ICC(1) < 0.45), with highly reliable group means on the school level (0.89 < ICC(2) < 0.97; LeBreton & Senter, 2008). All scales exhibited strong factorial invariance over time, making them suitable for longitudinal investigations. To answer our research questions, we employed a random intercept cross-lagged panel model (RI-CLPM) using manifest variables for principals’ leadership practices (PLP), teacher collaboration, and collective teacher efficacy (CTE) across three measurement occasions. RI-CLPM is a structural equation modelling approach and allows investigations of causal relationships between variables examining cross-lagged correlations, accounting for their time-invariant, trait-like nature (Mulder & Hamaker, 2021). For RQ1 and RQ2, we examined bivariate associations between teacher collaboration and CTE, as well as PLP and teacher collaboration, respectively. RQ3 involved mediation analysis to examine the indirect relationship between PLP and CTE, mediated via teacher collaboration. The final model demonstrated excellent fit to the data (χ² = 23.650, df = 31, p > .05, RMSEA = .00, CFI = 1.00, SRMR = .04).

Regarding RQ1, our findings indicate a reciprocal causal relationship between teacher collaboration and CTE. We observed a significant effect of teacher collaboration on CTE (T1–T2: β = .18, p < .01, T2–T3: β = .18, p < .01) and vice versa, with CTE affecting teacher collaboration (T1–T2: β = .25, p < .05, T2–T3: β = .24, p < .05), while controlling for baseline estimates. In essence, teachers engaged in frequent collaboration are more likely to experience heightened collective efficacy, and conversely, those reporting high levels of collective efficacy are more inclined to report increased collaboration. As per recommendations by Orth et al. (2022), these coefficients indicate large cross-lagged effects. For RQ2, we found a large significant effect for principals’ leadership practices (PLP) on teacher collaboration (T1–T2: β = .28, p < .01, T2–T3: β = .28, p < .01) accounting for baseline PLP estimates. This implies that teachers reporting higher levels of their principal’s leadership practices, specifically in fostering school-based collaboration, are more likely to observe increases in actual teacher collaboration. With regard to RQ3, our results suggest a significant indirect effect of PLP at T1 on CTE at T3 mediated through teacher collaboration at T2, with moderate effect size (βind = .05, p < .01). Principals facilitating structures for school-based collaboration seem to positively influence teachers’ actual collaboration, subsequently enhancing their perception of collective teacher efficacy. However, there are some minor methodological limitations to our study that we need to take into account when interpreting these findings, such as participant dropout at T3. Nonetheless, our study significantly contributes to the field of school leadership and school improvement research as it uses longitudinal data to shed light on the causal relationship between teacher collaboration and CTE. Additionally, we explore the pivotal role of principals’ leadership practices in fostering collaboration among teachers, thereby enhancing their motivation.

Bandura, A. (1997). Self-efficacy: The exercise of control. Freeman. Goddard, R., Goddard, Y., Sook Kim, E., & Miller, R. (2015). A theoretical and empirical analysis of the roles of instructional leadership, teacher collaboration, and collective efficacy beliefs in support of student learning. American Journal of Education, 121(4), 501-530. https://doi.org/10.1086/681925 Goddard, Y., & Kim, M. (2018). Examining connections between teacher perceptions of collaboration, differentiated instruction, and teacher efficacy. Teachers College Record, 120(1), 1-24. https://doi.org/10.1177/016146811812000102 Honingh, M., & Hooge, E. (2013). The effect of school-leader support and participation in decision making on teacher collaboration in Dutch primary and secondary schools. Educational Management Administration & Leadership, 42(1), 75–98. https://doi.org/10.1177/1741143213499256 Klassen, R. M., Usher, E. L., & Bong, M. (2010). Teachers’ collective efficacy, job satisfaction, and job stress in cross-cultural context. Journal of Experimental Education, 78(4), 464–486. https://doi.org/10.1080/00220970903292975 LeBreton, J. M., & Senter, J. L. (2008). Answers to 20 Questions About Interrater Reliability and Interrater Agreement. Organizational Research Methods, 11(4), 815–852. https://doi.org/10.1177/1094428106296642 Mulder, J. D., & Hamaker, E. L. (2021). Three extensions of the random intercept cross-lagged panel model. Structural Equation Modeling: A Multidisciplinary Journal, 28(4), 638-648. https://doi.org/10.1080/10705511.2020.1784738 Nájera Catalán, H. E. (2019). Reliability, Population Classification and Weighting in Multidimensional Poverty Measurement: A Monte Carlo Study. Social Indicators Research, 142(3), 887–910. https://doi.org/10.1007/s11205-018-1950-z Orth, U., Meier, L. L., Bühler, J. L., Dapp, L. C., Krauss, S., Messerli, D., & Robins, R. W. (2022). Effect size guidelines for cross-lagged effects. Psychological Methods. Advance online publication. https://doi.org/10.1037/met0000499 Tschannen-Moran, M., & Barr, M. (2004). Fostering student learning: The relationship of collective teacher efficacy and student achievement. Leadership and Policy in Schools, 3(3), 189–209. https://doi.org/10.1080/15700760490503706 Vangrieken, K., Dochy, F., Raes, E., & Kyndt, E. (2015). Teacher collaboration: A systematic review. Educational Research Review, 15, 17–40. https://doi.org/10.1016/j.edurev.2015.04.002 Wigfield, A., & Eccles, J. S. (2000). Expectancy–value theory of achievement motivation. Contemporary Educational Psychology, 25(1), 68–81. https://doi.org/10.1006/ceps.1999.1015