Trust in principals,leaders’ focus on instruction,teacher collaboration,and teacher self-efficacy: testing a multilevel mediation model

Abstract

A vast amount of contemporary research has been devoted to defining effective school leadership roles and practices and measuring their influence on school processes, structures and outcomes. This paper examines the effects of trust in principals and school leaders’ focus on instruction on teacher collaboration and teacher self-efficacy. Using teacher survey data collected from 45 lower secondary Turkish schools across the country in the 2018–2019 academic year, this study employs multilevel structural equation modelling with Bayesian estimation to investigate the structural relationships between trust in principals, alongside leaders’ focus on instruction, and teacher collaboration and teacher self-efficacy. The study confirms previous research, indicating that trust in principals has a crucial role in teachers’ emphasis on instruction and that these principals’ instructional leadership practices enhance teachers’ sense of efficacy both directly and indirectly through teacher collaboration. The research concludes that, while the way these practices are carried out might differ across contexts, the principals’ leadership practices focusing on teaching and learning still matter for teachers’ beliefs and practices in non-western countries. Implications for policy, practice and further research are discussed.

Keywords

Focus on instruction teacher collaboration teacher self-efficacy trust in principal

Introduction

A vast amount of contemporary research on educational management and leadership (EDLM) has been devoted to defining effective school leadership roles and practices and measuring their influence on school processes, structures and outcomes. The scholarly interest in understanding effective school leadership roles and practices has resulted in the development of several substantial leadership models, including, but not limited to, instructional leadership (Bellibas and Liu, 2017; Hallinger and Heck, 1996), transformational leadership (Leithwood and Jantzi, 2008; Sun and Leithwood, 2012) and distributed leadership (Heck and Hallinger, 2010; Liu et al., 2020; Ozdemir and Demircioglu, 2015). Such research over almost the past quarter-century has now confidently suggested that such school leadership occupies a central position to promote school improvement and student learning (Bryk et al., 2010; Hallinger and Heck, 1996).

Drawing upon the available literature, we know that principals’ explicit focus on instruction is a driving force for a school to achieve increased student learning outcomes (Bryk et al., 2010; Goddard et al., 2019; Özdemir and Yalçın, 2019; Robinson et al., 2008; Sebastian et al., 2017). This emphasis on improving teaching and learning also generates substantial potential to create a school environment in which teachers engage in collaborative work (Blase and Blase, 1999; Goddard et al., 2015), as well as experiencing more job satisfaction and feeling greater efficacy (Goddard et al., 2015). The current knowledge base in EDLM research, however, lacks empirical evidence regarding how effective leadership is defined, as well as whether, and to what extent, it influences the processes and outcomes of schools in the context of developing nations (Walker and Hallinger, 2015). Besides, only a little research has been conducted on the development and validation of complex multilevel models in which different variables such as trust in principal, teacher collaboration, focus on instruction, and teacher self-efficacy (TSE) is located. Hence, with the present study, we intend to center our focus on the impact of school leadership on teachers’ behaviours and emotions in Turkey, by examining the relations among some critical phenomena for school improvement, including teachers’ trust in principals, principals’ focus on instruction, and teacher collaboration and self-efficacy (e.g. Bryk et al., 2010; Goddard et al., 2015; Hallinger and Heck, 1996). More specifically, the current research aims to provide an understanding of the extent to which principals’ focus on instructional issues influences teacher collaboration and self-efficacy, as well as whether trust in principals plays a role in principals’ decisions to focus on instructional matters in Turkey.

Turkey is one of the developing countries (United Nations, 2019), and the education system consists of 12 years of compulsory education in the form of 4 (elementary) + 4 (middle) + 4 (high). Especially at the end of the second and third 4 years, students take a national exam. Their success in this exam has significant consequences for the future. Unlike those in many Anglo-Saxon countries, the Turkish education system has been profoundly centralized since the establishment of modern Turkey, and the Turkish Ministry of National Education (MoNE) sits at the top of this hierarchy. MoNE has long been responsible for the decision-making involved in all critical planning, staffing, coordination, budgeting and policy monitoring throughout the country. Despite the intention to empower local authorities and schools over the past decades, only a few policy changes have occurred to this end. For instance, in 2014, a policy shift took place that moved from the idea of national inspectors observing classrooms and providing teachers with feedback to a norm of school principals working with teachers for improving teaching and learning. Such a change in the national policy aimed to move the focus of principals from solely out-of-classroom managerial issues to involvement in the classroom through influencing core instructional practices. In this research, we are interested in examining the variation in principals’ focus on teaching, to understand better the extent to which principals influence the collaboration practices and self-efficacy of teachers. The findings of this study could contribute considerably to the growing knowledge base regarding the antecedents and outcomes of leadership in broader international contexts. It could also contribute to societies and countries with educational policies and structures similar to those of Turkey.

Conceptual framework

The conceptual framework of this study is derived from several decades of theoretical and empirical studies in the field of EDLM (see Figure 1). This body of knowledge is concerned predominantly with the antecedents (e.g., trust in principals) and outcomes (e.g., teacher collaboration, TSE, teacher learning, and student achievement) of leadership (e.g. Bellibas and Liu, 2017; Gumus et al., 2013; Klassen and Chiu, 2010; Shengnan and Hallinger, 2020; Tschannen-Moran and Hoy, 2007; Wu et al., 2020; Zheng et al., 2019). This previous work is divided into two related strands. The first line of research scrutinizes the cultural, institutional and societal contexts and their influence on various leadership roles, behaviours and practices. The second line of research has been interested in providing an understanding of how such roles, practices and behaviours could influence the school processes, culture and climate, as well as people’s behaviours, practices and outcomes. Consistent with this previous knowledge base, in the present study we centred our attention on a prominent topic: principals’ focus on instructional improvement, incorporating both its antecedents and outcomes. We first propose that teachers’ trust in principals is a prerequisite for school leaders to become productively involved in instructional matters. Second, we presume that principals’ focus on teaching would positively influence both teachers’ collaboration and their self-efficacy perceptions.

Figure 1.

Conceptual model.

The conceptual framework for this study involves seven hypotheses. The following section provides a discussion of the theoretical and empirical roots of each concept in the conceptual model and offers a basis for each hypothesis.

Teacher self-efficacy (TSE)

The concept of self-efficacy depends on Bandura’s (1977) social cognitive theory. He defines self-efficacy as individuals’ beliefs and thoughts about how well they can demonstrate the behaviours and efforts necessary to cope with the situations they face. Self-efficacy has garnered substantial attention in the educational sector and is often associated with teachers and their efforts to succeed in classroom practice. TSE refers to ‘a judgment of his or her capabilities to bring about desired outcomes of student engagement and learning, even among those students who may be difficult or unmotivated’ (Tschannen-Moran and Hoy, 2001: 783). Our definition of self-efficacy in the present study is influenced by Geijsel et al.’s (2009) work, in which the concept is characterized as the belief of teachers to effectively perform the teaching function.

Previous research has provided convincing evidence that the self-efficacy of teachers matters, both for their performance and for student learning outcomes. For example, Burić and Kim (2020) conducted research with 94 secondary school teachers and found a positive relationship between TSE and instructional quality. The difference in teachers’ perceptions of efficacy might be the main explanation for why a teacher may perform better than another with similar characteristics. The impact of self-efficacy can also influence psychosocial elements such as trust, openness, job satisfaction and professional motivation (Bümen, 2009). Moreover, the instructional methods and techniques used by teachers with strong self-efficacy beliefs are better planned, student-centered and based on humanist ideals (Goddard et al., 2004a). The study that was conducted in Jordan with both elementary and secondary school teachers offers evidence that teachers’ belief in their capability to deal with classroom problems has a significant and positive relationship with their classroom management skills (Abu-Tineh et al., 2011) and student achievement (Shahzad and Naureen, 2017). Zee and Koomen (2016) reviewed 40 years of research on TSE and found positive relationships between TSE and students’ academic achievement, their attitudes and classroom practices, their psychological well-being, job satisfaction and commitment. Some studies, conducted in Turkey, emphasized that the self-efficacy levels of elementary school teachers are higher than that of the teachers in secondary schools (Eker, 2014; Gökyer and Bakcak, 2018). In summary, strengthening teachers’ TSE perceptions would enhance their capacity to cope with problems within the school and classroom, improve the quality of their teaching, and increase their satisfaction with their profession, which could eventually improve student learning outcomes in schools.

In the present study, we are interested in the antecedents of TSE, examining multifaceted factors associated with the school organization and teacher characteristics to shed light on means through which teachers’ efficacy perceptions could be enhanced. Researchers have investigated various teacher demographic variables and found two key variables (gender and experience, respectively) that could account for teacher efficacy (Klassen and Chiu, 2010). Specifically, studies have found that female teachers have stronger self-efficacy perceptions than their male colleagues (Alwaleedi, 2017; Klassen and Chiu, 2010). Moreover, teachers’ years of teaching experience have been positively associated with their self-efficacy, suggesting increasing levels of confidence and satisfaction over early and mid-career stages (Klassen and Chiu, 2010). Based on the available literature, we expect that both teachers’ gender and job experience affect their self-efficacy (Hypotheses 1a and 1b).

Teacher collaboration

In organizational research, the concept of collaboration is based on social learning and adult learning theories. Within an educational context, these theories focus on improving the learning environment by creating opportunities for individuals to work collectively and learn from each other (Goddard et al., 2015; Wahlstrom and Louis, 2008). In education, teacher collaboration refers to teachers’ experiences in exchanging information with their colleagues, as well as their opportunities for learning from each other (Geijsel et al., 2009). Moreover, teacher collaboration is one of the most vital components of professional learning communities that foster change in teaching (Darling-Hammond, 2017) and student achievement (Goddard et al., 2007). In this study, we defined teacher collaboration based on Geijsel et al.’s (2009) study, which focused on the two key elements of exchanging ideas and engaging in reflective practice about classroom instruction. Goddard et al. (2015) emphasized that schools should have a collaborative culture for school improvement. A learning environment that makes an impact on teaching and learning should be cooperative and friendly, and promote collegial support within the school to solve problems (Wahlstrom and Louis, 2008).

Nielsen et al. (2008) emphasized the importance of three factors for supporting effective teacher collaboration: professional development embedded within school and classroom practices, clearly defined learning goals, and structures and processes that support teaching innovations. Teacher collaboration that focuses on classroom practices and student learning can promote professional learning, resulting in an improvement in instructional practices in primary schools (grades 1–8) in the Netherlands (Geijsel et al., 2009). This type of collaborative practice helps teachers transfer theories and methods of teaching to classroom practices. Better quality teaching, influenced by collegiality, also improves student learning outcomes and decreases student dropout rates and absenteeism (Chen et al., 2020). Additionally, peer mentoring, coaching and professional solidarity can help teachers in improving their problem-solving skills and generating better student learning outcomes in secondary schools (Goddard et al., 2004b). Teacher collaboration creates an opportunity for each teacher to share ideas and materials, discuss the issues they face, and receive support from their colleagues to overcome challenging circumstances. Such collegial support is essential for teacher professional learning that enables them to develop a greater sense of ability to resolve problems associated with teaching (Darling-Hammond, 2017). Several research findings support the notion that well-supported teacher collaboration can lead to increased TSE (Duyar et al., 2013; Puchner and Taylor, 2006; Tschannen-Moran and Hoy, 2007; Wu et al., 2020). Therefore, we propose that teacher collaboration will have a positive impact on TSE (Hypothesis 2).

Leaders’ focus on instruction

The idea of school principals’ focus on classroom instruction can be traced back to school effectiveness research spanning multiple decades, through which scholars concluded that schools that achieved higher student learning outcomes had principals who strived for improving teachers’ instructional practices and student achievement (Hallinger, 2011). Many studies have revealled that strong instructional leaders focus on instruction, especially by managing reading programmes and monitoring classroom processes (Blase and Blase, 1999; Bossert et al., 1982). Although some researches show that principals’ allocation of time to managerial leadership has a stronger impact on student achievement than their allocation of time to instructional leadership (e.g. Grissom and Loeb, 2011; Grissom, Loeb and Master, 2013; May, Huff and Goldring, 2012), many researchers provided evidence that instructional leadership is a driving force for student learning (Bryk et al., 2010; Hallinger, 2011; Robinson et al., 2008). Recent studies have shown that school principals frequently focus on the importance of creating high expectations for teachers, increasing student achievement, evaluating instructional practices, coordinating curriculum, emphasizing necessary skills, and monitoring students’ progress (Bellibas, 2015; Hallinger, 2011; Ozdemir, 2019). In the current study, the idea of principals’ focus on instruction is based on Supovitz, Sirinides, and May’s (2010) work that highlights principals’ endeavours to observe and evaluate classroom practices, monitor the quality of teaching, and account for school resources.

In recent years, studies have suggested a positive relationship between principals’ leadership behaviours and TSE (Leithwood and Jantzi, 2008; Zheng et al., 2019). Research demonstrates that principals’ focus on creating a vision and goals for student learning, mentoring teachers and students for success, and encouraging teacher professional development can enhance teachers’ sense of self-efficacy (e.g. Çalık et al., 2012; McGuigan and Hoy, 2006; Shengnan and Hallinger, 2020). Through this strong instructional leadership, teachers find themselves in a learning environment that nurtures their ability to deal with classroom issues and enhances their teaching practices in a way that leads to enhanced student learning outcomes (Darling-Hammond, 2017). More specifically, drawing on a multi-country dataset gathered from TALIS, Bellibas and Liu (2017) suggests that principals’ emphasis on critical instructional practices has a positive relationship with the three domains of TSE: instructional methods, classroom management and student engagement. Similarly, Duyar et al. (2013) concludes that principals’ classroom observation is directly related to well-built TSE. Based on such scholarly work, we presume that principals’ focus on instruction will have a positive impact on TSE (Hypothesis 3).

The literature suggests that principals influence classroom practices and student outcomes indirectly by creating a positive learning climate (Bossert et al., 1982; Hallinger, 2011). Their leadership practices are vital to developing a productive environment that motivates teachers to take on the responsibilities of school improvement (Ozdemir, 2019; Wahlstrom and Louis, 2008). A principal can and should establish, promote and sustain a school environment that boosts collegial interactions among the teaching staff. Research evidence shows that when principals emphasize school goals for teaching and student learning, work on instructional improvement, and foster a positive learning climate, teachers demonstrate more willingness to work together (Liu et al., 2020; Lockton, 2019). School leaders’ supervision of instruction is more likely to result in increased collaboration among primary school teachers as well (Gumus et al., 2013). Consonant with this literature, a growing number of scholars have recommended that principals should purposefully establish departmental or school-wide teacher groups and encourage these groups’ collaborative work toward the betterment of teaching practices, as well as student learning (Hallinger et al., 2019). We, therefore, propose that, when principals emphasize teaching and learning in their daily practices, teachers are more likely to collaborate for the purposes of professional learning (Hypothesis 4). Based on our assumption that principals’ focus on instruction will influence teacher collaboration that is positively related to TSE, we will also propose that there will be an indirect relationship between principals’ focus on instruction and TSE via teacher collaboration (Hypothesis 6).

Trust in principal

Trust often refers to ‘respect, personal regard, competence in core responsibilities, and personal integrity’ (Bryk and Schneider, 2002: 23). It is ‘a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behaviour of another’ (Rousseau et al., 1998: 394). Such positive expectations are comprised of several dimensions, including 'reliability, openness, competence, compassion …, or benevolence, integrity and competence' (Mishra and Mishra, 2013: 60). EDLM researchers suggest that trust between teachers and their principal is a cornerstone for all endeavours to promote school effectiveness and student achievement (e.g. Louis, 2007; Tschannen-Moran and Gareis, 2015), as trust in the principal simplifies relations, in addition to making school actors open-hearted (Berkovich, 2018). Bryk and Schneider (2002) note that relational trust levels between the principals and teachers are the heart of trust. Principals build a climate of trust through openness to communication, supporting cooperation, and providing for the professional development of the teacher (Blase and Blase, 1999; Hallinger et al., 2019). Principals should balance their task-oriented and relationship-oriented roles to create this climate (Tschannen-Moran and Gareis, 2015). In this study, the notion of trust in a principal is based on the conceptualization of Wahlstrom and Louis (2008), who define it as an atmosphere of trust where principals help teachers in instructional issues, give them individual support to enhance their teaching practices, and encourage collaborative work among them.

Teachers’ feelings and thoughts in relation to their principal are the result of emotional dynamics in the leader/follower relationship (Hallinger et al., 2019). School success increases when teachers have a high level of trust in their principals. Previous research has indicated that principals’ openness makes teachers more confident, and in the collaborative climate that such principals create, teachers can focus more on instruction (Louis, 2007). On the other hand, an atmosphere of distrust between principals and teachers blocks learning outcomes, and teachers and principals in such a climate cannot focus on instruction (Bryk and Schneider, 2002). Research indicated that trust in principals played a role of mediation in the relationship between distributed leadership and TSE of elementary school teachers (Zheng et al., 2019). Therefore, we suggest that trust in principals has a positive impact on principals’ focus on instruction (Hypothesis 5). Finally, when previous assumptions are held, we will expect that trust in principals has an indirect effect on TSE, created through principals’ focus on instruction and teacher collaboration (Hypothesis 7).

Method

In this study, we employed a cross-sectional survey design using quantitative research methods. In this section, we describe the sample and data collection procedure, introduce variables and measures, and explain the data analysis steps.

Sample

The study sample was comprised of teachers from lower secondary schools (5th–8th grades) in 12 regions in Turkey during the Spring semester of the 2018–2019 academic year. In this study, we used proportionate stratified sampling (Daniel, 2012), a useful strategy when the population involves various subgroups. Proportionate stratified sampling allowed us to determine an appropriate number of schools in each subgroup based on their number relative to the population. As an EU member, Turkey uses the Nomenclature of Territorial Units for Statistics (NUTS), a geocode standard employed in international exams such as PISA, TIMMS and PIAAC. We used this code to determine the school sample to obtain a sample that could represent the entire national population. Referring sampling showed a wide range from Turkey’s most disadvantaged regions to the most privileged areas. At the time of data collection, there were 18,745 lower secondary schools in Turkey.

Our data has a two-level hierarchical structure, consisting first of schools, and then of teachers within schools. Given this context, we used a two-stage sampling method. In the first stage, we selected the schools by using stratified random sampling via subgroups based on NUTS. The number of students in these schools ranged from 450 to 1,700. The majority of students (62%) have moderate academic success (their grade point average (GPA)is between 55 and 70). A total of 80% of students came from low- and middle-income families. We aimed to establish contact with at least 15 teachers from each school to reveal the common perception of the variables under investigation. In this study, we only focused on school principals and collected teachers’ opinions to determine the level of focus on instruction and principal trust, as well as behaviours of the school principals, collaboration practices and self-efficacy levels. We interviewed the school principals on the phone before sending our survey form to ensure data qualification. We also shared an official letter with the principals and teachers in the schools. During the data collection process, one teacher from each school was selected to distribute the scales and conduct the data collection process. They distributed scales to willing teachers to prevent principals’ bias in distributing the forms. The selected teachers filled out the scales, and the research team checked them simultaneously. Table 1 shows the distribution of the sample

Table 1.

The sampling of the study.

NUTS-1 region	All		Sampling
NUTS-1 region	Schools (N)	%	Schools (n)	%	Provinces
TR1 İstanbul<	1667	8.9	4	8.9	İstanbul
TR2 West Marmara	697	3.7	2	4.4	Balıkesir
TR3 Aegean	2269	12.1	5	11.1	İzmir-Muğla
TR4 East Marmara	1568	8.4	4	8.9	Bursa
TR5 West Anatolia	1464	7.8	3	6.7	Ankara and Karaman
TR6 Mediterranean	2497	13.3	6	13.3	Antalya-Isparta-Mersin
TR7 Middle Anatolia	1298	6.9	3	6.7	Aksaray-Niğde
TR8 West Black Sea	1259	6.7	3	6.7	Samsun-Çankırı
TR9 East Black Sea	817	4.4	2	4.4	Rize
TRA North-east Anatolia	1046	5.6	2	4.4	Elazığ-Ağrı
TRB Middle-east Anatolia	1401	7.5	4	8.9	Bitlis
TRC South-east Anatolia	2762	14.7	7	15.6	Gaziantep
Total	18,745	100	45	100

NUTS: Nomenclature of Territorial Units for Statistics.

As seen in Table 1, we collected data from 614 teachers from 45 lower secondary schools. Approximately 14 teachers from each school participated in this study. The participation rate of teachers was approximately 32%. Teachers in multilevel modelling, simulation studies show that 30 or more level-2 units are necessary to accurately estimate standard errors (SEs) (Maas and Hox, 2005). More than half of the teachers were female (60%), and 54% of them had 1–10 years of job experience. The majority of them (94%) had a bachelor’s degree, and 61% of them had worked at least 1–2 years in the same school. The distribution of core content areas was 32% Turkish, 23% math, 16% science, 12% social studies, 11% English language, and 6% other.

Variables and measures

We obtained data through four scales, including (i) TSE, (ii) teacher collaboration, (iii) focus on instruction, and (iv) teacher-principal trust. Before conducting the research, the team adapted these four measurement tools to the Turkish language and culture by using Hambleton and Patsula’s (1999) adaptation process. First, we asked three scholars who had experience in developing tests about educational administration and were familiar with both languages and cultures to translate each scale from English into Turkish. Then, the research team checked the translations and compiled them into a single form. Next, the other three scholars translated these modified instruments back into English. Finally, the research team made corrections on translation errors and the clarity of sentences that could lead to differences in meaning between the two languages.

After the translation process, we carried out a pilot study with 292 teachers from lower secondary schools. A multilevel confirmatory factor analysis (MCFA) was performed to validate each measure and examine aggregation for variables at the school level by using the intra-class correlation coefficient (ICC). Table 2 presents the fit indices and ICC results based on MCFA.

Table 2.

Fit indices and intra-class correlation coefficients (ICCs) based on multilevel confirmatory factor analysis (MCFA) results.

Measure	Fit indices								ICCs
Measure	x ²	df	x ²/df	RMSEA	CFI	TLI	SRMR_W	SRMR_B	ICC (1)	ICC (2)
TSE	3.918	3	1.306	.022	.999	.997	.008	.056	.13	.66
TC	54.306	13	4.177	.080	.956	.932	.034	.090	.06	.46
LFI	29.642	9	3.294	.061	.988	.973	.014	.016	.28	.80
TP	22.510	8	2.814	.054	.992	.980	.012	.014	.26	.83

Note: TSE: teacher self-efficacy; TC: teacher collaboration; LFI: leaders’ focus on instruction; TP: trust in principal; x ²: chi-square; df: degrees of freedom; RMSEA: root mean square error of approximation; CFI: comparative fit index; TLI: Tucker Lewis index; SRMR_W: standardized root mean square residual value for within; SRMR_B: standardized root mean square residual value for between; ICC: intra-class correlation coefficient.

The measures had acceptable fit indices if the x ²/df ratio was below 5; if the root mean square error of approximation (RMSEA), standardized root mean square residual value for within (SRMR_W) and standardized root mean square residual value for between (SRMR_B) were below .08; and if the comparative fit index (CFI) and Tucker Lewis index (TLI) were above .90 (Hu and Bentler, 1999). The intra-class correlation coefficient (ICC) (1) for each measure was above .05, indicating the suitability of aggregating teachers’ scores into the school level. For assessing the reliability of group-level means, the ICC (2) for each measure had to be between .46 and .83, demonstrating good indices (Fleiss, 1986).

The first of the four scales used for this project, the TSE scale, was created by van Woerkom (2003) in Dutch, and adapted into English by Geijsel et al. (2009). This scale measures the extent to which teachers perceive a sense of self-efficacy about their professional learning environment and development opportunities. Sample items from the scale include, ‘Do you feel that you are able to work effectively?’ ‘Are you satisfied with the quality of your work?’ and ‘Do you feel that you are being successful in your work?’ The scale was presented for this project as a 5-point Likert type (ranging from 1 = never to 5 = always), to measure the response frequency for the items.

The teacher collaboration scale was developed by Geijsel et al. (2009) to measure teachers’ experiences of professional collaboration that extend their levels of exchanging information and offer opportunities to learn from each other. Sample items from the scale include, ‘My colleagues give me positive feedback about my teaching,’ ‘My colleagues pass on to me things they have learned from further training’ and ‘My colleagues let me observe their lessons.’ The scale was prepared as a 5-point Likert scale (ranging from 1 = strongly disagree to 5 = strongly agree), to measure the response frequency of the items.

The focus on instruction scale, developed by Supovitz et al. (2010), includes five items for gauging teachers’ perceptions of principals’ expertise and focus on instruction. Sample items from the scale include ‘Knows what is going on in classrooms,’ ‘Monitors quality of instruction’ and ‘Encourages sharing of ideas.’ The scale was prepared as a 5-point Likert scale (ranging from 1 = never to 5 = always) to measure the response frequency.

The final scale used for the project, the teacher–principal trust scale, was developed by Wahlstrom and Louis (2008) to measure the trust relationship between principals and teachers. The scale consists of five items, including ‘I discuss instructional issues with my principal(s),’ ‘School’s principal(s) gives you individual support to help you improve your teaching practices,’ ‘School’s principal(s) develops an atmosphere of caring and trust’ and ‘School’s principal(s) encourages collaborative work among staff’. The scale was prepared for this project as a 5-point Likert scale (ranging from 1 = strongly disagree to 5 = strongly agree) to measure the response frequency for the items.

Data analysis

We analyzed the data according to the research hypotheses by using Mplus 8 software. There were no missing data since we collected the data using Google form. The data showed normal distribution based on the examination of the skewness and kurtosis coefficients and histogram, box-line and Q-Q graphs. The skewness and kurtosis coefficients of the groups were between –2 and +2 (George and Mallery, 2010), and the points were close to the 45-degree reference line on the Q-Q plots. In light of this information, the scores did not show a significant deviation from the normal distribution. We calculated descriptive statistics, including mean, standard deviation (SD), skewness and kurtosis. We also checked the Cronbach Alpha (α) reliability co-efficient. Pearson correlations were calculated to examine whether there was a relationship among variables.

We performed multilevel structural equation modelling (MSEM) by using Bayesian estimation with latent variables (Muthén and Muthén, 2010). The primary assumption of this analysis was that teachers’ perceptions in the same school are more homogeneous than those in different schools (Raudenbush and Bryk, 2002). MSEM is used for unbiased estimates for the relationship among variables. This method decreases bias in the estimation of direct and indirect effects and ensures accuracy in confidence interval (CI) coverage (Preacher et al., 2011). We employed the Bayesian estimation method (in Mplus) to strengthen the test of significance of effects because we selected 45 schools from 18,745 schools. Ten thousand iterations were used for this estimation, using items as observed and main constructs as latent variables. This model included TSE, gender and job experience at the individual level (Level 1). At the school level, the variables used were TSE, teacher collaboration, principals’ focus on instruction, and trust in principal. The fit indices used included chi-square, degree of freedom (df), RMSEA, SRMR, CFI, and TLI values (Hu and Bentler, 1999).

Results

Firstly, we conducted a Pearson correlation analysis. Table 3 presents the descriptive statistics and correlations among variables from this analysis

Table 3.

Descriptive statistics and Pearson correlations among variables.

Variable (within)	Mean	SD	α	S	K	TSE	JE	G
TSE	4.14	.61	.84	–1.05	1.86	1	–.025	–.090
JE	1.61	.73	–	.78	–.75		1	–.097*
G	–	–	–	–.35	–1.77			1
Variable (between)	Mean	SD	α	S	K	TSE	TC	LFI	TP
TSE	4.12	.26	.89	–.46	1.40	1	.381**	.315*	.421**
TC	3.81	.25	.88	.04	–.38		1	.371*	.349*
LFI	3.92	.54	.93	–1.10	1.60			1	.945**
TP	3.92	.56	.94	–1.11	1.51				1

Note: TSE: teachers’ self-efficacy; JE: job experience; G: gender (female is the reference group); TC: teacher collaboration; LFI: leaders’ focus on instruction; TP: trust in principal; SD: standard deviation; S: skewness; K: kurtosis; α: Cronbach alpha.

*p <.05; **p <.01.

The results showed that the skewness and kurtosis coefficients were between –2 and +2, indicating normal distribution of the variables. In terms of the internal consistency reliability of the scales, the Cronbach alphas were relatively high, ranging from .84 to .94 (Tabachnick and Fidell, 2013). Table 3 also presents a moderate and positive relationship between TSE and other school-level variables, including teacher collaboration (r = .38; p < .01), leaders’ focus on instruction (r = .32; p < .05), and trust in principal (r = .42; p < .01). This means that these variables are positively related.

Second, we performed MSEM by using Bayesian estimation to examine the direct and indirect relationship between these variables with 95% CI. First, we checked the model result, which indicated a good fit: x ² (df = 144) = 310.99; RMSEA = .043; CFI = .919; TLI = 902; SRMR_W = .008; SRMR_B = .096. Figure 2 shows the MSEM results with SEs. Table 4 presents the MSEM estimates, SEs, and CIs for the model.

Table 4.

Multilevel structural equation modelling (MSEM) estimates, standard errors (SEs) and confidence intervals (CIs) for the model.

Construct	Coefficient		95% confidence interval		P
Construct	Estimate	SE	Lower bound	Upper bound	P
Direct effects (within)
G → TSE	–.089	.045	–.179	–.005	.022
JE → TSE	–.007	.032	–.070	.057	.408
Direct effects (between)
TC → TSE	.382	.205	.063	.896	.010
PFI → TSE	.151	.088	.020	.328	.039
PFI → TC	.215	.091	.057	.416	.004
TP → LFI	.855	.078	.718	1.023	.000
Indirect effects
PFI → TC → TSE	.078	.051	.006	.201	.014
TP → LFI → TSE	.128	.076	.017	.283	.039
TP → LFI → TC	.184	.079	.048	.359	.004
TP → LFI → TC → TSE	.067	.044	.006	.174	.014
Total indirect effect
TP → TSE	.201	.075	.060	.357	.003

Note: This analysis is based on 10,000 iterations by using Bayesian estimation.

TSE: teacher self-efficacy; TC: teacher collaboration; LFI: leaders’ focus on instruction; TP: trust in principal.

Figure 2.

Model results.

At the individual teacher level, female TSE was lower than male teachers (β = –.09, p < .05, 95% CI [–.18, –.01]), confirming Hypothesis 1a. This means that, on average, male teachers have higher perceptions of their ability to deal with the problem and improve student learning. On the other hand, job experience (Hypothesis 1b) had no statistically significant effect on TSE (β = –.01, p > .05, 95% CI [–.07, .06]), implying that teachers’ sense of efficacy does not change with an increase in years of teaching experience.

Between schools, teacher collaboration has a moderate and statistically significant positive effect on TSE (β = .38, p < .01, 95% CI [.06, .90]), confirming Hypothesis 2. This finding suggests that the teacher community established through collaboration and cooperation enhances TSE. Additionally, principals’ focus on instruction (Hypothesis 3) has a small and statistically significant positive direct impact on TSE (β = .15, p < .05, 95% CI [.02, .33]), presumably because principals’ supporting instruction and providing useful feedback increases teachers’ confidence in overcoming problems in their classroom. This finding illustrates that principals’ focus on instruction has a small and statistically significant positive direct effect on teacher collaboration (β = .22, p < .01, 95% CI [.06, .42]), confirming Hypothesis 4. This suggests that principals encourage teachers to take responsibility out of the classroom to share experimental and reflective practices with their colleagues. Moreover, trust in principals (Hypothesis 5) has a large and statistically significant positive impact on principals’ focus on instruction (β = .86, p < .01, 95% CI [.72, 1.02]). This means that a successful and supportive trust atmosphere between principals and teachers encourages sharing instructional practices openly and interacting with each other honestly.

Finally, we also examined the indirect relationship between principals’ focus on instruction and TSE (Hypothesis 6). Principals’ focus on instruction had a small and indirect impact on the study sample TSE through the encouragement of teacher collaboration (β = .08, p < .05, 95% CI [.01, .20]). This means that the increase in principals’ focus on instruction leads to small improvements in teacher collaboration; such a small increase in teacher collaboration was associated with a slight improvement in teachers’ sense of self-efficacy. Additionally, we tested the indirect effect of trust on TSE. Teacher–principal trust had an indirect impact on TSE in this study, influenced through principals’ focus on instruction and teacher collaboration (Hypothesis 7). The total indirect effect is .20 (p < .01, 95% CI [.06, .36]). It suggests that the increase in teacher–principal trust contributes to substantial improvements in principal focus on instruction and teacher collaboration; such a considerable increase in principal focus on instruction together with collaboration was associated with a slight improvement in teachers’ sense of self-efficacy.

Discussion

In this section of the paper, we begin with a discussion of the limitations regarding variables and analysis for this study, followed by the interpretation of the main findings, and concluding with recommendations for policy, practice and future research.

Study limitations

Before interpreting the study’s findings, it is vital to acknowledge the limitations of the current research. First, data for this study was collected from a total of 45 schools. Although the number of schools was large enough to perform a multilevel model, it is still small considering the total number of schools in the country. Additionally, while proportionate sampling was used to ensure a sample that represented the entire national population, we believe that the generalization of these findings should still be made with caution. We conducted a mediation model using cross-sectional data, which could lead to biased estimates of the coefficients (Maxwell and Cole, 2007). To minimize these limitations, we used the Bayesian estimation method to strengthen the calculation of direct and indirect effects.

Second, this research is based on a cross-sectional survey design, meaning that we collected survey data at one point in time. Therefore, the relationships between variables should not be interpreted as causal effects. We recommend that future studies collect longitudinal data (i.e. administer surveys during at least two-time points), which enables to use the data collected during the first survey administration to control for unobserved differences among study participants at time one. Moreover, we find that both (a) teacher collaboration and (b) principals’ focus on instruction are associated with TSE. But note that it could be that TSE seems to contribute to (a) and (b) instead. Due to the cross-sectional nature of this design, the direction of these associations is unclear.

Third, we relied on teachers’ self-reported perceptions. When self-reporting, participants may provide answers to each survey item based on what their colleagues or principals want to hear, instead of what they believe themselves. It is this potential for social desirability bias that may weaken the validity of the assessment (Bandura, 1997). To avoid this bias, we distributed online surveys so that teachers could ensure confidentiality and provide more objective responses. We used MSEM to block biased estimates for the relationship among variables as well. It would be helpful in the future to collect data not only from teachers but also from principals on their perceptions of their leadership practices and trust relations with teachers (Ham, Duyar and Gumus, 2015).

Interpretation of findings

Our first hypothesis predicted a significant relationship between teachers’ gender and their self-efficacy. The research findings illustrate that, on average, female teachers have a sense of self-efficacy that is lower than that of their male colleagues. This means that male teachers possess a stronger belief in their capability to overcome classroom problems and perform well in their teaching practices. This research finding, however, contradicts the results of previous studies (Alwaleedi, 2017; Klassen and Chiu, 2010), which provided evidence that female teachers have higher self-efficacy beliefs than male teachers. The reason for this could be explained by the differences in the cultural contexts of the countries under study. Turkey is an eastern country, where a patriarchal approach is a predominant feature of society. In such societies, the primary duties of female teachers are described as a mother, housewife and spouse, while career development activities are perceived as male work (İnandı et al., 2009). Female teachers may have lower perceptions of self-efficacy because of these more rigid gender roles. Additionally, female teachers could have additional obstacles to developing themselves professionally when compared with their male counterparts. Male teachers might find more time and support for professional growth and practice, both within the school and outside of it in their home lives.

The second part of the first hypothesis predicted that job experience would have a statistically significant effect on TSE. The research results, however, suggest that years of teaching experience have no significant impact on TSE. The hypothesis concerning the relationship between teachers’ job experience and TSE was supported by previous research finding that teachers’ job experience has a significant effect on TSE (e.g. Klassen and Chiu, 2010; Penrose et al., 2007). However, our findings indicated that teachers’ length of experience did not have a significant impact on TSE in Turkey. This could be because there is no effective, well-organized professional development programme or monitoring system to observe teaching practices at the school level due to the centralized nature of the education system in the country (Bümen, 2009). Research indicates that when principals enact instructional leadership such as developing vision, monitoring teaching and providing professional learning opportunities, teachers are likely to develop increased self-efficacy perceptions (Bellibas and Liu, 2017). Whereas teachers in Turkey are not adequately supported with professional learning opportunities, resulting in only a slight variation in TSE levels across individuals with varying job experience. More individualized and school-based professional development is needed to focus on diversifying teachers’ classroom practices and improving student learning outcomes (Bümen et al., 2012). Another possible explanation for the lack of influence of experience on self-efficacy could be attributed to the particular methods of analysis employed in this study. Our analysis assumes a linear relationship between years of teaching experience and teachers’ TSE perceptions – whereas previous research, such as Klassen and Chiu’s (2010) study, has suggested a non-linear relationship between the two, providing evidence that self-efficacy increases until the late mid-career and then demonstrates a trend of decline. We believe that future studies based on non-linear assumptions could provide more reliable and valid results regarding the relationship between experience and self-efficacy for members of the teaching profession.

Our second assumption was that a positive relationship would exist between teacher collaboration and self-efficacy. The results of the analysis supported this assumption, indicating that the more teachers collaborate, the higher perceptions of self-efficacy they develop. Schools with strong supportive relationships among colleagues provide opportunities for teachers to increase their sense of self-efficacy. Our finding is supported by the previous literature (Duyar et al., 2013; Puchner and Taylor, 2006; Tschannen-Moran and Hoy, 2007). These studies claim that teachers in a supportive climate feel confident and that this climate increases their belief that they can accomplish educational tasks in the best way possible. However, Tschannen-Moran and Hoy (2007) revealed the relationship between these two variables at the individual level based on the opinions of the teachers in their studies. Examining teacher cooperation at an individual level means that their personal perception of collaboration influences their self-efficacy. By treating collaboration as a school-level variable, this research implies that collaboration arising from the collective opinion of all teachers within schools is related to individual TSE, which is a more accurate way of measuring collaboration.

The results also provide evidence to support our third hypothesis, suggesting that principals’ focus on instruction has a statistically significant positive direct impact on TSE. This indicates that school leaders can directly bolster teachers’ belief regarding their capabilities by observing their classroom and focusing on student achievement. Consistently, the literature demonstrates that teachers develop a stronger sense of efficacy when their principals emphasize, encourage and work on the improvement of teaching and learning (Bellibas and Liu, 2017; Shengnan and Hallinger, 2020). However, Wu et al. (2020) found a non-significant relationship between principal leadership and TSE in the United States by using PISA 2015 data. Within Turkey’s centralized school culture, principals have traditionally played the role of ‘great man/woman’ or ‘solo’ leader in their schools (Ozdemir and Demircioglu, 2015), like in China (Shengnan and Hallinger, 2020). Therefore, principals have the potential to use legal power on teachers’ behaviour (Sezgin and Koşar, 2010). Instructional leadership practices, emphasizing ‘solo’ leadership (Gumus et al., 2018) that facilitate the development of teacher beliefs, increase TSE, while leaders who fail to exhibit these behaviours decrease their self-efficacy (Bellibas and Liu, 2017).

Our fourth finding expresses that principals’ focus on instruction has a positive and significant direct effect on teacher collaboration. This demonstrates that principals’ focus on instruction encourages teachers to be more productive in their classes, emboldening them to take responsibility outside the classroom. This finding is parallel to those of several previous studies (Liu et al., 2020; Lockton, 2019; Ozdemir, 2019; Wahlstrom and Louis, 2008). These studies claim that principals who observed classrooms and created a supportive learning atmosphere encouraged teachers to share their reflective ideas and to set up a supportive culture and a community based on solidarity. Another possible reason to explain the positive relationship between the principal focus on instruction and teacher collaboration could be that lower secondary school principals feel pressure on rising 8th-grade students’ national exam results in Turkey (Sezgin and Koşar, 2010). Principals who are interested in increasing student achievement in high stake tests emphasize not only good quality teaching but also school-level collaboration among teachers.

The fifth hypothesis suggests that trust in principals has a significant and positive impact on principals’ focus on instruction. This means that teachers’ trust in principals is vital for principals’ focus on instruction. This finding overlaps with Bryk and Schneider’s (2002) and Louis’ (2007) studies, which provided evidence that creating a culture of trust within the school enables principals to deal with teachers’ resistance to classroom observation and feedback. However, an atmosphere of distrust between principals and teachers prevents principals from supporting teaching and learning activities in classrooms. This means that principals who are open to communication, open-hearted, supportive, cooperative and dynamic in facilitating teachers’ professional development can quickly build relational trust with teachers; thus, they can easily observe classrooms and teachers will openly accept their feedback and recommendations about classroom practices (Darling-Hammond, 2017). Prior studies have emphasized the importance of ‘relational leadership’, which refers to creating learning-oriented norms of trust, mutual respect and open discussion about difficult issues (Hallinger et al., 2019; Zheng et al., 2019). Principals can further develop their trust relationship with their teachers using these principles of accountability and transparency.

Finally, our findings provide evidence to suggest that there is an indirect relationship between principals’ focus on instruction and TSE established through teacher collaboration, as well as an indirect relationship between trust in principal and TSE established through teacher collaboration and principals’ focus on instruction. This finding supports the available literature, which indicates that principals who focused on teaching and gained the trust of their teachers would ensure teacher collaboration and TSE (Tschannen-Moran and Gareis, 2015). The relationship between teachers and principals based on trust seems to be supportive and productive both for teachers and principals. This type of partnership supports principals’ endeavors to work closely with teachers to improve teaching and learning. Supported by principals’ instructional leadership, a trust-based school environment builds a collaborative school culture that enhances teachers’ beliefs in their skills to tackle educational issues and promote student learning.

Conclusion

EDLM researchers have focused on understanding the effects of instructional leadership on school processes, structures, culture, teaching and learning for the last decades, concluding that it has an indirect effect on teaching and student learning through multifaceted means including collaboration, self-efficacy and commitment. In the present study, we examined the direct and indirect relationships between the variables of trust and instruction-focused leadership and the resulting practices of teacher collaboration and TSE. Our study supported previous research findings, highlighting that school principals who focused on teaching and gained the trust of their teachers would ensure teacher collaboration and TSE – elements that are key to teachers’ instructional practices and student learning.

Our research extends such findings beyond western countries and suggests that principals’ emphasis on building trust and improving teaching could make a difference in developing non-western countries. Research involving more countries is needed to claim an established relationship among such variables. Additionally, further research could examine whether the critical concepts investigated in this research might influence student learning outcomes to verify the relationship between leadership and student learning in a global context.

The finding of this study also supports the recent policy moves by the Turkish MoNE that emphasize and encourage more involvement by principals in teaching and learning activities. Our research indicates that more effective principals build trust and emphasize teaching to create a school climate conducive to increased collaboration, which in turn promotes the quality of teaching practices and student learning. The interaction among increased trust, emphasis on instruction, and collaboration in a school could enhance teachers’ beliefs regarding what they can do in the classroom. Principals’ instructional leadership seems critical in these processes. However, emphasizing principals’ roles as instructional leaders in policy documents alone may not provide sufficient impetus to encourage principals to focus on teaching and learning. If principals are not prepared and trained for such roles, the frequency of such supportive practices would still be rare, and their quality and basis in established theory and research could be in question. Policymakers should consider principals’ instruction-related functions in recruitment, preparation, induction and professional development as a substantial policy step to improve teacher professional capacity that is key to effective teaching and enhanced student learning.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Ömür Çoban

Nedim Özdemir

Mehmet Şükrü Bellibaş

References

Abu-Tineh

Khasawneh

Khalaileh

(2011) Teacher self-efficacy and classroom management styles in Jordanian schools. Management in Education 25(4): 175–181.

Alwaleedi

(2017) Impact of demographic variables in the development of teachers’ self-efficacy beliefs in the context of Saudi Arabia. Asian Social Science 13(1): 1–10.

Bandura

(1977) Self-efficacy: toward a unifying theory of behavioral change. Psychological review 84(2): 191–215.

Bandura

(1997) Self-Efficacy: The Exercise of Control. New York, NY: W. H. Freeman.

Bellibas

(2015) Principals’ and teachers’ perceptions of efforts by principals to improve teaching and learning in Turkish middle schools. Educational Sciences: Theory & Practice 15(6): 1471–1485.

Bellibas

Liu

(2017) Multilevel analysis of the relationship between principals’ perceived practices of instructional leadership and teachers’ self-efficacy perceptions. Journal of Educational Administration 55(1): 49–69.

Berkovich

(2018) Typology of trust relationships: profiles of teachers’ trust in principal and their implications. Teachers and Teaching 24(7): 749–767.

Blase

(1999) Principals’ instructional leadership and teacher development: teachers’ perspectives. Educational Administration Quarterly 35(3): 349–378.

Bossert

Dwyer

Rowan

, et al. (1982) The instructional management role of the principal. Educational Administration Quarterly 18(3): 34–64.

10.

Bryk

Schneider

(2002) Trust in Schools: A Core Resource for Improvement. New York: Russell Sage Foundation.

11.

Bryk

Sebring

Allensworth

, et al. (2010) Organizing Schools for Improvement: Lessons from Chicago. Chicago: University of Chicago Press.

12.

Burić

Kim

(2020) Teacher self-efficacy, instructional quality, and student motivational beliefs: an analysis using multilevel structural equation modeling. Learning and Instruction 66(2020): 1–12. DOI: 10.1016/j.learninstruc.2019.101302.

13.

Bümen

(2009) Possible effects of professional development on Turkish teachers’ self-efficacy and classroom practice. Professional Development in Education 35(2): 261–278.

14.

Bümen

Ateş

Çakar

Ural

Acar

(2012) Türkiye bağlamında öğretmenlerin mesleki gelişimi: Sorunlar ve öneriler. Milli Eğitim 194: 31–50.

15.

Chen

Elchert

Asikin-Garmager

(2020) Comparing the effects of teacher collaboration on student performance in Taiwan, Hong Kong and Singapore. A Journal of Comparative and International Education 50(4): 515–532.

16.

Çalık

Sezgin

Kavgacı

, et al. (2012) Examination of relationships between instructional leadership of school principals and self-efficacy of teachers and collective teacher efficacy. Educational Sciences: Theory & Practice 12(4): 2487–2504.

17.

Daniel

(2012) Choosing the type of probability sampling. In: Daniel

(ed) Sampling Essentials: Practical Guidelines for Making Sampling Choices. Thousand Oaks, CA: SAGE Publications, Inc.

18.

Darling-Hammond

(2017) Teacher education around the world: what can we learn from international practice? European Journal of Teacher Education 40(3): 291–309.

19.

Duyar

Gumus

Bellibas

MŞ

(2013) Multilevel analysis of teacher work attitudes: The influence of principal leadership and teacher collaboration. International Journal of Educational Management 27(7): 700–719.

20.

Eker

(2014) Sınıf öğretmenlerinin öz-yeterlilik inanç düzeyleri üzerine bir araştırm.a. Uşak Uüniversitesi Sosyal Bilimler Dergisi 7(1): 162–178.

21.

Fleiss

(1986) Reliability of measurement. In: Fleiss

(ed) The Design and Analysis of Clinical Experiments. New York: Wiley, pp.1–12.

22.

Geijsel

Sleegers

Stoel

, et al. (2009) The effect of teacher psychological, school organizational and leadership factors on teachers’ professional learning in Dutch schools. The Elementary School Journal 109(4): 406–427.

23.

George

Mallery

(2010) SPSS for Windows Step by Step. A Simple Study Guide and Reference. Boston, MA: Pearson Education.

24.

Goddard

Bailes

Nichols

(2019) From school leadership to differentiated instruction: a pathway to student learning in schools. The Elementary School Journal 120(2): 197–219.

25.

Goddard

Sook Kim

, et al. (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.

26.

Goddard

Tschannen-Moran

(2007) A theoretical and empirical investigation of teacher collaboration for school improvement and student achievement in public elementary schools. Teachers College Record 109(4): 877–896.

27.

Goddard

Hoy

(2004a) Collective efficacy beliefs: Theoretical developments, empirical evidence, and future directions. Educational Researcher 33(3): 3–13.

28.

Goddard

LoGerfo

Hoy

(2004b) High school accountability: the role of perceived collective efficacy. Educational Policy 18(3): 403–425.

29.

Gökyer

Bakcak

(2018) Ortaokul öğretmenlerinin özyeterlik düzeyleri: elaziğ ili örneği. Turkish Journal of Educational Studies 5(3): 82–98.

30.

Grissom

Loeb

(2011) Triangulating principal effectiveness: How perspectives of parents, teachers, and assistant principals identify the central importance of managerial skills. American Educational Research Journal 48(5): 1091–1123.

31.

Grissom

Loeb

Master

(2013) Effective instructional time use for school leaders: longitudinal evidence from observations of principals. Educational Researcher 42(8): 433–444.

32.

Gumus

Bellibas

Esen

, et al. (2018) A systematic review of studies on leadership models in educational research from 1980 to 2014. Educational Management Administration & Leadership 46(1): 25–48.

33.

Gumus

Bulut

Bellibas

(2013) The relationship between principal leadership and teacher collaboration in Turkish primary schools: A multilevel analysis. Education Research and Perspectives 40(1): 1–29.

34.

Hallinger

(2011) Leadership for learning: Lessons from 40 years of empirical research. Journal of Educational Administration 49(2): 125–142.

35.

Hallinger

Heck

(1996) Reassessing the principal’s role in school effectiveness: a review of empirical research, 1980-1995. Educational Administration Quarterly 32(1): 5–44.

36.

Hallinger

Liu

Piyaman

(2019) Does principal leadership make a difference in teacher professional learning? A comparative study China and Thailand. Compare: A Journal of Comparative and International Education 49(3): 341–357.

37.

Ham

Duyar

Gumus

(2015) Agreement of self-other perceptions matters: analyzing the effectiveness of principal leadership through multi-source assessment. Australian Journal of Education 59(3): 225–246.

38.

Hambleton

Patsula

(1999) Increasing the validity of adapted tests: myths to be avoided and guidelines for improving test adaptation practices. Journal of Applied Testing Technology 1(1): 1–13.

39.

Heck

Hallinger

(2010) Testing a longitudinal model of distributed leadership effects on school improvement. The Leadership Quarterly 21(5): 867–885.

40.

Bentler

(1999) Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal 6(1): 1–55.

41.

İnandı

Özkan

Peker

, et al. (2009) Kadın öğretmenlerin kariyer geliştirme engelleri [Career development barriers of women teachers]. Mersin Üniversitesi Eğitim Fakültesi Dergisi 5(1): 77–96.

42.

Klassen

Chiu

(2010) Effects on teachers’ self-efficacy and job satisfaction: Teacher gender, years of experience, and job stress. Journal of Educational Psychology 102(3): 741–756.

43.

Leithwood

Jantzi

(2008) Linking leadership to student learning: the contributions of leader efficacy. Educational Administration Quarterly 44(4): 496–528.

44.

Liu

Bellibas

MŞ

Gumus

(2020) The effect of instructional leadership and distributed leadership on teacher self-efficacy and job satisfaction: Mediating roles of supportive school culture and teacher collaboration. Educational Management Administration & Leadership. DOI: 10.1177/1741143220910438.

45.

Lockton

(2019) Chasing joint work: administrators’ efforts to structure teacher collaboration. School Leadership & Management 39(5): 496–518.

46.

Louis

(2007) Trust and improvement in schools. Journal of Educational Change 8(1): 1–24.

47.

Maas

Hox

(2005) Sufficient sample sizes for multilevel modeling. Methodology 1(3): 86–92.

48.

Maxwell

Cole

(2007) Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods 12(1): 23–44.

49.

May

Huff

Goldring

(2012) A longitudinal study of principals’ activities and student performance. School Effectiveness and School Improvement 23(4): 417–439.

50.

McGuigan

Hoy

(2006) Principal leadership: creating a culture of academic optimism to improve achievement for all students. Leadership and Policy in Schools 5(3): 203–229.

51.

Mishra

(2013) The research on trust in leadership: The need for context. Journal of Trust Research 3(1): 59–69.

52.

Muthén

(2010) Mplus user’s Guide. Los Angeles, CA: Muthén & Muthén.

53.

Nielsen

Randall

Yarker

, et al. (2008) The effects of transformational leadership on followers’ perceived work characteristics and psychological well-being: a longitudinal study. Work & Stress 22(1): 16–32.

54.

Ozdemir

Demircioglu

(2015) Distributed leadership and contract relations: evidence from Turkish high schools. Educational Management Administration & Leadership 43(6): 918–938.

55.

Ozdemir

(2019) Principal leadership and students’ achievement: Mediated pathways of professional community and teachers’ instructional practices. KEDI Journal of Educational Policy 16(1): 81–104.

56.

Özdemir

Yalçın

(2019) Examination of the relationships between academic achievement and the variables at the levels of school and students in secondary schools: two-level path analysis. Education and Science 44(200): 93–116.

57.

Penrose

Perry

Ball

(2007) Emotional intelligence and teacher self-efficacy: the contribution of teacher status and length of experience. Issues in Educational Research 17(1): 107–126.

58.

Preacher

Zhang

Zyphur

(2011) Alternative methods for assessing mediation in multilevel data: the advantages of multilevel SEM. Structural Equation Modeling 18(2): 161–182.

59.

Puchner

Taylor

(2006) Lesson study, collaboration and teacher efficacy: stories from two school-based math lesson study groups. Teaching and Teacher Education 22(7): 922–934.

60.

Raudenbush

Bryk

(2002) Hierarchical Linear Models: Applications and Data analysis Methods. Thousand Oaks, CA: Sage Publications, Inc.

61.

Robinson

VMJ

Lloyd

Rowe

(2008) The impact of leadership on student outcomes: an analysis of the differential effects of leadership types. Educational Administration Quarterly 44(5): 635–674.

62.

Rousseau

Sitkin

Burt

, et al. (1998) Not so different after all: a cross-discipline view of trust. Academy of Management Review 23(3): 393–404.

63.

Sebastian

Huang

Allensworth

(2017) Examining integrated leadership systems in high schools: connecting principal and teacher leadership to organizational processes and student outcomes. School Effectiveness and School Improvement 28(3): 463–488.

64.

Sezgin

Koşar

(2010) İlköğretim okulu müdürlerinin güç stilleri ile öğretmenlerin örgütsel bağlılığı arasındaki ilişki. Türk Eğitim Bilimleri Dergisi 8(2): 273–296.

65.

Shahzad

Naureen

(2017) Impact of teacher self-efficacy on secondary school students’ academic achievement. Journal of Education and Educational Development 4(1): 48–72.

66.

Shengnan

Hallinger

(2020) Unpacking the effects of culture on school leadership and teacher learning in China. Educational Management Administration & Leadership Epub ahead of print 21 January 2020. DOI: 10.1177/1741143219896042. 174114321989604.

67.

Sun

Leithwood

(2012) Transformational school leadership effects on student achievement. Leadership and Policy in Schools 11(4): 418–451.

68.

Supovitz

Sirinides

May

(2010) How principals and peers influence teaching and learning. Educational Administration Quarterly 46(1): 31–56.

69.

Tabachnick

Fidell

(2013) Using Multivariate Statistics. Boston: Pearson.

70.

Tschannen-Moran

Gareis

(2015) Principals, trust, and cultivating vibrant schools. Societies 5: 256–276.

71.

Tschannen-Moran

Hoy

(2001) Teacher efficacy: capturing an elusive construct. Teaching and Teacher Education 17(7): 783–805.

72.

Tschannen-Moran

Hoy

(2007) The differential antecedents of self-efficacy beliefs of novice and experienced teachers. Teaching and Teacher Education 23(6): 944–956.

73.

United Nations (2019) World Economic Situation and Prospects 2019. New York: United Nations.

74.

van Woerkom

(2003) Critical Reflection at Work: Bridging Individual and Organisational Learning. Enschede, the Netherlands: Twente University.

75.

Wahlstrom

Louis

(2008) How teachers experience principal leadership: the roles of professional community, trust, efficacy, and shared responsibility. Educational Administration Quarterly 44(4): 458–495.

76.

Walker

Hallinger

(2015) A synthesis of reviews of research on principal leadership in East Asia. Journal of Educational Administration 53(4): 554–570.

77.

Shen

Zhang

, et al. (2020) Examining the effect of principal leadership on student science achievement. International Journal of Science Education 42(6): 1017–1039.

78.

Zee

Koomen

HMY

(2016) Teacher self-efficacy and its effects on classroom processes, student academic adjustment, and teacher well-being: synthesis of 40 years of research. Review of Educational Research 86(4): 981–1015.

79.

Zheng

Yin

(2019) Exploring the relationships among instructional leadership, professional learning communities and teacher self-efficacy in China. Educational Management Administration & Leadership 47(6): 843–859.

80.

Zheng

Yin

Liu

(2019) The relationship between distributed leadership and teacher efficacy in china: the mediation of satisfaction and trust. The Asia-Pacific Education Researcher 28(6): 509–518.