Linking distributed leadership with collective teacher innovativeness: The mediating roles of job satisfaction and professional collaboration

Abstract

Collective teacher innovativeness has emerged as a recent topic of interest in both international policy documents and scholarly research. However, only a few studies have focused on the factors that enable collective teacher innovativeness, particularly in terms of revealing whether and to what extent school leadership might influence this construct. Existing literature suggests that distributed leadership can influence teachers’ instructional practices, emotions, and daily activities, all of which could be important to teacher innovativeness. The present study, therefore, aimed to explore the effect of distributed leadership in fostering collective teacher innovativeness, as well as the mediating roles played by job satisfaction and professional collaboration, by using the Teaching and Learning International Survey 2018 dataset. The findings of the partial mediation structural equation model analysis revealed that distributed leadership has both direct and indirect effects on collective teacher innovativeness, as mediated by job satisfaction and professional collaboration. The results provide evidence of how principal leadership practices impact teachers’ changes in behaviors through professional attitudes and practices.

Keywords

Collective teacher innovativeness distributed leadership job satisfaction professional collaboration Teaching and Learning International Survey (TALIS)

Introduction

The history of educational reforms provides ample evidence that teachers are the most important factor in the success of such initiatives, regardless of the goals, resources, and technology of the reform attempts (Cuban, 1990; Serdyukov, 2017). As the core agents of the educational process, teachers not only make significant contributions to the reform processes by gaining new knowledge, adapting to proposed changes, and utilizing new technologies but also initiate and implement change in their work environments (Fullan, 1993; van der Heijden et al., 2018). Consequently, teacher innovativeness—defined as the receptivity, openness, and willingness of teachers to embrace change (Fullan, 2016; Goldsmith, 1986; Patterson et al., 2005)—has captured the attention of both policymakers and scholars in recent years (Buske, 2018; Schwabsky et al., 2019).

Although innovativeness does not necessarily produce innovation, which refers to generating new ideas or further improvement of an existing product, process, or method (Vieluf et al., 2012), it is seen as a crucial component in creating an innovative working environment and eventually bringing about innovation (Anderson and West, 1998; Blömeke et al., 2021). Innovation requires three main steps: the formation of an idea, the implementation of the idea, and the generation of resulting outcomes (Serdyukov, 2017). These steps depend on a multitude of contextual and human factors that should be aligned with the notion of innovativeness (Fullan, 2016). Leadership emerges as an important factor in such a process since it influences the structure, roles, and relationships in the organizations.

Distributed leadership can play an especially important role in reinforcing teacher innovativeness through its influence on teachers’ instructional practices, emotions, and daily activities (O'Shea, 2021a). First, distributed leadership can enable organizational members to combine their knowledge and work together interactively by ensuring a collaborative working environment (Gronn, 2002). In addition, by creating a climate where teachers can participate in the decision-making process regardless of their status within the school, demonstrate leadership behaviors, and have more autonomy (O'Shea, 2021a), distributed leadership reinforces teacher job satisfaction (JS) (Liu and Werblow, 2019; Liu et al., 2021; Torres, 2019) and provides an environment for teachers to experience increased collaboration (O'Shea, 2021a). Such factors could be crucial in bolstering teachers’ innovativeness (Neto et al., 2017).

Drawing on the relationships among the aforementioned variables, this study empirically examines the effect of distributed leadership on collective teacher innovativeness through the mediating roles of teacher JS and professional collaboration (PC) by employing cross-national data from the latest cycle of the Teaching and Learning International Survey (TALIS). The two research questions that guided this analysis are as follows:

To what extent is distributed leadership associated with collective teacher innovativeness?

To what extent do PC and JS mediate the effect of distributed leadership on collective teacher innovativeness?

Theoretical foundation

The following section outlines the theoretical foundation of the study. First, we discuss the conceptual and operational meanings of the selected variables, to provide context for the analyses made in the study. Second, we explore the effect of distributed leadership on collective teacher innovativeness, along with the potential mediating roles that JS and PC play in this relationship.

Collective teacher innovativeness

Innovation refers to the application of novel practices, ideas, and methods to make significant improvements to an organization or system through targeted outcomes (Cohen and Ball, 2007; Gopalakrishnan and Damanpour, 1997). Researchers have highlighted the role of innovation in organizational competition and survival in a variety of social science disciplines, including in education (Thurlings et al., 2015; Westphal et al., 1997). Innovation in education emerges in areas such as new pedagogical approaches, teaching methods, and organizational structures that seek to generate significant change in educational outcomes (Serdyukov, 2017). Innovation, however, does not emerge inherently and requires certain conditions. One of the most important conditions is being open to change and new ideas, which refers to innovativeness. Teacher innovativeness, therefore, has received great attention in the relevant literature because of the pivotal role that teachers play in educational innovations (Chou et al., 2019; Kern and Graber, 2018; Thurlings et al., 2015).

In his definition of teacher innovativeness, McGeown (1980) states that the concept encompasses teachers’ adoption and internalization of innovations, as well as their constant involvement in change-related professional activities. He further asserts that the successful implementation of innovation is dependent on teachers’ competencies, perceptions, attitudes, and values. It should be noted, however, that innovation agendas generally come from the policymakers and administrators, and teachers are often expected to embrace what is ordered and mandated (McLeod and Shareski, 2018) given the rigid hierarchical structures of most school systems. Even more, teachers’ risk-taking and brave attempts on behalf of innovation may create envy among peers; this ultimately causes “a crab bucket culture,” where some teachers try to undermine individual innovation-related attempts (Margolis, 2012).

The conceptualization and measurement of innovativeness have been discussed both at the individual (Rogers, 1995) and collective level (Buske, 2018; Moolenaar et al., 2014; Schwabsky et al., 2019). At the individual level, innovativeness involves an openness to new ideas and innovations (Buske, 2018), while at the collective level it connotes a mutual climate toward pursuing innovation created by members of the organization (Nsenduluka and Shee, 2009). Blömeke et al. (2021) discussed the operationalization of innovativeness in four categories; individual attitude (Hurt et al., 1977), psychological climate characteristics (Anderson and West, 1998), organizational climate characteristics, and organizational leadership characteristics. According to the authors, individual innovativeness refers to a personal attitude towards doing something via statements such as “I am receptive to new ideas” (Hurt et al., 1977); psychological climate characteristics are identified through organizational members’ assessment of their organization through statements such as “This team is open and responsive to change” (Anderson and West, 1998), but analysis of the responses at individual (member) level; organizational climate characteristics is again measured by the statements such as “This team is open and responsive to change” to be assessed by individual members then aggregated to and analyzed on organizational level; organizational leadership characteristics can be considered as organizational leader's merits in innovativeness.

Buske (2018) conceptualizes collective teacher innovativeness as both the holistic characteristics and common awareness of teaching staff, focusing on innovations taking place in their school setting. Compared to individual innovativeness, teachers’ collective innovativeness may generate a ripple effect in educational outcomes by exceeding the sum of individual attitudes or actions (Buske, 2018); this assumption is consistent with the power of working-group dynamics, as reflected in the concepts of collective self-efficacy versus individual self-efficacy (Bandura, 1977) and organizational learning versus individual learning (Argyris and Schön, 1978). All domains of schooling can benefit from the strength of unified team efforts (Donohoo et al., 2018); hence, collective teacher innovativeness can be considered a novel concept that differs from individual teacher innovativeness. In brief, collective teacher innovativeness is more like a climate characteristic than an individual attitude, as emphasized in the literature (Ainley and Carstens, 2018; Blömeke et al., 2021).

The current study used collective teacher innovativeness as a focused concept defined by teachers’ collective endeavors to develop novel ideas in teaching and learning, openness to change, quests for new modes of problem-solving, and support of each other in the application of new ideas (OECD, 2019a).

Distributed leadership

Since the millennium, interest in distributed leadership has grown steadily within the field of educational management and leadership (Crawford, 2012; Gronn, 2002; Gümüs et al., 2018; Hartley, 2007; Hoyle and Wallace, 2005; Spillane and Healey, 2010). On the one hand, research data have accumulated to support the conclusion that distributed leadership contributes school effectiveness and improvement (Bektaş et al., 2022; Gronn, 2002; Heck and Hallinger, 2009; Leithwood et al., 2007; Spillane, 2006, 2012). Yet, on the other hand, there exists research reporting conflicting results regarding the effects of distributed leadership on student outcomes (e.g., Huang et al., 2019; Karadag, 2020; Tan, 2018).

This understanding of leadership proposes that a single leader within a rigid hierarchical structure may not successfully address the complex issues of schools (Harris, 2008; Torres, 2019); thus, the cooperation of stakeholders in the process of leading may be more effective. However, Spillane's (2006) practice-centered perspective on distributed leadership suggests a wider conceptualization of the construct, and articulates that leadership practice, encapsulating both the thinking and the activity, comes out through the interaction of the leader, their followers, and the situation itself. Further, some scholars (e.g., Crawford, 2012; Mifsud 2017a) question distributed leadership notion, and argue its genuinity and contribution to the field. These concerns also propose that distributed leadership might have emerged as a response to the growing demand for equity and inclusiveness within the schooling environments (Hartley 2007; Mifsud 2017b).

The two decades of research in this area have highlighted a few common features (Liu and Watson, 2020; Liu and Werblow, 2019). Bennett et al. (2003) outlined three distinctive characteristics of the concept of distributed leadership: (a) though traditional notes of leadership originate from individual-level properties, distributed leadership emphasizes dynamic interaction within groups; (b) while distributed leadership may broaden the traditional leadership net, the concept itself has no built-in limitations how wide its borders should be drawn; and (c) distributed leadership defends the idea that diverse types of expertise are often spread among many individuals rather than a select few in a given organization, indicating that a “concertive dynamic” would be more than the sum of the individual contributions when coordinated synergistically.

In the same vein, the distributed leadership perspective in school settings relies on the idea that school staff and stakeholders should participate in leadership activities to achieve mutual goals (Diamond and Spillane, 2016; Gronn, 2002). Accordingly, this study defined distributed leadership as the capacity of schools to maintain active participation of staff, students, and parents in the decision-making process (OECD, 2019a), underpinning its concertive or conjoint characteristic and distinct distribution of expertise and perspectives across individuals (Printy and Liu, 2021). Consistent with some recent research reporting on distributed leadership (e.g., Liu and Werblow, 2019; Sun and Xia, 2018), our study adopts this operationalization in treating distributed leadership as a feature of school capacity, supporting the joint participation of teachers, parents, and students in the school-wide decision-making processes.

Job satisfaction

JS is identified as employees’ positive or pleasurable emotional status stemming from their views of their job experiences (Locke, 1976). Most of the research on teacher JS is shaped by Herzberg et al.'s (1959) “two-factor theory,” which involves satisfying factors (motivators) pertaining to higher-order needs and dissatisfying factors (hygiene factors) related to lower-order needs (Dinham and Scott, 1998). Satisfying factors encompass intrinsic features of work such as recognition, achievement, and chances for progress, while dissatisfying factors refer to extrinsic features like policies, payments, job conditions, and interpersonal relationships (Bogler, 2001; Dinham and Scott, 2000).

Regarding the sources of JS and dissatisfaction among teachers, Dinham and Scott (1998) reported a three-domain model surrounding (a) intrinsic issues to the role of teaching, (b) factors extrinsic to the school, and (c) school-based factors. Intrinsic factors are associated with teachers’ experiences pertaining to the teaching profession, such as self-growth, working with students, and learners’ academic development. Factors extrinsic to the school refer to influences from the society, government, and education system, which can include increased expectations of schools, the rapid pace of educational change, negative images of teachers depicted in the media, and a lack of support services for teachers. Finally, school-based factors include school leadership, relationships with colleagues and parents, school climate, and school reputation. Relying on the transactional model of stress and coping (Lazarus and Folkman, 1984), O'Shea (2021b) describes school-based factors, particularly teachers’ daily interactions and relationships, as important sources of positive or negative emotions, which thereby impact JS. In a recent study conducted by Toropova et al. (2021), certain school working conditions (teacher workload, teacher cooperation, and teacher's perception of student discipline) and teacher characteristics (teacher self-efficacy, teacher's exposure to professional development, and gender of teacher) are associated with JS. Similarly, using TALIS 2013 data from 34 countries, Sims (2017) reported that working conditions like effective professional development, teacher cooperation, and school leadership are significant correlates of JS. Likewise, school climate factors, principal support, and resource adequacy are reported to positively influence JS while work pressure was negatively associated with JS (Aldridge and Fraser, 2015).

Drawing upon the definitions above, this study conceptualized JS as teachers’ satisfaction with the aspects pertaining to their profession and features related to their school environment. Thus, teachers’ profession-related satisfaction involves their overall happiness with their job, while their school-related satisfaction includes whether they would change their school if possible, their delight in working at their school, and their recommendation of the workplace as a good school (or not) (OECD, 2019a).

Professional collaboration

At its broadest, PC can be defined as the cooperative interaction of a professional group in all activities that are required to achieve a shared goal (Brouwer et al., 2012; Somech, 2005). Stemming from the notion of a “school as a learning organization” (Huber, 2008: 167), as well as the concept's shared practice orientation, PC has recently gained attention from educational scholars, who have applied it to the context of school settings (e.g., Lu and Hallinger, 2018; Meyer et al., 2022; Nguyen et al., 2021). Researchers have used a variety of similar terms to conceptualize and delineate PC as well, including professional learning communities, collegiality, teacher collaboration, teacher teams, teacher learning groups, communities of practice, and professional networks (Doppenberg et al., 2012; Levine and Marcus, 2010; Meyer et al., 2022; Vangrieken et al., 2015).

The majority of the existing conceptualizations of PC propose several shared characteristics: (a) PC entails a group of teachers working together to achieve a common school objective (Meyer et al., 2022; Nguyen and Ng, 2020); (b) collaboration is designed to generate work interdependence of teachers completing a task (Huber, 2008; Somech, 2005; Truijen et al., 2013); (c) though teachers are increasingly expected by school administrations to collaborate (Hallinger and Kulophas, 2020), genuine collaboration is founded on the willingness and co-equality of teachers (Seashore-Louis et al., 2003); (d) ongoing and long-term interactions among teachers are needed to achieve powerful collaboration in schools (Levine and Marcus, 2010); and () school tasks need to be meaningful, valuable, and worthwhile for teachers to collaborate (Somech, 2005).

In practice, a PC of teachers takes several forms in “varying interdependency degrees” (Nguyen and Ng, 2020: 640). Little (1990) delineated a relevant typology with four categories of collaboration. While the first type of collaboration, storytelling and scanning for ideas, illustrates the lowest levels of interdependency and collective autonomy in the continuum, the fourth type, joint work, represents the highest interdependence and consists of activities like evaluating and adapting a plan, and contributing to school development. Doppenberg et al. (2012) added another collaboration category to the framework, collegial support, which consists of collegial visitation and coaching activities.

This study defined PC according to the framework adopted by the Organization for Economic Co-operation and Development (OECD) for the TALIS 2018 cycle (2019a), which outlined the concept as collaborative practices with a high level of interdependency, including teaching jointly as a team in the same class, providing feedback to other teachers about their practices, engaging in shared activities across different classes, and taking part in collaborative professional learning.

Linking distributed leadership to collective teacher innovativeness

Leadership is seen as an important enabling factor for creating a school environment that supports closer teacher interaction for a shared purpose and the generation of novel ideas. Spillane (2005) claims that the distribution of leadership within a school can create synergistic interaction among principals, teachers, and situations, which may lead to a collective consciousness toward innovative behaviors among organizational members. In addition, scholars have emphasized that providing greater autonomy to teachers through a distributive perspective can then enable those teachers to employ their professional expertise, affinity, and creativity toward improving the school environment (Amels et al., 2020; Seashore-Louis and Lee, 2016) and to adopt innovative teaching practices (O'Shea, 2021a). In such a context, the interaction between leaders and followers becomes more important than individual actions. These previous studies indicate that distributed leadership practices likely create a school climate that fosters collective innovativeness among teachers.

We have proposed teachers’ JS and PC as mediators of the hypothesized relationship between distributed leadership and collective innovativeness. First, the relevant literature illustrates the impact of distributed leadership in creating an attractive and productive work environment for teachers. For example, Torres (2019) empirically associates higher levels of distributed leadership with higher levels of JS and PC among teachers, arguing that schools with greater leadership distribution are likely to foster a culture of mutual responsibility and collaboration, thus increasing teachers’ participation in school decision-making and decreasing job turnover. Supporting this finding, Liu et al. (2021) reported that distributed leadership is directly and positively associated with teacher JS. Similarly, Liu and Werblow (2019) stated that shared-decision making, which is a dimension of distributed leadership (Hairon and Goh, 2015), is positively related to teacher's JS; and O'Shea (2021b) stressed the importance of daily close interactions, as sources of positive and negative emotions, within the school in creating satisfying job environment for teachers.

Moreover, teacher JS is considered an important enabling factor of innovative school environments. The extant literature has documented the critical roles that JS and motivation play in creating productive behavior among organizational members (Jex and Britt, 2008; Spector, 1997; Yousef, 2000). JS is reported as a key predictor in building teachers’ readiness for educational change initiatives (Kondakci et al., 2016) and facilitating continuous change behaviors that promote the ongoing improvement of work practices (Kondakci et al. 2019). Some evidence also indicates that JS may increase teachers’ entrepreneurial and innovative activities (Neto et al., 2017). The respective literature strongly supports the assumption that if teachers become more satisfied with their profession and their workplace, they are more likely to manifest innovative behaviors.

Previous research also suggests a potential association between teachers’ active collaboration in professional issues and their innovative attitudes. Innovation emerges from stakeholders’ inspirational ideas and flourishes through constant creative mental activity (Brewer and Tierney, 2012) and organizational members’ eagerness to share knowledge with their colleagues (Lin, 2007). This indicates the importance of cooperative interaction for generating innovative ideas. While active collaboration motivates teachers to move beyond their own experiences and capacities and benefit from others’ ideas, low levels of teacher collaboration may discourage them from taking personal risks, leading to isolation (Goddard et al., 2007). Consequently, it is assumed that active professional engagement among teachers will likely contribute to the emergence of creative ideas, and thereby collective innovativeness.

The present study

The body of literature discussed above suggests a potential conceptual model for this study, as depicted in Figure 1. According to the literature, distributed leadership is paramount for creating shared decision-making and collective vision, as well as fostering positive attitudes toward educational change (Bush and Glover 2012; Harris, 2008; Heck and Hallinger, 2009; Kondakci et al., 2016; Spillane, 2012). However, despite its potential power for informing policy and practice, research on the association between distributed leadership and collective teacher innovativeness remains scant.

Figure 1.

Conceptual model.

The available literature also asserts that work-related factors such as JS and PC are influential in creating a productive work atmosphere and innovation-stimulating attitude (Brewer and Tierney, 2012; Jex and Britt, 2008; Lin, 2007; Lu and Campbell, 2021; Neto et al., 2017). At the same time, existing research evidence illustrates clear links between distributed leadership and both JS and PC (Kondakci et al., 2016; Liu et al., 2021; Torres, 2018). Based on the literature presented here, we suggest that distributed leadership is likely to foster collective teacher innovativeness both directly and indirectly, through the mediating effects of JS and PC.

Methodology

Data source and sample

We employed a secondary analysis of the TALIS 2018 dataset collected by the OECD between September 2017 and July 2018 (OECD, 2019a). The TALIS is a large-scale international study providing multi-country data on teacher- and school-level factors such as school leadership, teachers’ professional learning environment, school culture, instructional practices, and teaching conditions, as well as their impacts on school effectiveness (OECD, 2019b).

The TALIS 2018 cycle covers the ISCED Level 1 (primary education), Level 2 (lower secondary education), and Level 3 (upper secondary education) categories, as well as a category linking the TALIS and PISA 2018 databases called the TALIS-PISA link (OECD, 2019a). Since all countries took part and administered the instruments at the lower secondary level, the ISCED Level 2 core survey (IDPOP) was used in the current study. Choosing the Level 2 core survey allowed us to represent all countries at the same educational level. Further, as previous research (Liu and Werblow, 2019; Moolenaar and Sleegers, 2010; Printy and Liu, 2021; Schwabsky et al. 2019; Spillane and Healey, 2010) showed that leadership is contextual and leadership practice and school climate may vary based on several contextual variables such as school level, school policies, and school composition, we preferred to proceed with Level 2 core survey within the study.

Within the TALIS 2018 procedure, both teachers and their principals in the same schools were surveyed through a canonical probability cluster sampling method (OECD, 2019a). This study used the international teacher public-use file (TTGINTT3), including data from 153,681 teachers from 47 participating countries (Iceland does not present any data available for public access) focused on teacher perspectives. Both publicly and privately managed schools were included in the study, though a small number of private schools (e.g., those under a different administration, implementing a different education system, etc.) were excluded by the OECD when determining the target population (OECD, 2019a).

Variables

Dependent variable

The outcome variable of the study is collective teacher innovativeness – CTI (ω = 0.90). The TALIS 2018 defined CTI as teachers’ openness to developing innovative practices, as well as their perceptions of incentives for innovation adoption (OECD, 2019a). The OECD recently began to consider innovation as a “cross-cutting” issue for the learning environment in schools and added the theme into the TALIS 2018 cycle for the first time (OECD, 2019b). CTI is a latent variable (T3TEAM) consisting of four items (see Appendix) measured on four-point Likert scales, asking teachers to rate their responses between “strongly disagree” (1) and “strongly agree” (4).

Mediating variables

The mediating variables in this study were teacher JS and PC. The TALIS 2018 defined JS using three subscales: JS with the work environment (T3JSENV), JS with the profession (T3JSPRO), and satisfaction with target class autonomy (T3SATAT). We used the JS with work environment subscale T3JSENV (ω = 0.79) in the present study, as we focused on teachers’ satisfaction with their current work environment. This subscale was based on four items (see Appendix) measured on a four-point Likert scale, varying between “strongly disagree” (1) and “strongly agree” (4). One item (TT3G53C, “I would like to change to another school if that were possible”) was reverse coded.

PC between teachers was measured using the “professional collaboration in lessons among teachers” subscale (T3COLES) of the teacher co-operation scale (T3COOP). The PC subscale included four items (see Appendix) with response options measured on a six-point answer format from “never” (1) to “once a week or more” (6). The OECD calculated ω values for the T3COLES subscale across ISCED 2 Level participant countries ranged from 0.54 (Slovenia) to 0.77 (United Arab Emirates), while the overall ω coefficient for the current study was 0.63.

Independent variable

The independent variable for our conceptual model was teacher-reported distributed leadership (DL). The TALIS teacher questionnaire includes several items relevant to teachers’ perceptions of distributed leadership. We used three items from the “participation among stakeholders” subscale (T3STAKE) (ω = 0.83), asking teachers whether their school provides staff (TT3G48A), parents or guardians (TT3G48B), and students (TT3G48C) with opportunities to actively participate in school decisions. All items were measured using a four-point Likert-type scale, with response options ranging from “strongly disagree” (1) to “strongly agree” (4).

Control variables

We also included control variables to avoid the potential presence of confounding bias. Teacher gender (TT3G01, binary), total years of teaching experience (TT3G11B, metric), level of education (TT3G03, nominal), and country (IDCNTRY, nominal) were controlled for their probable effects on the outcome variable of the model. Previous research reports mixed results about the effect of gender on teacher innovativeness (e.g., Carmeli et al., 2006; Nguyen et al., 2021). Besides teacher gender, some research demonstrates that as teachers become more experienced, they tend to exhibit less innovativeness (Thurlings et al., 2015). Consistent with previous research (Yang and Huang, 2008), the educational level of teachers was also controlled for within the model. Given that each country has distinct political, social, and cultural features that may influence teacher innovativeness, the country variable was controlled to address its variation on the dependent variable, as well.

We created dummy variables for teachers’ level of education to include in the analysis by using “ISCED Level 6: bachelor's or equivalent” as the reference group (the other two groups were “ISCED 2011 Level 0–5: pre-tertiary or short-cycle tertiary education” and “ISCED 2011 Level 7–8: graduate education or equivalent,” respectively). The participating country variable was also treated as a control variable when creating dummy variables to represent subgroups within the structural model.

Analytical approach

We used single-level scales, subscales, and/or items derived from the TALIS 2018 within the study. The dataset was prepared utilizing IBM SPSS version 23 and analyzed in Mplus version 8.5 (Muthén and Muthén, 2017) by maximum likelihood, the method of choice for estimating continuous outcome variables (Maydeu-Olivares, 2017). A partial mediation structural equation model (SEM) was performed to identify the hypothesized direct and indirect effects of distributed leadership on collective teacher innovativeness. Before the main analysis, descriptive and preliminary statistical analyses were conducted to check the assumptions underlying multivariate analysis (Tabachnick and Fidell, 2012).

As the number of missing values for the study variables varied between 2.7% (TT3G33A) and 5.2% (TT3G03), full information maximum likelihood (FIML) estimation was run while accommodating the missing data (Enders, 2001). Standardized z-scores were calculated to determine the univariate outliers of each variable and 1170 cases exceeding the ± 3.29 range were identified and removed. To find multivariate outliers, the Mahalanobis distance values were computed at p < 0.001 level and compared against the respective critical values of the χ² values. 371 cases with exceeding critical χ² values were detected and excluded, as suggested by Leys et al. (2019). The skewness and kurtosis values were then checked to address univariate normality. All modeled variables met the univariate normality assumption, ranging from −0.380 to 0.594 for skewness and −0.030 to 0.879 for kurtosis (e.g., Chou and Bentler, 1995). When the bivariate correlations between predictor variables were checked, no possible issues of multicollinearity existed within the dataset (see Table 1), as all values were below the generally accepted threshold level of 0.80 (Grewal et al., 2004).

Table 1.

Means, standard deviations, and bivariate correlations of study variables.

Variable	Mean	SD	CTI	JS	PC	DL
1. Collective teacher innovativeness	2.97^a	0.610	(0.90)
2. Job satisfaction	3.12^a	0.553	0.307^b	(0.79)
3. Professional collaboration	2.83^c	1.089	0.273^b	0.124^b	(0.63)
4. Distributed leadership	2.86^a	0.583	0.359^b	0.371^b	0.219^b	(0.83)

Notes. Corresponding McDonald's omega reliability coefficients in parenthesis (Dunn et al., 2014).

4-point Likert-type scale.

All correlations are statistically significant at 0.01 level.

6-point Likert-type scale.

Since the current study relied on a single data source, we conducted Harman's (1960) single factor test to identify common method variance (e.g., Podsakoff et al., 2003). Common method variance is specified by the presence of a single factor accounting for the majority of shared variance among study variables. If this covariance ranges between >10% and <50%, it can be ignored, as covariance at this level typically causes no material biases or further distortion in the analysis (Lance et al., 2010). The test yielded a 31.04% variance for unrotated one-factor solution in the present study, indicating no serious method bias in the dataset.

Though the coefficient α is the most popular measure of choice for estimating reliability, it computes the lower bound estimations of the reliability (Dunn et al., 2014). Accordingly, as the McDonald's omega coefficient (ω) relies on a more sensible measure of reliability, we reported ω for the scale reliability. Finally, a mediated SEM using the maximum likelihood estimation method was conducted to test the structural relationships among the study variables. The direct and indirect effects of teacher-reported distributed leadership on collective teacher innovativeness were tested while controlling for teachers’ gender, total years of experience, level of education, and country. We calculated and reported the standardized regression coefficients and standard errors to present comparable effect sizes (Nezlek, 2011), indicating the proportion of variance explained and the relative importance of the variables (Kline, 2020).

Descriptive statistics and preliminary results

The means, standard deviations, bivariate correlations, and reliability coefficients of the study variables are summarized in Table 1 below.

As presented in Table 1, mean values varied from 2.83 to 3.12, while standard deviations were between a moderate (0.55) to high (1.09) range. Bivariate correlations revealed that all study variables were statistically significant and positively correlated to each other at the 0.01 level with varying strengths.

After the preliminary analyses, a measurement model was tested as a primary step of the conceptual SEM model. We evaluated the model fit using the indices generated by Mplus, including root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR), and comparative fit index (CFI). RMSEA and SRMR values smaller than 0.08, and CFI values larger than 0.90 are broadly considered acceptable indices (e.g., Browne and Cudeck, 1993; Marsh et al., 2004). We reported yet ignored the χ² values since it simply rejects models with large sample sizes (Hu and Bentler, 1999). Accordingly, the measurement model fit the data well (χ² = 21,3800.559 [df = 84], p < 0.001, CFI = 0.98, RMSEA = 0.04, SRMR = 0.03). The following section presents the direct, indirect, and total effects of the structural model.

Results

To test the proposed direct and indirect effects of distributed leadership on teacher collective innovation, a mediated SEM was developed. This model incorporated teacher-reported distributed leadership as the independent variable, JS and PC as mediating variables, and collective teacher innovativeness as the outcome variable.

The conceptual model showed a good fit for the data (χ² = 67,210.350 [df = 415], p < 0.001, CFI = 0.88, RMSEA = 0.03, SRMR = 0.03). We then examined the paths and standardized coefficients among the modeled variables. Kline's (2005) guideline on path coefficients’ effect size interprets values of ≥0.10, ≥0.30, and ≥0.50 as small, medium, and large effects, respectively. Table 2 displays the standardized regression coefficients and corresponding standard errors for each path.

Table 2.

Results for the standardized direct, indirect, and total effects of distributed leadership on collective teacher innovativeness through JS and professional collaboration.

	β	SE	p
Standardized direct effects
DL → CTI	0.286	0.003	0.000
DL → JS	0.397	0.003	0.000
DL → PC	0.228	0.003	0.000
PC → CTI	0.147	0.003	0.000
JS → CTI	0.196	0.003	0.000
JS → PC	0.040	0.003	0.000
Standardized total indirect effects
DL → JS → CTI	0.078	0.001	0.000
DL → PC → CTI	0.034	0.001	0.000
DL → JS → PC → CTI	0.002	0.000	0.000
Standardized total effects
DL → CTI	0.399	0.003	0.000

Abbreviations. DL: distributed leadership; JS: job satisfaction; PC: professional collaboration; CTI: collective teacher innovativeness; SE: standard error.

Notes. The model was controlled for by teacher gender, total years of experience, level of education, and country.

β = standardized regression coefficients. All paths are statistically significant at 0.001 level.

As shown in Table 2, distributed leadership had a moderate and significant relationship with collective teacher innovativeness (β = 0.286, SE = 0.003, p < 0.001). This indicated that school principals’ distributed leadership behaviors, as reported by individual teachers, impacted teachers’ perceptions of innovative practices in their schools. The results also revealed that distributed leadership was a strong predictor of JS (β = 0.397, SE = 0.003, p < 0.001), implying that teachers felt more satisfied in schools where principals demonstrated greater distributed leadership behaviors. Likewise, distributed leadership was positively linked with teachers’ PC at a moderate level (β = 0.228, SE = 0.003, p < 0.001), indicating that distributed leadership as reported by teachers influenced their PC practices, such as providing feedback to other teachers, teaching jointly, and engaging in learning activities together.

PC among teachers had a weak yet significant relationship with collective teacher innovativeness (β = 0.147, SE = 0.003, p < 0.001), suggesting that teacher collaboration influenced their perceptions of collective innovativeness in their professional environment. Likewise, JS was a weak predictor of the outcome variable of this study, collective teacher innovativeness (β = 0.196, SE = 0.003, p < 0.001), indicating that teachers with higher levels of JS tended to perceive a more innovative environment in their schools. Furthermore, JS was found to have a weak yet significant relationship with PC (β = 0.040, SE = 0.003, p < 0.001). This means that teachers who were satisfied with their work were more likely to perceive higher levels of PC.

Next, the standardized indirect effects within the structural model were examined (see Figure 2). Distributed leadership was found to positively predict collective teacher innovativeness through JS (β = 0.078, SE = 0.001, p < 0.001). Though the mediation effect was small, this result implies that teachers who rated more distributed leadership behaviors and practices in their schools also felt more satisfied in their work, which in turn increased their perceptions of innovation adoption among colleagues. Similarly, PC had a weak yet significant mediator role in the relationship between distributed leadership and collective teacher innovativeness (β = 0.034, SE = 0.001, p < 0.001). Analysis of indirect paths revealed that distributed leadership practices in school settings positively influenced teachers’ perception of PC and that this influence might increase their perceptions of openness to innovative practices among their colleagues. Moreover, a significant association was found between distributed leadership and collective teacher innovativeness through the path of JS and PC, though with a very small effect size (β = 0.002, SE = 0.001, p < 0.001). Overall, the mediating variables of the present study, JS and PC, accounted for 11.4% of the total variation in collective teacher innovativeness.

Figure 2.

Structural equation model results.

We then examined the control variables of the structural model. The results showed that female teachers had a higher level of perception of team innovativeness (β = −0.035, SE = 0.003, p < 0.001), implying that male teachers are more likely to perceive the degree of collective teacher innovativeness in their current schools lower than their female colleagues. Total years of teaching experience were significantly and positively related to collective teacher innovativeness as well (β = 0.057, SE = 0.003, p < 0.001). This meant that as teachers gained more experience, they were likely to perceive their working environment as more innovative. Results also denoted that teachers with graduate degrees perceived their schools as less innovative than their colleagues who had only achieved a bachelor's degree (β = −0.023, SE = 0.003, p < 0.001), while those with a pre-tertiary degree perceived their schools as more innovative than their colleagues with a bachelor's degree (β = 0.014, SE = 0.003, p < 0.001).

In brief, the SEM analysis revealed that the total effects of teacher-reported distributed leadership on collective teacher innovativeness were moderate, statistically significant, and positive (β = 0.399, SE = 0.003, p < 0.001). Comprised of weak to moderate effect sizes (i.e., β = 0.002 to β = 0.397), the partial mediation model hereby explained 40% of the total variation in collective teacher innovativeness and provided further evidence of the relationship between leadership practices and innovative learning environments in schools from teachers’ perspectives.

Discussion and conclusions

The current study examines the role of distributed leadership in enhancing collective teacher innovativeness using an international secondary data set from the TALIS 2018, compiled by the OECD. The mediating roles of two frequently studied teacher variables, JS and PC, are also investigated in the assumed relationship. The results of our analysis highlight a few key findings. First, distributed leadership has a moderate and significant direct association with collective teacher innovativeness. Second, both mediating variables, JS and PC, have weak yet significant relationships with collective teacher innovativeness. Last, both teacher JS and PC play a significant yet weak mediating role in the relationship between distributed leadership and collective teacher innovativeness. Before interpreting these findings and discussing their possible implications, we wish to mention the limitations of this study and offer suggestions for future research.

Limitations and future research

First, the measures of variables in this study have certain shortcomings. The constructs developed in this study are based on the OECD's TALIS 2018, which does not always include enough items to represent a broader understanding of the relevant constructs. For example, the items used in this study for the distributed leadership construct only represent the participative decision-making aspect. While the active participation of stakeholders in school decisions is critical for distributed leadership and reflects some understandings of distributed leadership, there are additional elements of the construct that should be considered (Mayrowetz, 2008; Printy and Liu, 2021). It is also important to note that all measures used in this study, including the items regarding the participation of parents and students in school decisions, are based on the teacher questionnaire. Therefore, triangulation of data was not possible. We encourage OECD to consider adding parent and student surveys to future TALIS cycles to provide a more comprehensive understanding of organizational and classroom level aspects of schools. Lastly, the cross-sectional nature of the analysis limits the possibility of establishing a causal relationship among the studied variables. We, therefore, suggest longitudinal and experimental studies to confirm the associations among the variables found in the study. Moreover, in future studies, using multilevel models focusing on teachers’ and principals’ perspectives may provide different insights. Last, the field will benefit from additional qualitative investigations of the role of school leadership in creating and sustaining teacher innovation and change.

Interpretations and implications

Given the available evidence, our results offer important insights toward fostering a school environment conducive to collective teacher innovativeness. Overall, the results support the idea that distributed leadership might play a role in promoting innovative teacher behavior in schools (Gronn, 2002). This finding is consistent with the extant literature defining collective teacher innovativeness as the unified efforts of organizational members towards a shared vision in terms of innovation (Buske, 2018; Nguyen et al., 2021), which implies that school leadership that demonstrates more distributive characteristics is likely to support the interaction of teachers (Spillane, 2005) and thereby enhance creative ideas and an innovative atmosphere among teachers. This result may also indicate that the essence of teacher innovation might lie in “innovation process variables” such as work environment characteristics, manager/supervisor support for innovation, and participative safety, as suggested in an early meta-analysis study by Hülsheger et al. (2009).

The direct effect of distributed leadership in this study could be related to both trust and openness to share ideas/information, which are critical factors promoting innovation in any working environment, including schools (Clegg et al., 2002; Dovey, 2009; Moolenaar and Sleegers, 2010; Schwabsky et al., 2019). The leadership literature has provided evidence that distributed leadership can strengthen trust among employees and enable openness and willingness to generate, exchange, share, and try new ideas, information, and solutions (Adıgüzelli, 2016; Bektaş et al., 2022; Smylie et al., 2007). More specifically, our conceptualization of distributed leadership, which is based on participatory decision-making, could be closely related to the creation of an environment in which teachers share ideas and information on various subjects with the confidence that they will be heard. These environments facilitate the emergence of new and innovative ideas (Hülsheger et al., 2009). Such a work environment usually encourages sharing and disseminating ideas and information, willingness to take risks, open communication, discussion of opposing viewpoints, allowing questions, and feeling heard and understood, all of which are essential components of innovation and creativity in school settings (Somech and Naamneh, 2019). Moreover, participative decision-making practices enable the generation of innovative ideas, which can then be converted into coordinated team action—that is, teachers first interpret the process behind the innovation and then transfer it into their teaching, pedagogical work, procedures, and projects. We, therefore, argue that distributed leadership could form a functional policy tool, particularly in reform environments where innovation and change are desirable.

About the significant relationship between collective teacher innovativeness and this study's mediating variables, JS, and PC, the previous literature affirms the impact of cooperative interaction of a professional group on creating a shared task within a workplace (Brouwer et al., 2012; Somech, 2005; Truijen et al., 2013) and enhancing knowledge-sharing climate and thereby organizational innovation (Lin, 2007). In a similar vein, greater motivation and JS function as critical enabling factors for generating innovation-oriented behavior among organizational members (Jex and Britt, 2008; Spector, 1997; Yousef, 2000), as well as in supporting continuous change behaviors of teachers (Kondakci et al., 2019).

Though their impacts were relatively small in size, JS and PC were found to be significant mediators of the relationship between distributed leadership and collective teacher innovativeness. Based on the theoretical connections as well as the existing literature, which confirms the close relationships of these constructs with the dependent and independent variables (Liu and Watson, 2020; Liu and Werblow, 2019; Liu et al., 2021; Nguyen et al., 2021; Somech and Naamneh, 2019; Torres, 2019), we expected such findings, though the magnitude of their mediating role was smaller than we anticipated. The small magnitude could be attributed to the nature of JS as well as PC fostered by distributed leadership practices. Participating in decision-making processes could increase the level of collaboration for professional purposes among teachers, though such collaboration activities might focus more on existing practices rather than the emergence of new ideas and innovations in certain contexts. Similarly, teacher JS might even deteriorate the willingness to innovate in certain contexts or for certain individuals if it provides a high-level appreciation for existing structures and practices. In short, while both mediating variables were found to be related to collective teacher innovativeness, their existence may not always guarantee higher levels of innovation. Thus, future studies should investigate the moderating roles that various organizational and individual characteristics play in this relationship.

Conclusions

Overall, our model was able to explain 40% of the total variation in collective teacher innovation. These findings provide strong evidence regarding the positive relationship between distributed leadership and collective teacher innovativeness, indicating that leadership practices influence the tendency and capability of teachers to explore and implement new learning to take collective action in educational processes. However, additional important contextual factors could explain the remaining percentage of collective teacher innovativeness. Some of these potentially critical factors, such as trust, commitment, external support/demands, etc., might also operate as mediators in the relationship between school leadership and teacher innovativeness. Furthermore, it is also important to investigate the associations between various other leadership practices and models—such as transformational, instructional, and teacher leadership—and teacher innovativeness. We, therefore, suggest that future research employ a variety of leadership models, alongside additional contextual factors, to better explain the mechanisms that link school leadership with collective teacher innovativeness.

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

Hilal Buyukgoze

Omer Caliskan

Sedat Gümüş

Author biographies

Hilal Buyukgoze is a research assistant in the Department of Educational Sciences at Hacettepe University, Turkey. She received her PhD in Educational Administration from Hacettepe University in 2021. Her research interests include school improvement, educational leadership, and higher education.

Omer Caliskan is currently Alexander von Humboldt fellow at the International Center for Higher Education Research (INCHER) at the University of Kassel, Germany. He is an associate professor of educational administration at Yozgat Bozok University, Turkey. He received his PhD in Educational Administration and Planning from Middle East Technical University in 2017. His research interests include education policy, educational leadership, and higher education.

Sedat Gümüş is an associate professor in the Department of Educational Policy and Leadership at the Education University of Hong Kong, Hong Kong. His current research focuses on the relationship between school leadership and various teacher and student outcomes as well as the contextualization of leadership models and practices.

References

Adıgüzelli

(2016) Examining the relationship between distributed leadership and organizational trust according to opinions of teachers. Egitim ve Bilim 41: 269–280.

Ainley

Carstens

(2018) Teaching and Learning International Survey (TALIS) 2018 Conceptual Framework (OECD Education Working Papers, 187). OECD Publishing.

Aldridge

Fraser

(2015) Teachers’ views of their school climate and its relationship with teacher self-efficacy and job satisfaction. Learning Environments Research 19(2): 291–307.

Amels

Krüger

Suhre

, et al. (2020) The effects of distributed leadership and inquiry-based work on primary teachers’ capacity to change: Testing a model. School Effectiveness and School Improvement 31(3): 468–485.

Anderson

West

(1998) Measuring climate for work group innovation: Development and validation of the team climate inventory. Journal of Organizational Behavior 19(3): 235–258.

Argyris

Schön

(1978) Organizational Learning: A Theory of Action-Perspective. Reading, MA: Addison-Wesley.

Bandura

(1977) Self-efficacy: Toward a unifying theory of behavioural change. Psychological Review 84(2): 191–215.

Bektaş

Kılınç

AÇ

and Gümüş S (2022) The effects of distributed leadership on teacher professional learning: mediating roles of teacher trust in principal and teacher motivation. Educational studies 48(5): 602–624.

Bennett

Wise

Woods

, et al. (2003) Distributed Leadership: A Review of Literature. Nottingham, UK: National College for School Leadership: The Open University. Retrieved from http://oro.open.ac.uk/8534/1/bennett-distributed-leadership-full.pdf.

10.

Blömeke

Nilsen

Scherer

(2021) School innovativeness is associated with enhanced teacher collaboration, innovative classroom practices, and job satisfaction. Journal of Educational Psychology 113(8): 1645–1667.

11.

Bogler

(2001) The influence of leadership style on teacher job satisfaction. Educational Administration Quarterly 37(5): 662–683.

12.

Brewer

Tierney

(2012) Barriers to innovation in the US education. In: Wildavsky

Kelly

Carey

(eds) Reinventing Higher Education: The Promise of Innovation. Cambridge, MA: Harvard Education Press, pp.11–40.

13.

Brouwer

Brekelmas

Nieuwenhuis

, et al. (2012) Community development in the school workplace. International Journal of Educational Management 26(4): 403–418.

14.

Browne

Cudeck

(1993) Alternative ways of assessing model fit. In: Bollen

Long

(eds) Testing Structural Equations Models. Newbury Park, CA: Sage, pp.136–162.

15.

Bush

Glover

(2012) Distributed leadership in action: Leading high-performing leadership teams in English schools. School Leadership and Management 32(1): 21–36.

16.

Buske

(2018) The principal as a key actor in promoting teachers’ innovativeness: Analyzing the innovativeness of teaching staff with variance-based partial least square modeling. School Effectiveness and School Improvement 29(2): 262–284.

17.

Carmeli

Meitar

Weisberg

(2006) Self-leadership skills and innovative behavior at work. International Journal of Manpower 27(1): 75–90.

18.

Chou

Bentler

(1995) Estimates and tests in structural equation modeling. In: Boyle

(ed) Structural Equation Modeling: Concepts, Issues and Applications. Thousand Oaks, CA: Sage, pp.37–56.

19.

Chou

Shen

Hsiao

, et al. (2019) Factors influencing teachers’ innovative teaching behaviour with information and communication technology (ICT): The mediator role of organisational innovation climate. Educational Psychology 39(1): 65–85.

20.

Clegg

Unsworth

Epitropaki

, et al. (2002) Implicating trust in the innovation process. Journal of Occupational and Organizational Psychology 75(4): 409–422.

21.

Cohen

Ball

(2007) Educational innovation and the problem of scale. In: Schneider

McDonald

(eds) Scale Up in Education: Ideas in Principle. UK: Rowman and Littlefield Publishers, pp.19–36.

22.

Crawford

(2012) Solo and distributed leadership: Definitions and dilemmas. Educational Management Administration and Leadership 40(5): 610–620.

23.

Cuban

(1990) Reforming again, again, and again. Educational Researcher 19(1): 3–13.

24.

Diamond

Spillane

(2016) School leadership and management from a distributed perspective: A 2016 retrospective and prospective. Management in Education 30(4): 147–154.

25.

Dinham

Scott

(1998) A three domain model of teacher and school executive career satisfaction. Journal of Educational Administration 36: 362–378.

26.

Dinham

Scott

(2000) Moving into the third, outer domain of teacher satisfaction. Journal of Educational Administration 38(4): 379–396.

27.

Donohoo

Hattie

Eells

(2018) The power of collective efficacy. Educational Leadership 75(6): 40–44.

28.

Doppenberg

Den Brok

Bakx

AWEA

(2012) Collaborative teacher learning across foci of collaboration: Perceived activities and outcomes. Teaching and Teacher Education 28: 899–910.

29.

Dovey

(2009) The role of trust in innovation. The Learning Organization 16(4): 311–325.

30.

Dunn

Baguley

Brunsden

(2014) From alpha to omega: A practical solution to the pervasive problem of internal consistency estimation. British Journal of Psychology 105: 399–412.

31.

Enders

(2001) The performance of the full information maximum likelihood estimator in multiple regression models with missing data. Educational and Psychological Measurement 61(5): 713–740.

32.

Gümüş

Bellibaş

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

33.

Fullan

(1993) Why teachers must become change agents. Educational Leadership 50(6): 12–17.

34.

Fullan

(2016) The New Meaning of Educational Change. New York, NY: Teachers College Press.

35.

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.

36.

Goldsmith

(1986) Convergent validity of four innovativeness scales. Educational and Psychological Measurement 46(1): 81–87.

37.

Gopalakrishnan

Damanpour

(1997) A review of innovation research in economics, sociology and technology management. Omega 25(1): 15–28.

38.

Grewal

Cote

Baumgartner

(2004) Multicollinearity and measurement error in structural equation models: Implications for theory testing. Marketing Science 23(4): 519–529.

39.

Gronn

(2002) Distributed leadership. In: Leithwood

Hallinger

(eds) Second International Handbook of Educational Leadership and Administration. Dordrecht: Kluwer Academic, pp.653–696.

40.

Hairon

Goh

(2015) Pursuing the elusive construct of distributed leadership: Is the search over? Educational Management Administration & Leadership 43(5): 693–718.

41.

Hallinger

Kulophas

(2020) The evolving knowledge base on leadership and teacher professional learning: A bibliometric analysis of the literature, 1960–2018. Professional Development in Education 46(4): 521–540.

42.

Harman

(1960) Modern Factor Analysis. Chicago: University of Chicago Press.

43.

Harris

(2008) Distributed School Leadership: Developing Tomorrow’s Leaders. Abingdon, Oxford: Routledge.

44.

Hartley

(2007) The emergence of distributed leadership in education: Why now? British Journal of Educational Studies 55(2): 202–214.

45.

Heck

Hallinger

(2009) Assessing the contribution of distributed leadership to school improvement and growth in math achievement. American Educational Research Journal 46(3): 659–689.

46.

Herzberg

Mausner

Snyderman

(1959) The Motivation to Work (2nd ed). New York, NY: John Wiley and Sons.

47.

Hoyle

Wallace

(2005) Educational Leadership: Ambiguity, Professionals and Managerialism. London: Sage.

48.

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.

49.

Huang

Tang

, et al. (2019) Singapore’s school excellence model and student learning: Evidence from PISA 2012 and TALIS 2013. Asia Pacific Journal of Education 39(1): 96–112.

50.

Huber

(2008) School development and school leader development: New learning opportunities for school leaders and their schools. In: Lumby

Crow

Pashiardis

(eds) International Handbook on the Preparation and Development of School Leaders. New York, NY: Routledge, pp.163–175.

51.

Hülsheger

Anderson

Salgado

(2009) Team-level predictors of innovation at work: A comprehensive meta-analysis spanning three decades of research. Journal of Applied Psychology 94(5): 1128–1145.

52.

Hurt

Joseph

Cook

(1977) Scales for the measurement of innovativeness. Human Communication Research 4(1): 58–65.

53.

Jex

Britt

(2008) Organizational Psychology: A Scientist Practitioner Approach. Hoboken, NJ: Wiley.

54.

Karadag

(2020) The effect of educational leadership on students’ achievement: A cross-cultural meta-analysis research on studies between 2008 and 2018. Asia Pacific Education Review 21: 49–64.

55.

Kern

Graber

(2018) Understanding teacher change: A national survey of US physical educators. Research Quarterly for Exercise and Sport 89(1): 80–90.

56.

Kline

(2005) Principles and Practice of Structural Equation Modeling. New York, USA: Guilford Press.

57.

Kline

(2020) Becoming a Behavioral Science Researcher: A Guide to Producing Research That Matters. New York, USA: Guilford Press.

58.

Kondakci

Zaim Kurtay

and Caliskan O (2019) Antecedents of continuous change in educational organizations. International Journal of Educational Management 33(6): 1366–1380.

59.

Kondakci

Zayim

Beycioglu

, et al. (2016) The mediating roles of internal context variables in the relationship between distributed leadership perceptions and continuous change behaviours of public school teachers. Educational Studies 42(4): 410–426.

60.

Lance

Dawson

Birkelbach

, et al. (2010) Method effects, measurement error, and substantive conclusions. Organizational Research Methods 13(3): 435–455.

61.

Lazarus

Folkman

(1984) Stress, Appraisal, and Coping. New York, USA: Springer.

62.

Leithwood

Mascall

Strauss

, et al. (2007) Distributed leadership to make schools smarter: Taking the ego out of the system. Leadership and Policy in Schools 6(1): 37–67.

63.

Levine

Marcus

(2010) How the structure and focus of teachers’ collaborative activities facilitate and constrain teacher learning. Teaching and Teacher Education 26: 389–398.

64.

Leys

Delacre

Mora

, et al. (2019) How to classify, detect, and manage univariate and multivariate outliers, with emphasis on pre-registration. International Review of Social Psychology 32(1): 1–10.

65.

Lin

(2007) Knowledge sharing and firm innovation capability: An empirical study. International Journal of Manpower 28(3/4): 315–332.

66.

Little

(1990) The persistence of privacy: Autonomy and initiative in teachers’ professional relations. Teachers College Record 91(4): 509–536.

67.

Liu

Bellibaş

and Gümüş S (2021) 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 and Leadership 49(3): 430–430.

68.

Liu

Watson

(2020) Whose leadership role is more substantial for teacher professional collaboration, job satisfaction and organizational commitment: A lens of distributed leadership. International Journal of Leadership in Education. Epub ahead of print 18 November 2020. DOI: https://doi.org/10.1080/13603124.2020.1820580.

69.

Liu

Werblow

(2019) The operation of distributed leadership and the relationship with organizational commitment and job satisfaction of principals and teachers: A multi-level model and meta-analysis using the 2013 TALIS data. International Journal of Educational Research 96: 41–55.

70.

Locke

(1976) The nature and causes of job satisfaction. In: Dunette

(ed) Handbook of Industrial and Organizational Psychology. Chicago: Rand McNally, pp.1297–1349.

71.

Campbell

(2021) Conceptualising innovation and professional learning in the Hong Kong context. Practice 3(1): 67–72.

72.

Hallinger

(2018) A mirroring process: From school management team cooperation to teacher collaboration. Leadership and Policy in Schools 17(2): 238–263.

73.

Margolis

(2012) Hybrid teacher leaders and the new professional development ecology. Professional Development in Education 38(2): 291–315.

74.

Marsh

Hau

Wen

(2004) In search of golden rules: Comment on hypothesis-testing approaches to setting cut-off values for fit indexes and dangers in overgeneralizing Hu and Bentler’s (1999) findings. Structural Equation Modeling 11(3): 320–341.

75.

Maydeu-Olivares

(2017) Maximum likelihood estimation of structural equation models for continuous data: Standard errors and goodness of fit. Structural Equation Modeling: A Multidisciplinary Journal 24(3): 383–394.

76.

Mayrowetz

(2008) Making sense of distributed leadership: Exploring the multiple usages of the concept in the field. Educational Administration Quarterly 44(3): 424–435.

77.

McGeown

(1980) Dimensions of teacher innovativeness. British Educational Research Journal 6(2): 147–163.

78.

McLeod

Shareski

(2018) Different Schools for a Different World: School Improvement for 21st Century Skills, Global Citizenship, and Deeper Learning. Bloomington, IN: Solution Tree Press.

79.

Meyer

Richter

Hartung-Beck

(2022) The relationship between principal leadership and teacher collaboration: Investigating the mediating effect of teachers’ collective efficacy. Educational Management Administration and Leadership 50(4): 593–612.

80.

Mifsud

(2017a) Distributed leadership in a Maltese college: The voices of those among whom leadership is ‘distributed’ and who concurrently narrate themselves as leadership ‘distributors’. International Journal of Leadership in Education 20(2): 149–175.

81.

Mifsud

(2017b) Distribution dilemmas: Exploring the presence of tension between democracy and autocracy within a distributed leadership scenario. Educational Management Administration and Leadership 45(6): 978–1001.

82.

Moolenaar

Daly

Cornelissen

, et al. (2014) Linked to innovation: Shaping an innovative climate through network intentionality and educators’ social network position. Journal of Educational Change 15(2): 99–123.

83.

Moolenaar

Sleegers

PJC

(2010) Social networks, trust, and innovation. How social relationships support trust and innovative climates in Dutch schools. In: Daly

(ed) Social Network Theory and Educational Change. Cambridge, MA: Harvard University Press, pp.97–114.

84.

Muthén

(2017) Mplus Statistical Analysis with Latent Variables: User’s Guide (8th ed). Los Angeles, CA: Muthén & Muthén.

85.

Neto

Rodrigues

Panzer

(2017) Exploring the relationship between entrepreneurial behavior and teachers’ job satisfaction. Teaching and Teacher Education 63: 254–262.

86.

Nezlek

(2011) Multilevel Modeling for Social and Personality Psychology. London: Sage.

87.

Nguyen

(2020) Teacher collaboration for change: Sharing, improving, and spreading. Professional Development in Education 46(4): 638–651.

88.

Nguyen

Pietsch

and Gümüş S (2021) Collective teacher innovativeness in 48 countries: Effects of teacher autonomy, collaborative culture, and professional learning. Teaching and Teacher Education 106: 103463.

89.

Nsenduluka

Shee

(2009) Organisational and group antecedents of work group service innovativeness. Journal of Management and Organization 15(4): 438–451.

90.

OECD (2019a) TALIS 2018 technical report. Available at: https://www.oecd.org/education/talis/TALIS_2018_Technical_Report.pdf (accessed 11 March 2021).

91.

OECD (2019b) TALIS 2018 Results (Volume I): Teachers and school leaders as lifelong learners. Available at: https://www.oecd.org/education/talis-2018-results-volume-i-1d0bc92a-en.htm (accessed 11 March 2021).

92.

O’Shea

(2021a) Distributed leadership and innovative teaching practices. International Journal of Educational Research Open 2: 100088.

93.

O’Shea

(2021b) How relationships impact teacher job satisfaction. International Journal of Modern Education Studies 5(2): 280–298.

94.

Patterson

West

Shackleton

, et al. (2005) Validating the organizational climate measure: Links to managerial practices, productivity and innovation. Journal of Organizational Behavior 26(4): 379–408.

95.

Podsakoff

MacKenzie

Lee

, et al. (2003) Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology 88(5): 879–903.

96.

Printy

Liu

(2021) Distributed leadership globally: The interactive nature of principal and teacher leadership in 32 countries. Educational Administration Quarterly 57(2): 290–325.

97.

Rogers

(1995) The Diffusion of Innovations (4th ed). New York, USA: The Free Press.

98.

Schwabsky

Erdogan

Tschannen-Moran

(2019) Predicting school innovation: The role of collective efficacy and academic press mediated by faculty trust Journal of Educational Administration 58(2): 246–262.

99.

Seashore-Louis

Anderson

Riedel

(2003) Implementing Arts for Academic Achievement: The Impact of Mental Models, Professional Community and Interdisciplinary Teaming. Minnesota, USA: Center for Applied Research and Educational Improvement. Available at: https://hdl.handle.net/11299/143717.

100.

Seashore-Louis

Lee

(2016) Teachers’ capacity for organizational learning: The effects of school culture and context. School Effectiveness and School Improvement 27(4): 534–556.

101.

Serdyukov

(2017) Innovation in education: What works, what doesn’t, and what to do about it? Journal of Research in Innovative Teaching and Learning 10(1): 4–33.

102.

Sims

(2017) TALIS 2013: Working Conditions, Teacher Job Satisfaction and Retention. Statistical Working Paper. UK Department for Education.

103.

Smylie

Mayrowetz

Murphy

, et al. (2007) Trust and the development of distributed leadership. Journal of School Leadership 17(4): 469–503.

104.

Somech

(2005) Teachers’ personal and team empowerment and their relations to organizational outcomes: Contradictory or compatible constructs? Educational Administration Quarterly 41(2): 237–266.

105.

Somech

Naamneh

(2019) Subject coordinators as boundary managers: The impact on team learning and organizational outcomes. Educational Management Administration and Leadership 47(1): 56–73.

106.

Spector

(1997) Job Satisfaction: Application, Assessment, Causes, and Consequences. Thousand Oaks, CA: Sage.

107.

Spillane

(2005) Distributed leadership. The Educational Forum 69(2): 143–150.

108.

Spillane

(2006) Distributed Leadership. San Francisco, CA, USA: Jossey-Bass.

109.

Spillane

(2012) Distributed Leadership (Vol. 4). San Francisco, CA, USA: John Wiley & Sons.

110.

Spillane

Healey

(2010) Conceptualizing school leadership and management from a distributed perspective: An exploration of some study operations and measures. The Elementary School Journal 111(2): 253–281.

111.

Sun

Xia

(2018) Teacher-perceived distributed leadership, teacher self-efficacy and job satisfaction: A multilevel SEM approach using the 2013 TALIS data. International Journal of Educational Research 92: 86–97.

112.

Tabachnick

Fidell

(2012) Using Multivariate Statistics (6th ed). Boston: Allyn and Bacon.

113.

Tan

(2018) Examining school leadership effects on student achievement: The role of contextual challenges and constraints. Cambridge Journal of Education 48(1): 21–45.

114.

Thurlings

Evers

Vermeulen

(2015) Toward a model of explaining teachers’ innovative behavior: A literature review. Review of Educational Research 85(3): 430–471.

115.

Toropova

Myrberg

Johansson

(2021) Teacher job satisfaction: The importance of school working conditions and teacher characteristics. Educational Review 73(1): 71–97.

116.

Torres

(2018) Distributed leadership and teacher job satisfaction in Singapore. Journal of Educational Administration 56(1): 127–142.

117.

Torres

(2019) Distributed leadership, professional collaboration, and teachers’ job satisfaction in US schools. Teaching and Teacher Education 79: 111–123.

118.

Truijen

KJP

Sleegers

PLC

Meelissen

MRM

, et al. (2013) What makes teacher teams in a vocational education context effective? A qualitative study of managers’ view on team working. Journal of Workplace Learning 25(1): 58–73.

119.

Van der Heijden

HRMA

Beijaard

Geldens

JJM

, et al. (2018) Understanding teachers as change agents: An investigation of primary school teachers’ self-perception. Journal of Educational Change 19(3): 347–373.

120.

Vangrieken

Dochy

Raes

, et al. (2015) Teacher collaboration: A systematic review. Educational Research Review 15: 17–40.

121.

Vieluf

Kaplan

Klieme

, et al. (2012) Teaching Practices and Pedagogical Innovations: Evidence From TALIS. Paris: OECD Publishing.

122.

Westphal

Gulati

Shortell

(1997) Customization or conformity? An institutional and network perspective on the content and consequences of TQM adoption. Administrative Science Quarterly 42(2): 366–394.

123.

Yang

Huang

Y-F

(2008) A study of high school English teachers’ behavior, concerns and beliefs in integrating information technology into English instruction. Computers in Human Behavior 24: 1085–1103.

124.

Yousef

(2000) Organizational commitment and job satisfaction as predictors of attitudes toward organizational change in a non-western setting. Personnel Review 29: 567–592.