Promoting 21^st century pedagogy: can school autonomy and school-based professional development processes make a difference?

ABSTRACT

Public schools are expected to ensure that students who graduate the kindergarten (for 5–6 year olds) through twelfth grade (for 17–18 year olds) system are adept in 21^st century skills to be successful in the new workforce landscape. Hence, schools need to establish progressive teaching and learning processes that significantly differ from the traditional ones. Since learning environments designed for knowledge construction promote self-regulated and self-directed learners, the current study attempts to assess to what extent do teachers’ perceptions of their school’s autonomy and satisfaction with school-based professional development (PD) processes foster 21^st century oriented pedagogy. Data were collected from 645 teachers in 64 Israeli public junior high schools through questionnaires. A positive association was found between teachers’ perceived school autonomy and their satisfaction with school-based PD processes. High teacher satisfaction with PD was positively associated with progressive pedagogy. Teachers’ satisfaction with PD mediated the effect of school autonomy on using technology in teaching, on 21^st century pedagogical knowledge, and on teachers’ professional competence. School autonomy with the mediation of teachers’ satisfaction with PD processes makes a positive contribution to teachers’ professional competence, knowledge related to progressive teaching and the use of technology. Theoretical and practical implications are further discussed.

KEYWORDS:

• School autonomy
• 21st century pedagogy
• professional development
• teachers

Introduction

A growing number of scientists, politicians, and educators share the view that the development of students’ 21^st century skills is highly significant for their future success (Dishon & Gilead, 2021; Grannäs & Stavem, 2021; Jacobson-Lundeberg, 2016; Kay & Greenhill, 2011). Various conceptions about 21^st century skills have been proposed by organisations such as the European Union (EU) (2006), the Organization for Economic Co-operation and Development (OECD) (2005) and Partnership for 21^st Century Skills (P21) (2015), suggesting that these skills will be included in the core of educational processes (Ananiadou & Claro, 2009; González-Salamanca, Agudelo, & Salinas, 2020). When reviewing the various conceptualisations, several key skills are revealed that include ICT-related skills, collaboration, communication, social and cultural competencies, creativity, critical thinking and problem-solving (Haug & Mork, 2021; National Research Council, 2012; Van Laar, Van Deursen, Van Dijk, & De Haan, 2017; Van Laar, Van Deursen, Van Dijk, & de Haan, 2020; Voogt & Roblin, 2012). These skills are vital since employers demand fewer employees with basic skill sets and more people with complex thinking and communication skills (Mahmud & Wong, 2022), as well as technological skills (Laleye, 2022) to meet the needs of the modern workplace. Specifically, workers will need to develop their cognitive, intrapersonal, interpersonal, and technical skills (Ananiadou & Claro, 2009), which are the three classified broad domains of the 21^st century skills (Haug & Mork, 2021; National Research Council, 2012). The rapid advancements in technology and globalisation demand that employees will be flexible and perpetual learners to keep up with new developments (Johnson, Peters-Burton, & Moore, 2016). They will need to solve complex problems, to think critically, to effectively communicate with individuals coming from a variety of cultural and social contexts, work in collaboration with others, rapidly adapt to changing conditions, and acquire new information relevant to the present and future endeavours (González-Pérez & Ramírez-Montoya, 2022; National Research Council, 2011).

These newly created circumstances place public schools in a position where they are expected to ensure that students who graduate the K-12 system are adept in 21^st century skills so that they can be successful in this new workforce landscape (Sulaiman & Ismail, 2020). Schools should allow students to construct new knowledge themselves rather than reproduce or consume information (Prettyman, Ward, Jauk, & Awad, 2012; Stehle & Peters-Burton, 2019). Learning environments that are designed for knowledge construction promote self-regulated and self-directed learners as well as build perseverance and a deeper understanding of the content (Carpenter & Pease, 2013; Ross, Pirraglia, Aquilina, & Zulla, 2022), allowing the development of students’ skills in line with the demands characterising the 21^st century.

Previous research findings suggest that work autonomy positively contributes to workplace creativity (Amabile, Conti, Coon, Lazenby, & Herron, 1996; Li, Li, & Chen, 2018; Sia & Appu, 2015), leading to greater team effectiveness for projects pursuing radical innovations (Patanakul, Chen, & Lynn, 2012). Additional empirical evidence indicates a positive relationship between teacher autonomy and 21^st century teacher skills (Koçak & Karatepe, 2022). In line with these findings, the current study argues that school autonomy through increased degrees of freedom it provides to teachers may promote their ability to exercise agency and initiate teaching and learning processes, which better correspond to 21^st century demands. Autonomous schools may be more responsive, flexible and open to learn from others in educating children for the unpredictability and complexity of 21^st century challenges (Hargreaves, 2010; Mulgan, 2007; Suggett, 2015).

One aspect of public schooling, assumed to equip schools with flexibility and sufficient degrees of freedom that may enable them to meet this challenge, relates to their autonomy and degrees of freedom. School autonomy usually refers to the “decision-making structures in schools and their capacity to be self-governing and/or self-administering” (Thompson, Mockler, & Hogan, 2022, p. 86). Increased school autonomy allows school-level educators to shape teaching processes in ways that increase their compatibility to these newly created circumstances as well as to their community of teachers, students and parents. Hence, it is considered a reform strategy that enables public education systems to improve their standards and outcomes (Hashim, Torres, & Kumar, 2021). The current study attempts to assess the extent to which school autonomy may be a catalyst for 21^st century oriented teaching and learning processes using teachers’ perceptions of their satisfaction with PD as mediating the relationship between perceived school autonomy and pedagogical outcomes.

The current study was conducted in Israel. Hence, we frame the context by describing the Israeli education system, the centralisation-decentralisation issue that affects school autonomy, the status of the teaching profession and curricular reforms, including a reference to 21^st century pedagogy.

The Israeli education system

The Israeli education system reflects the heterogeneous structure of the society. It is composed of a Jewish majority combining people from various origins (immigrants who came from Europe, North Africa and America), and a minority (about 20%) of ethnic and religious Palestinian Arabs (Yemini & Addi-Raccah, 2013). Consequently, the school system comprises four sectors, three of which are public, free of charge and supervised by the state: the Jewish secular schools, the Jewish religious schools, and the Palestinian Arabs schools. A fourth sector comprises the semi-private ultra-Orthodox Jewish schools, which are only partially supervised by the state (ibid.). Since the establishment of the State of Israel (1948), a highly centralised system of government was established for allocating, among others, the educational resources, creating a hierarchical structure, where the Ministry of Education had highly significant impact on almost all aspects of the education system (Bogler, 2014).

Following research evidence testifying to teachers’ burnout which began to accumulate during the 1970s, a number of reforms were initiated aiming to decentralise the education system and increase the degrees of freedom granted to school-level educators (Volansky, 2007). Nevertheless, only around the 1990s following the state’s economic growth education was prioritised and reforms, incorporating information and communication technology to teaching and learning, were initiated. This era was characterised by greater emphasis on liberalism, equality, civil rights, market economy and privatisation processes, which prompted the implementation of policies that stressed school autonomy, accountability, and the strengthening of professional development programmes for teachers (Bogler, 2014). This was also the time when the Ministry of Education implemented and later on expanded the School-Based Management policy aimed to empower schools and increase principals and teachers’ autonomy and independence (Volansky, 2007), as has been the case in other countries (e.g. Keddie et al., 2022; Wilkins, 2015).

This was followed by two reforms implemented in the last decades in the elementary and middle schools (“Pedagogical Horizon”, 2008) and in the high schools (“Courage to Change”, 2012) attempting to promote teachers’ professional status. At the same time, the “Pedagogical Horizon” – Education for Thinking policy was introduced, attempting to promote 21^st century skills in teaching and learning processes. This policy emphasised the teaching of higher order thinking skills across the curriculum as a main universal educational goal to be implemented in the school system via three channels: curriculum, professional development of educational staff and assessment (Nir, Ben-David, Bogler, Inbar, & Zohar, 2016). During this time, a National Plan to Adapt the Education System to the 21^st century was introduced by the Ministry of Education integrating ICT (information and communications technology) based learning and teaching towards the 21^st century skills defined in accordance with the Assessment and Teaching of 21st Century Skills policy (ATC21S). The ICT reform included characteristics of “deep learning, learning for understanding and fostering higher order thinking, as well as critical thinking, problem-solving skills, creative thinking and metacognitive thinking” (Nir, Ben-David, Bogler, Inbar, & Zohar, 2016, p. 1239). In 2013, a national reform, entitled “Moving to meaningful learning” was introduced, strongly emphasising the 21^st century skills pedagogy. It was oriented towards a progressive pedagogy as it focused on meaningful learning and in-depth understanding (Ministry of Education, Israel, 2013).

It should be noted that studying the relationship between school autonomy and 21^st century pedagogy, while examining a mechanism that facilitates this relationship, is of considerable interest to international scholars who study educational structures similar to those in Israel (i.e. they experience the transition from a centralised to decentralised system and a transition from traditional to 21^st century pedagogy). “Generally outlined pedagogical reforms empower educators with the opportunity to act autonomously, prompting pedagogical creativity, individuality, and teamwork while simultaneously promoting the culture of leadership and internal accountability” (Bros & Schechter, 2022, p. 492). The Israeli case exemplifies these changes because of the reforms that were implemented throughout the years.

School autonomy and teachers’ professional development (PD)

Policymakers all over the world, for several decades, have initiated school autonomy policies intending to promote school educators’ accountability in their national education systems (Briggs & Wohlstetter, 2003; Caldwell, 2008; Cheng, Ko, & Lee, 2016; Hashim, Torres, & Kumar, 2021; Moradi, Hussin, & Barzegar, 2012). In essence, all these policies propose granting schools greater freedom in governance and decision-making (Imants, Zwart, & Breur, 2016) and decreasing central control (Dimmock, 2013; Neeleman, 2019). This tendency is based on the assumption that local autonomy and increased degrees of freedom at school level will improve public education by creating conditions for school educators to better respond to local needs (Keddie, 2017) and increase the relevance and effectiveness of public education (Arar & Nasra, 2020; Caldwell, 2016; Carr-Hill, Rolleston, Schendel, & Waddington, 2018; Dimmock, 2013; Nir, 2012). This claim follows an understanding that policymakers at the national level cannot be cognisant of the unique circumstances within which each school operates. Therefore, promoting schools’ autonomy will better allow them to provide adequate solutions to a wide array of needs and problems and diverse student characteristics.

Although all these reform policies use the same terminology, their exact configuration, scope and meaning vary according to the unique features characterising different contexts (Hashim, Torres, & Kumar, 2021; OECD, 2012). Hence, their impact on public education is subject to the governance mechanisms of the systems within which they are implemented. The differences of contextual features have direct implications for the potential impact this reform policy has in different educational contexts as the wide array of research findings indicates. Studies that attempted to evaluate the contribution of school autonomy to students’ learning, school effectiveness and educational equity have produced a mixed picture with findings that may be divided into two main categories. The first category contains evidence indicating the benefits of increased autonomy for schools and education systems. Research shows that school autonomy allows more effective pedagogy (Ugarte, Urpí, & Costa-París, 2022) as it amplifies the correspondence between school services and the particular needs of the community and the children that each school serves (Nir, 2012; Wohlstetter, Wenning, & Briggs, 1994). School autonomy also plays a major role in the promotion of teachers’ professional competence as it allows better adjustment of PD processes to the unique pedagogical challenges teachers are facing. It appears that teachers’ PD is most beneficial when it is school-based, focused on students’ learning, and linked to the school’s curricula (Darling-Hammond & Sykes, 1999; Garet, Porter, Desimone, Birman, & Yoon, 2001; Korthagen, 2017; Nir & Bogler, 2008; Sancar, Atal, & Deryakulu, 2021). Other findings further suggest that school autonomy enables schools to tailor instructional programmes to the teachers’ needs and puts greater emphasis on teachers’ discretion and creativity considered significant for the promotion of their PD (Lomos, Hofman, & Bosker, 2011; Newmann, King, & Youngs, 2000). Additional studies testify to the positive effects school autonomy has on students’ learning outcomes (Hanushek, Link, & Woessman, 2013; Mourshed, Chijioke, & Barber, 2010; Woessmann, Ludemann, Schutz, & West, 2007). Urban charter schools (Finn, Manno, & Vanourek, 2000), which have greater flexibility while operating as schools of choice oriented towards achieving educational objectives, increase teachers’ professional development time (Angrist, Pathak, & Walters, 2013; Dobbie & Fryer, 2013) and improve the academic achievements of students of colour (Abdulkadiroglu, Angrist, Narita, & Pathak, 2017). Moreover, the increase of school autonomy in areas such as curricula, assessments, and resource allocation seems to be positively associated with school performance when coupled with close public scrutiny of school test score performance (Hanushek, Link, & Woessman, 2013; Wöbmann, Lüdemann, Schütz, & West, 2008).

Contrary to the findings indicating the positive effect autonomy has on school performance and outcomes, some studies testify to the limitations and negative implications of autonomy for public schooling. Research has shown that school autonomy reforms tend to constrain the autonomy of teachers, subjecting them to increased control of their work and to stricter monitoring (Skerritt, 2023). One issue of concern involves educational equity. Research findings indicate that school autonomy increases competition and stratification within the system, promoting advantaged schools and weakening disadvantaged schools (Keddie et al., 2020, 2022; Lamb, Jackson, Walstab, & Huo, 2015). Fundraising activities, common in schools that are self-managed, have led to growing resource inequities among schools operating in a high or low socioeconomic environment (Yoon, Young, & Livingston, 2020), making public school systems less equitable. School autonomy harms the caring relationships between students and teachers due to prioritisation of success in examinations (Lewis & Pearce, 2021). It leads to a teach-to-the-test mentality (Hursh, McGinnis, Chen, & Lingard, 2019) due to schools’ need to compete for students and succeed in the high stakes exams that continue to dominate pedagogical considerations (Kauko & Salokangas, 2015; Salokangas & Ainscow, 2018). Hence, in many cases, school autonomy initiatives are found to have no effect or a negative effect on students’ learning outcomes (Jensen, 2013).

It is evident that research literature provides contradictory findings regarding the impact of autonomy on public education systems. Much of it is related to the way autonomy reforms are articulated and to the governance mechanisms in each national context within which they are implemented. Based on research findings indicating that centralised administration is perceived by teachers as a barrier when attempting to transform their teaching according to students’ needs in the 21^st century (Ben-David Kolikant, 2019), the current study aims to investigate the extent to which autonomy has the potential to promote 21^st century oriented teaching and learning.

21st century oriented pedagogy: a shift in teaching paradigm

Today’s global economy is characterised by interconnectedness; hence, the nature of the modern workplace and the professional challenges it presents requires fewer employees with only basic skill sets and more employees with higher-order thinking, communication, and collaboration skills (Kim, Raza, & Seidman, 2019; Saavedra & Opfer, 2012). These newly created circumstances suggest the need to develop students’ 21^st century skills emphasising their ability to think critically, their problem-solving capacity, their creativity, and their ability to work and communicate effectively with others.

As key socialisation agents, public schools are the main arena where students’ 21^st century skills should be acquired and established. Practically, schools need to introduce a major change in teaching and learning practices, primarily abandoning the dominant and traditional paradigm of “transmission” (Auslander, Myers, Bingham, & Tanguay, 2023; Peterson, Fennema, Carpenter, & Loef, 1989), where transfer of factual knowledge is performed through lectures and textbooks (Sibanda & Marongwe, 2022). Via the transmission paradigm, students learn new knowledge but they do not gain much practice in applying the knowledge to new contexts or using higher-order thinking skills (Zohar, 2023). Therefore, the traditional paradigm is not an effective way to develop students’ 21^st century skills (Boix-Mansilla & Jackson, 2011; Schwartz & Fischer, 2006).

Although it is clear that developing students’ 21^st century skills requires a major shift of the current teaching paradigm, this change is rather rare. Most teachers have not yet attained the expertise needed to develop their students’ 21^st century skills (Rotherham & Willingham, 2010, Toker, 2022), and they continue to conduct whole class instruction based on the transmission of knowledge (Bühler, Jelinek, & Nübel, 2022). This reality may be related to the fact that shifting from traditional to progressive teaching is a “Third-order” type of change (Blyth, 2013; Hall, 1993), which always has been difficult for implementation in public education systems (Cuban, 2013). Moreover, it seems that the transmission paradigm demands less pedagogical expertise from teachers than does the “constructivist” paradigm through which students actively, rather than passively, gain skills and knowledge (Schleicher, 2012).

It should be noted, however, that the practice of 21^st century skills may be affected by the existing standardised tests because teachers often prioritise testing over teaching. The prioritisation of tests suggests that there is less time for innovative instruction and for the curriculum involved with 21^st century pedagogy (Nahar, 2023). Furthermore, the emphasis on creativity, critical thinking, problem solving and overreliance on digital and technological literacy may not align well with the rigid structures of standardised testing methods. Another challenge in successfully practicing 21^st century pedagogy may lie in the lack of context-specific understanding of the needed teaching practices of this pedagogy, and the significant ways to support teachers in their professional development processes (Kim, Raza, & Seidman, 2019).

Researchers and practitioners agree that constructing educational processes aimed to develop 21^st century skills requires a strong human capital base (Barber & Mourshed, 2007). Moreover, teachers cannot cultivate students’ 21^st century skills if they only master basic low-order thinking skills themselves. Hence, there is a need to build the capacity of both the novice and the experienced teachers in developing students’ 21^st century skills through PD (Saavedra & Opfer, 2012).

To shift from the transmission to the constructivist paradigm, education systems need to invest in the transformation of teachers’ orientation and professional capacities. This requires time (Garet, Porter, Desimone, Birman, & Yoon, 2001; Sancar, Atal, & Deryakulu, 2021; Spiteri & Chang Rundgren, 2017), and the success of this shift depends on the way PD processes are designed and implemented. Teachers are less likely to change their orientation and practice as a result of low-order learning activities that occur via instructors’ presentation or the memorising of new knowledge (Desimone, Porter, Garet, Yoon, & Birman, 2002; Wayne, Yoon, Zhu, Cronen, & Garet, 2008). Moreover, it appears that teachers hardly change their instructional habits and strategies following one-time workshops (Gümüş & Bellibaş, 2023; Hawley & Valli, 1999), or when they are passive during PD activities and not asked to learn in ways that reflect how they should teach (Darling-Hammond & Sykes, 1999; Loyalka, Popova, Li, & Shi, 2019). Teachers’ PD processes seem to be most effective when they involve actual teaching materials (Putnam & Borko, 2000), when teachers from their school participate collectively (Garet, Porter, Desimone, Birman, & Yoon, 2001; Wayne, Yoon, Zhu, Cronen, & Garet, 2008), and when the design of the activities are school-based (Keese, Waxman, Asadi, & Graham, 2022; Wideen, Mayer-Smith, & Moon, 1998). Therefore, developing teachers’ capacity to employ 21^st century oriented pedagogy in their classrooms is more likely to occur when PD processes are school-based and school conditions support the implementation of what has been learned (Saavedra & Opfer, 2012).

Most studies reporting on the impact of autonomy on teaching and school effectiveness focus on students’ performance in learning assignments taught via the transmission paradigm. Less is known, however, about the extent to which school autonomy is translated to changes in teachers’ orientation that meet the needs of the modern workplace. The current study attempts to fill this gap by offering an answer to the following question: To what extent do teachers’ perceptions of their school’s autonomy and their satisfaction with PD processes support the application 21^st century oriented pedagogy? Specifically, our study attempts to test the following hypotheses:

Hypothesis 1:

School autonomy will be positively related to professional competence.

Hypothesis 2:

School autonomy will be positively related to 21^st century pedagogical knowledge, to 21^st century pedagogical practices and to the use of technology in teaching.

Hypothesis 3:

Teachers’ satisfaction with school-based PD will mediate the relationship between school autonomy and professional competence.

Hypothesis 4:

Teachers’ satisfaction with school-based PD will mediate the relationship between school autonomy and 21^st century oriented pedagogy.

Method

Sample

The purpose of this study was to assess the extent to which teachers’ perceptions of their school autonomy may explain their pedagogical orientation with particular emphasis on progressive pedagogy intended to develop students’ 21^st century skills.

The study is based on a nationwide sample of Israeli public junior high schools (grades 7^th through 9^th). The choice of junior high schools for the current study was supported by the fact that national policies oriented towards the promotion of school autonomy were not implemented in these schools at the time we collected our data. This allowed us to measure the relation between teachers’ perceived autonomy of their schools and pedagogical qualities excluding external influences initiated by the Ministry of Education. It should be noted, however, that a reform, entitled “Meaningful Learning”, started to be implemented in the education system around that time. This reform emphasises aspects of progressive pedagogy such as fostering students’ high-order thinking skills, inquiry-based learning, and relevant learning that are among the 21^st century skills (Bros & Schechter, 2022).

Data were collected from 64 schools located in four educational districts. In each school, at least 10 teachers completed the research questionnaire, and the dataset is based on the responses of 645 teachers. All schools were affiliated with the Jewish sector: 46% secular, 21% traditional, 33% religious schools.

Teachers’ average age was 42 years (SD = 11.1) with an average of 14.5 years of teaching experience (SD = 10.9). Fifty-five percent of the teachers were men. As for teachers’ education, 22% had a professional certificate, 11% earned a college degree, 28% earned a university Bachelor’s degree, 37% had a Master’s degree and 2% had a Ph.D.

Data collection

All data were collected in a single school year. Based on formal approval obtained from the Ministry of Education, questionnaires were administered in each school by trained research assistants. The research assistants distributed the questionnaires to teachers in the schools during the break and collected them by the end of the day to ensure confidentiality. Respondents were assured that anonymity is guaranteed.

The research instrument

School autonomy: This variable was measured using a scale developed by Schulz, Ainley, and Fraillon (2011) comprised of four sub-scales: the degree of school autonomy in core issues such as curriculum planning, teachers’ appointment, or budget allocation (12 items) (“How much autonomy does this school have in relation to curriculum planning?”); teachers’ involvement in school’s core issues such as establishing school priorities, resolution of conflicts or improvement of school governance (7 items) (“In your opinion, how many teachers in this school put forward useful suggestions for improving governance?”); parental involvement in issues such as school projects, school parental meetings or voting in school council elections (5 items) (“In your opinion, how many parents of students in this school participate in the school parent association?”); and, reliance on students’ opinions regarding various issues such classroom rules, school’s rules or teaching and learning materials (4 items) (“How much are students” opinions taken into consideration when decisions are made about classroom rules?’). The reliability reported for the whole scale is α = .87.

Satisfaction with school-based PD processes: This five-item scale, developed by Nir and Bogler (2008), measures teachers’ satisfaction with PD processes and programmes initiated by their schools (“The PD process in my school promotes my professional abilities”). The reliability reported for the whole scale is α = .81.

Teachers’ professional competence: This 32 item scale was originally developed by Van de Grift (2007), and was later modified and validated for school settings in studies conducted in various European countries (Maulana, Helms-Lorenz, & Van de Grift, 2017; Van de Grift, 2014; Van de Grift & Van der Wal, 2011; Van de Grift, Van der Wal, & Torenbeek, 2011). The scale measures five dimensions of effective teaching: safe and stimulating climate (3 items) (“Promotes the mutual respect and interest of students”); classroom management (3 items) (“Ensures efficient classroom management”); clear and activating instruction (4 items) (“Checks whether students understand the lesson content”); teaching learning strategies (3 items) (“Fosters critical thinking in students”); adaptation of teaching (3 items) (“Adapts the assignments and processing to the relevant differences between students”). The reliability reported for the items comprising the scale in various studies is above .70, indicating that all five dimensions reflect aspects of teachers’ instructional conduct.

21^st century pedagogy: This scale was based on items taken from the Teaching and Learning International Survey (TALIS) initiated by the Organization for Economic Co-operation and Development (OECD). Three groups of items representing 21^st century pedagogy were adopted: 21^st century pedagogical practices (“My role as a teacher is to facilitate students” inquiry’; α = .93); 21^st century pedagogical knowledge (“Thinking and reasoning processes are more important than specific curriculum content”; α = .91); and use of technology in teaching (“Students should use ICT [information and communication technology] for projects or classwork”; α = .72).

Analytical strategy

Data from the 645 participating teachers were nested within data from 64 schools, thereby requiring a multilevel approach (MLM). Intraclass correlations (ICC) for the mediator, teachers’ satisfaction with PD processes, and the dependent variables – teacher’s progressive pedagogy aspects and professional competence ranged in the present study from .01 to .58, which justified the use of a MLM analytical approach. Data were analysed using multilevel structural equation modelling (MSEM) in Mplus 7.2 (Muthén & Muthén, 2014). This method offers considerable advantages over conventional multilevel modelling procedures by allowing the simultaneous estimation of complex MLM mediation models with multiple outcome variables (Preacher, Zyphur, & Zhang, 2010).

Adopting notation proposed by Krull and MacKinnon (2001), the study tested a 2-2-1 multilevel mediation model, in which the predictor school autonomy and the mediator teachers’ satisfaction with PD processes were conceptualised at the school level (level 2), and the dependent variables measuring teachers’ progressive pedagogy and professional competence were conceptualised at the teacher level (level 1). We allowed teachers’ progressive pedagogy variables to covariate to account for their potential associations.

The goodness of fit was evaluated with the Tucker – Lewis index (TLI), comparative fit index (CFI), root mean square error of approximation (RMSEA), and the standardised root mean square residual (SRMR) (Browne & Cudeck, 1993; Hu & Bentler, 1995). Values of .95 or greater on the CFI and TLI, whereas RMSEA and SRMR less than .06 are often taken as evidence of excellent fitting models (Hu & Bentler, 1999).

To properly disentangle between-school variances from within-school variance and avoid a conflated bias, the values of the predictor and mediator were centred on their school mean (i.e. teachers’ deviation from school mean) (Zhang, Zyphur, & Preacher, 2009). Missing data were handled using full-information maximum likelihood estimation.

To test for the significance of the mediation effects, we followed Tofighi and MacKinnon’s (2016) recommendation and calculated a Monte Carlo-based parametric bootstrap approach using Mplus to estimate the 95% confidence interval (CI) of the unstandardised indirect effects of school autonomy on teachers’ progressive pedagogy through teachers’ capacity building. When the value zero is not included in the 95% CI, the mediation effect is considered significant at α < .05.

Aggregation tests

To justify the appropriateness of aggregating teachers’ ratings of school autonomy and teachers’ satisfaction with PD processes as reflecting a collective school experience, it is necessary to demonstrate that teachers are in agreement (e.g. teachers within a school have a similar experience of their school’s autonomy). To this end, we assessed both within-school agreement and between-school variability. First, we calculated ICC1. In general, ICC1 may be interpreted as the proportion of variance in teachers’ ratings that is attributed to systematic between-school differences compared to the total variance in ratings. Values as small as .10 may provide prima facie evidence of a medium group effect size, while values above .25 indicate a large effect size (LeBreton & Senter, 2008). ICC1 values were moderate and significant for school autonomy (ICC1 = .11, p = .003) and teachers’ satisfaction with PD processes (ICC1 = .18, p < .001). Next, we calculated ICC2 to determine the reliability of school means. ICC2 estimates the stability (i.e. reliability) of mean ratings from different teachers. The ICC2 for school autonomy (ICC2 = .54) and teachers’ satisfaction with PD processes (ICC2 = .68) indicated reasonable agreement (LeBreton & Senter, 2008). We also calculated the within-group agreement (rwg; James, Demaree, & Wolf, 1993) of the measures to assess whether teachers belonging to the same school were uniform in their school ratings to such an extent that their perceptions could be perceived as shared. Rwg for school autonomy (rwg=.86) and teachers’ satisfaction with PD processes (rwg=.75) reached the required minimum (rwg >.70; LeBreton, Burgess, Kaiser, Atchley, & James, 2003). Finally, results of a one-way random ANOVA (i.e. WABA test) indicated significant between-school differences for school autonomy, F(63,581) = 2.11, p < .001 and teachers’ satisfaction with PD processes, F(63,581) = 3.03, p < .001. In sum, aggregating school autonomy and teachers’ satisfaction with PD processes to school-level constructs was empirically justified.

Results

Descriptive statistics

Table 1 presents descriptive statistics for and correlations among the study’s main variables at the between-school level. The predictor, school autonomy, was positively associated with the mediator, teachers’ satisfaction with PD processes, but not significantly associated with any of the dependent progressive pedagogy variables or professional competence. This suggests that the effect of school autonomy on progressive pedagogy and professional competence is most likely indirect. The mediator, teachers’ satisfaction with PD processes, is positively associated with teachers’ professional competence, use of technology in teaching, and 21^st century pedagogical knowledge, but not significantly associated with 21^st century pedagogical practices. Among the dependent pedagogy variables, correlations were weak to moderate which indicates that they capture distinguished facets of pedagogy. Moderate significant associations were found between teachers’ professional competence, using technology in teaching, and 21^st century pedagogical knowledge.

Table 1. Descriptive statistics, correlations, and reliabilities (diagonal) for all research variables at the between-school level.

	1	2	3	4	5	6
1. School’s autonomy	.89
2. Teachers’ satisfaction with PD processes	.33**	.90
3. Teachers’ professional competence	.18	.54**	.97
4. 21^st century pedagogical practices	.23	.11	.09	.93
5. Using technology in teaching	−.15	.38**	.51**	−.02	.72
6. 21^st century pedagogical knowledge	−.13	.57**	.64**	−.03	.62**	.91
Mean	4.46	5.42	5.14	4.04	4.96	5.51
SD	0.37	0.60	0.54	0.44	0.61	0.83

*p < .05, **p<.01.

Multilevel mediation analyses

Results of the MSEM mediation analysis fit the data well (chi-square = 607.55, df = 42, p < .001; TLI = 1.06; CFI = 1.00; RMSEA < .01; SRMR < .01). Table 2 presents a detailed description of the direct and indirect effects of school autonomy and teachers’ satisfaction with PD processes on teachers’ progressive pedagogy variables. The results are summarised in Figure 1. Results of the hypothesised mediation analysis revealed a positive association between school autonomy and teachers’ satisfaction with PD processes. Specifically, the average score for the teachers’ satisfaction with PD processes variable was higher in schools that were more autonomy supportive. Higher teachers’ satisfaction with PD processes, in turn, was positively associated with the following aspects of school progressive pedagogy: using technology in teaching, 21^st century pedagogical knowledge, and teachers’ professional competence.

Figure 1. Direct and indirect effects of school autonomy and teachers’ satisfaction with PD processes on teachers’ progressive pedagogy and professional competence: standardised MSEM coefficients

Note: For indirect paths, the main indirect paths are presented in bold, while the non-significant paths are presented by a segmented line.

For the sake of clear representation, only significant direct paths are presented. *p<.05, **p<.01, ***p<.001.

Table 2. Direct and indirect effects of school autonomy and teachers’ satisfaction with PD processes on teachers’ progressive pedagogy and professional competence: multilevel mediation analysis.

	β	B	SE	T	p	95%CI
Mediator model (DV = Teachers’ satisfaction with PD processes)
School autonomy	.39	.63	.17	3.63	<.001	[0.29, 0.97]
Dependent variable model (DV = Using technology in teaching)
School autonomy	−.34	−.65	.18	3.67	<.001	[−1.00, −0.30]
Teachers’ satisfaction with PD processes	.41	.48	.12	3.98	<.001	[0.24, 0.72]
Dependent variable model (DV = 21^st century pedagogical practices)
School autonomy	.22	.26	.17	1.59	.112	[−0.06, 0.59]
Teachers’ satisfaction with PD processes	−.07	.05	.09	.58	.562	[−0.24, 0.13]
Dependent variable model (DV = Teachers’ professional competence)
School autonomy	−.08	−.11	.11	1.06	.290	[−0.32, 0.10]
Teachers’ satisfaction with PD processes	.63	.57	.11	5.30	<.001	[0.36, 0.79]
Dependent variable model (DV = 21^st century pedagogical knowledge)
School autonomy	−.31	−.92	.19	4.78	<.001	[−1.30, −0.54]
Teachers’ satisfaction with PD processes	.72	.99	.16	6.42	<.001	[0.69, 1.30]
Indirect effects of school autonomy via teachers’ capacity on:					B	95%CI
Using technology in teaching					.30	[0.07, 0.54]
21^st century pedagogical practices					−.03	[−0.14, 0.08]
Teachers’ professional competence					.36	[0.12, 0.60]
21^st century pedagogical knowledge					.63	[0.24, 1.02]

To test the significance of the indirect effects of school autonomy on progressive pedagogy and professional competence through teachers’ satisfaction with PD processes, we calculated 95% CIs. Cases in which the 95% CI did not include zero indicate a significant mediation effect. Results presented at the bottom of Table 2 indicate that teachers’ satisfaction with PD processes significantly mediated the effect of school autonomy on using technology in teaching, 21^st century oriented pedagogical knowledge, and teachers’ professional competence. However, the indirect effects of school autonomy on teaching 21^st century practices was not significant.

Furthermore, inspection of Table 2 revealed additional direct effects of school autonomy on using technology in teaching and on 21^st century pedagogical knowledge. In both cases, these relationships were negative. Given that school autonomy had a positive indirect effect on using technology in teaching and on 21^st century pedagogical knowledge, these negative direct effects may indicate suppression (MacKinnon, Krull, & Lockwood, 2000).

Discussion and conclusion

The current study attempted to assess the extent to which teachers’ perceptions of their school’s autonomy foster 21^st century oriented pedagogy using their satisfaction with PD processes conducted in their school as a mediator. The findings obtained support most of our hypotheses: the direct relationship between school autonomy and teachers’ professional competence was not statistically significant (Hypothesis 1); school autonomy was negatively related to the use of technology in teaching and to 21^st century pedagogical knowledge, and was not related to 21^st century pedagogical practices (Hypothesis 2); teachers’ satisfaction with PD processes mediated the relationship between school autonomy and professional competence (Hypothesis 3); and, teachers’ satisfaction with PD processes mediated the relationship between school autonomy and the use of technology in teaching and 21^st century pedagogical knowledge, and not the 21^st century pedagogical practices variable (Hypothesis 4). The findings allow five main conclusions that have both theoretical and practical implications.

Initially, in line with previous research findings, indicating that teachers’ PD is most beneficial when it is school-based (Nir & Bogler, 2008; Darling-Hammond & Sykes, 1999, Garet et al., 200; Sutherland, Markauskaite, & Cruickshank, 2023), our study shows that teachers’ perceptions of their school autonomy positively influence their satisfaction with the PD processes they experience in their school. It appears that the greater the local influence on the design and implementation of PD processes, the more relevant are these processes to local needs, and, therefore, the greater the teachers’ satisfaction with these processes.

A second conclusion refers to the direct negative associations found between teachers’ perceived school autonomy and two of the 21^st century oriented pedagogy characteristics. These findings seem to contradict the research body indicating that school autonomy allows effective pedagogy (e.g. Ugarte, Urpí, & Costa-París, 2022; Wohlstetter, Wenning, & Briggs, 1994). A potential explanation to these less expected findings may lie in the noteworthy results, in which school autonomy was found to have a positive indirect effects on both outcomes: using technology in teaching and 21^st century pedagogical knowledge. These positive indirect effects suggest that although autonomy has the potential to shape pedagogical processes according to school needs and local circumstances, it cannot by itself promote teachers’ 21^st century pedagogical knowledge or increase their use of technology in teaching. School’s extended degrees of freedom must be translated to teachers’ competencies before it can change teachers’ attitudes and the quality of teaching and learning processes. This conclusion is in line with Caldwell (2005) who asserts that to allow school autonomy to meet its potential, teachers’ PD processes should focus on “ … building the intellectual capital of the school, and involves the creation, dissemination, and utilisation of professional knowledge that takes account of a rapidly expanding knowledge base, and the need for access to that base in a timely and easily understandable manner” (p. 18).

The positive association found between school autonomy and teachers’ professional competence mediated by teachers’ satisfaction with PD processes leads to our third conclusion. This finding echoes the explanation provided above, indicating that teachers’ professional competence – teachers’ perceptions of their pedagogical and didactic behaviour (Van de Grift, 2007) – is strongly influenced by the school’s autonomy to shape and execute PD processes that correspond to the school’s needs and interests. Thus, teachers’ satisfaction with these processes induces them to develop their professional competence, as it has been demonstrated with relation to progressive pedagogy. This finding supports the tendency to conduct on-the-job training processes within the school context rather than use the services of external agencies, which provide continuing education programmes that do not take into account the unique features and circumstances of each particular teacher and school.

A fourth conclusion follows the finding showing that teachers’ satisfaction with PD processes conducted in their school mediated the effect of school autonomy on using technology in teaching and on 21^st century pedagogical knowledge, reflecting teachers’ acquaintance with the principles and assumptions of progressive pedagogy. Acknowledging that teachers typically differ in their technological competencies, our findings suggest that technological training processes should be tailored in accordance with the needs characterising teachers in each particular school to promote the effectiveness of these processes. It appears that school autonomy opens teachers’ minds and promotes their willingness to be exposed to ideas and knowledge relevant to 21^st century pedagogy. Apparently, working in a school considered autonomous promotes teachers’ psychological openness which reflects “ … readiness to perceive ongoing experience accurately, without distorting or attempting to avoid the experience, and a willingness to assimilate novel experiences into self-structures” (Hodgins & Knee, 2002, p. 88).

Our last conclusion refers to the finding showing that teachers’ satisfaction with PD processes did not mediate the relationship between school autonomy and 21^st century pedagogical practices. Contrary to teachers’ knowledge regarding 21^st century pedagogy, the 21^st century pedagogical practices variable stands for teachers’ perceptions of the extent to which instruction based on progressive pedagogical principles is actually practiced in their lessons. This finding may indicate the discrepancy existing between teachers’ awareness and theoretical knowledge regarding progressive teaching and their employment of these principles in their classrooms (Rotherham & Willingham, 2010; Toker, 2022). While school autonomy sets the conditions allowing schools to shape and transform teaching in ways that better meet the needs of the modern workplace, our findings suggest that teachers’ perceptions regarding their ability to initiate progressive teaching practices is a necessary though not sufficient condition. Teachers may need more time (Garet, Porter, Desimone, Birman, & Yoon, 2001; Sancar, Atal, & Deryakulu, 2021), further PD, and a supportive environment before applying 21^st century pedagogy in their classrooms.

Our findings show that school autonomy, along with teachers’ school-based PD processes, may play a major role in transforming teaching according to the assumptions and principles that guide progressive teaching. Currently, however, it seems that these qualities mostly shape teachers’ awareness of the ideas characterising progressive pedagogy, and that teachers need additional conditions, such as to experience more PD processes oriented towards 21^st century pedagogy before they can promote substantial changes in their progressive-oriented practices. Another possible explanation is that the application of teachers’ progressive knowledge is hindered because of a non-supportive environment. For example, Zohar (Zohar, 2023; Zohar & Alboher Agmon, 2018) found that teachers who expressed substantial knowledge of 21^st pedagogies refrained from applying them in their practice in a school culture dominated by high-stakes testing.

As with any piece of research, this study is not free of limitations, offering opportunities for future studies. One limitation lies in the fact that the study is based on single-source data, mainly teachers’ perceptions. Although the study attempted to assess teachers’ perspectives regarding the constructs comprising the research model, relying exclusively on their perceptions might have a confounding effect on the findings (Podsakoff & Organ, 1986).

Another related limitation may be evident in the fact that our study is based on teachers’ perceptions and self-reports rather than on their actual performance. Since meaning and action are collectively negotiated and context-dependent, self-reports of attitudes and behaviours are of limited value in explaining what people do since what people say may often be a poor predictor of what they do (Jerolmack & Khan, 2014).

Adopting a qualitative research method may provide a way to cope with these limitations. Using a case-study approach may allow deeper insights into the nuances in the implementation of principles guiding progressive teaching and learning and the way teachers perceive and are influenced by their school autonomy. These insights may also reveal the extent to which problem solving, creativity and high-order thinking skills are apparent in classroom teaching, and may shed more light on the relationship between school autonomy and pedagogy.

Previous findings have indicated that schools’ potential to benefit from autonomy and extended degrees of freedom depends on various contextual features such as the education system’s level of centralisation (Nir, 2009; Nir, Kondakci, & Emil, 2017), a country’s level of development (Hanushek, Link, & Woessman, 2013) or schools’ socioeconomic circumstances (Armstrong & Ainscow, 2018; Keddie, 2017; Nordholm, Wermke, & Jarl, 2022). Therefore, it is suggested that future studies will focus on schools’ contextual features to allow better understanding regarding the unique features characterising school autonomy policies and their impact on pedagogy.

Acknowledgement

Prof. Dan Inbar passed away before the submission of the final version of this manuscript. Dan was a dear colleague and friend. This article is dedicated to his memory.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was supported by a research grant received from the Chief Scientist at the Israeli Ministry of Education.

Notes on contributors

Adam Nir

Adam Nir is Professor of Educational Administration Policy and Leadership and the Abraham Shiffman Chair in Secondary Education at the Hebrew University of Jerusalem Israel. He is the former Chair of the Department of Education at the Hebrew University and the former elected President of the International Society for Educational Planning (ISEP). His research interests include school autonomy, School-Based Management, educational planning, leadership and human resource management in public education.

Ronit Bogler

Ronit Bogler is a full professor at the Open University of Israel, the Department of Education and Psychology. Her research interests include educational leadership, organizational citizenship behavior, citizenship pressure, empowerment, organizational and professional commitment.

Dan Inbar

Dan Inbar was a full professor in the Administration, Policy, and Leadership in Education program at the Hebrew University of Jerusalem Israel. He was the President of the International Society for Educational Planning (ISEP) in 1994-95. Prof. Inbar passed away in 2022.

Anat Zohar

Anat Zohar is a full professor at the Hebrew University of Jerusalem. She is also a faculty member at the Mandel School for Educational Leadership. Her main academic interests are the development of students’ higher order thinking and metacognition, teachers’ knowledge and professional development, and the challenges involved in scaling up instructional innovations across whole school systems. She served as Director of Pedagogical Affairs in the Israeli Ministry of Education, leading a pedagogical reform aimed at teaching for higher order thinking and deep understanding. In 2009-2010 she was a member in the Institute for Advanced Study, Princeton, NJ. https://en-anatzohar.huji.ac.il/

Adi Ben-David

Adi Ben David (https://education.huji.ac.il/adi-ben-david) is a lecturer and academic coordinator for natural sciences & mathematics education at the Hebrew University of Jerusalem. She is also a lecturer in master's degree and pedagogical instructor at the David Yellin Academic College of Education. Her main academic interests are science education; teacher education; higher-order thinking and metacognition in learning and instruction; pedagogical innovation and the development of 21st–century skills. From 2010–2016 she was an academic coordinator and lecturer for continuing teacher education at the Department of Teaching Studies and Professional Development, at the Hebrew University of Jerusalem.