Transformational leadership, professional learning communities, teacher learning and learner centred teaching practices; Evidence on their interrelations in Mozambican primary education

Studies in Educational Evaluation 60 (2019) 14–31

Contents lists available at ScienceDirect

Studies in Educational Evaluation journal homepage: www.elsevier.com/locate/stueduc

Transformational leadership, professional learning communities, teacher learning and learner centred teaching practices; Evidence on their interrelations in Mozambican primary education

T

Hans Luytena, , Manuel Bazob ⁎

a

Faculty of Behavioural, Management and Social Sciences, Department of Research Methodology, Measurement and Data Analysis, University of Twente, Enschede, Netherlands b Faculdade de Educação, Universidade Eduardo Mondlane, Maputo, Mozambique

ARTICLE INFO

ABSTRACT

Keywords: Transformational leadership Professional learning communities Teacher learning Learner centred teaching practices Structural equation modelling

This study examines the routes via which transformational leadership, professional learning communities and teacher learning lead to more learner centred teaching practices. By means of structural equation modelling (path analysis with latent variables) a conceptual model was tested, using questionnaire data from 518 teachers in 95 Mozambican primary schools. The data analysis shows that the effect of transformational leadership on teaching practices is substantial but indirect, running via professional learning communities and teacher learning. All in all, the impact of professional learning communities on teaching practices is particularly strong. Its effect is both direct and indirect, running via teacher learning. The findings show the relevance of the concepts transformational leadership and professional learning communities in an African context.

1. Introduction Educational reform is a complex undertaking and its success ultimately hinges on the capacity of teachers to realize changes in their teaching practices. In the end, reform can only be accomplished if teachers acquire new skills and adopt them in their daily work. This may be realized through teacher learning, which primarily focuses on development of individual skills (Kwakman, 2003), but mutual consultation, information exchange and collective decision making in professional learning communities is also seen as a highly relevant factor in this respect (Stoll, Bolam, McMahon, Wallace, & Thomas, 2006; Toole & Louis, 2002). School leaders are generally believed to be a major factor in the change process as well, as effective leadership may provide the right conditions for a collaborative school culture and stimulate teachers to extend and improve their skills. The importance of school leadership in educational change processes is widely acknowledged. A large number of research literature reviews and meta-analyses underscore its relevance. Theoretical models that describe the impact of leadership emphasize its indirect nature (Bossert, Dwyer, Rowan, & Lee, 1982; Hallinger & Heck, 1996; Leithwood & Levin, 2005), but empirical studies that explicitly assess indirect effects and specify mediating variables are relatively scarce (; Krüger, Witziers, & Sleegers, 2007;

Sebastian & Allensworth, 2012; Ten Bruggencate, Luyten, Scheerens & Sleegers, 2012). The present study addresses the effects of transformational leadership, professional learning communities and teacher learning on changes in teaching practices towards a more learner centred approach in Mozambican primary schools. As such it evaluates the potential of these theoretical concepts in realizing educational change in Mozambican primary education. The current paper contributes to the research literature as it provides an empirical assessment of the paths through which transformational leadership, professional learning communities and teacher learning directly and indirectly affect teaching practices. Specific attention is paid to the interrelations between transformational leadership, professional learning communities and teacher learning. Furthermore, the study relates to primary education in Mozambique, whereas the bulk of research on transformational leadership and professional learning communities has been conducted in western countries. Therefore the present study provides information on the relevance of conceptual models on transformational leadership, professional learning communities and teacher learning in a non-western context. The model to be tested postulates that both teacher learning and professional learning communities promote change toward more learner centred teaching practices. It furthermore posits a

⁎ Corresponding author at: Faculty of Behavioural, Management and Social Sciences, Department of Research Methodology, Measurement and Data Analysis, University of Twente, P.O. box 217, 7500, AE, Enschede, Netherlands. E-mail address: [email protected] (H. Luyten).

https://doi.org/10.1016/j.stueduc.2018.11.002 Received 25 November 2017; Received in revised form 10 November 2018; Accepted 13 November 2018 0191-491X/ © 2018 Elsevier Ltd. All rights reserved.

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

positive effect of professional learning communities on teacher learning. In addition, it is hypothesized that both professional learning communities and teacher learning mediate the effect of transformational leadership on teaching practices. Bazo (2011) reported a number of multilevel regression analyses with regard to the effects of transformational leadership in Mozambique. In this paper we present a re-analysis of the data collected for this study in 2008–2009, using structural equation modelling (path analysis with latent variables). Thus we are able to test the empirical validity of a conceptual model that posits indirect effects of transformational leadership via professional learning communities and teacher learning on teaching practices. By including latent variables in our statistical model, we take into account that the concepts transformational leadership and professional learning communities should be conceived as multidimensional constructs. In addition, our data analyses control for possibly confounding effects of contextual and antecedent factors (e.g. school location, teacher education). The analyses also take into account the hierarchical data structure (i.e. teachers nested within schools).

result, students go to school either in the morning or in the afternoon. More than one in five teachers in our sample work in more than one shift (see Table 2). Authors like Davies and Ellison (1997) and Beck (1999) argue that the results of (educational) decentralisation policies strongly depend on the capabilities of individual (school) leaders. This has also been acknowledged in official documents of the Mozambique education ministry (Ministério da Educação e Cultura, 2006; Ministério da Educação, 2004; Nhavoto, Buendia, & Bazo, 2009). In particular, school leaders have been expected to stimulate the professional development of their teaching staff. This is clearly illustrated in the following quotation:

1.1. Context: educational policy in Mozambique

1.2. Theoretical background

Although concepts like transformational leadership and organizational learning were originally developed in western contexts, the work of Nguni (2005) and Nguni, Sleegers, and Denessen, (2006) demonstrates their usefulness in an African environment, namely Tanzania. Their findings suggest a universal relevance of these concepts across a wide range of contexts and societies. Like Tanzania, Mozambique is situated in southeast Africa. It gained political independence in 1975 after nearly five centuries of Portuguese domination. During the 1980s the country was plagued by a civil war, which ended in 1992. Since then, the Mozambique administration has stimulated the decentralization of the national education system. Autonomy and decision-making responsibility of schools and local authorities has steadily increased. The aim of this policy has been to improve education and foster student learning. A problem in this respect is that instruction in Mozambican schools traditionally tended to be teacher centred drill and repeat practice rather than learner centred. Teaching was mainly focused on memorization and mechanized procedures. As a result, few pupils showed higher levels of comprehension or the ability to apply their knowledge in wider contexts. In reaction, the ministry of education developed a new, more learner centred primary education curriculum. On a general level, the aim of learner centred teaching is that students take more responsibility of their own learning process (Middlewood & Burton, 2001). More specifically, the curriculum includes instruction in mother tongue (instead of Portuguese) in the early grades and allows for coverage of content that is relevant for the particular needs in different parts of the country. This mainly relates to the natural resources in separate regions, e.g. oil and gas in the northern provinces vs. agriculture and fishing in the south. The primary education curricula were designed by the ministry of education taking into account the needs and resources in each region. The local communities were involved in the design of the curricula. The implementation of the curriculum started in 2004 (Ministério da Educação, 2004) and required teachers to adopt new instruction methods. In order to meet the learning needs of the teachers in this respect, the government launched new training programs, including in-service teacher training. The extent to which teachers actually changed their teaching practices towards learner centred teaching was assessed in the present study through a standardized questionnaire. Primary education in Mozambique is free and compulsory and comprises seven years of schooling, with the ages of the students ranging approximately from 6 to 13. The first five years make up the lower level (EP1). The upper level (EP2) comprises two final years. Because of the shortage of schools (and high numbers of students) primary schools often work in two shifts, typically 7.30–12.30 and 13.00-17.30. As a

The theoretical framework that underlies the present study posits that transformational leadership indirectly stimulates changes in teaching practices via professional learning communities and teacher learning. The dependent variable in our conceptual model relates to changes in teaching practices that fit a more learner centred curriculum as envisaged by the Mozambique ministry of education (Ministério da Educação, 2004). This means that teachers are expected to stimulate their pupils to focus on the application of practically relevant knowledge rather than on memorization of factual knowledge. Teachers are also expected to use a wider range of instructional approaches instead of mainly resorting to lecturing. This ambition has set high demands on the development of new skills on the part of the teachers, as success of educational reform is to a large extent determined by the teachers’ capacities to develop and adopt new instruction approaches. Learning efforts by teachers in order to improve their teaching skills are usually referred to as professional development. With regard to these efforts, three categories can be discerned, namely keeping up with new developments in the professional field, experimenting with new approaches in practice and individual reflection and discussion with colleagues (Kwakman, 2003). Beyond the development of individual skills, educational reform requires collaborative learning among teachers within schools. The basic idea behind this is that through interaction and exchange of experiences all teachers in a school profit from the individual knowledge and skills of their colleagues (Stoll et al., 2006). This can be realized through mutual consultation, information exchange and collective decision making in professional learning communities (Toole & Louis, 2002). Both professional learning communities and teacher learning are generally considered as factors that can stimulate changes in teaching practices, but the relative weight of both factors in promoting changes has to our knowledge hardly been addressed in empirical research. The present study yields an empirical assessment of the effects of both professional learning communities and teacher learning on new teaching practices. Our conceptual model posits substantial effects of both teacher learning and professional learning communities. Following Kwakman (2003) we consider professional development to consist of learning activities that aim to improve individual skills, whereas the basic idea behind professional learning communities is that through interaction and exchange of experiences all teachers in a school profit from the knowledge and skills of their colleagues (Stoll et al., 2006). The essential difference between teacher learning and professional learning communities boils down to whether the focus is on individual vs. collective knowledge and skills. However, as Kwakman (2003) points out, collaborative activities are part of teacher learning as

“Substantial investments in teacher training can have very little impact if school directors do not create a climate supportive of innovation and collaboration in their schools. Moreover, qualified school directors can provide supplementary “on the job” training for their teachers, either formally or informally (e.g., through classroom observation and subsequent discussion)” (Ministério da Educação, 1998, p. 15).

15

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

• Culture: the extent to which the principal promotes an atmosphere

well. Although teacher learning and professional communities are considered closely related concepts, it remains unclear whether teacher learning should be conceived as an outcome of professional learning communities. Darling-Hammond, Chung Wei, Andree, Richardson, and Orphanos, (2009) note that professional learning communities stimulate teacher learning, but an effect in the opposite direction might be at work as well. In that case, teacher learning also stimulates collaborative learning and professional learning communities (Castaneda & Rios, 2007), which implies a case of reciprocal causation. Still, the opportunities for a community to influence individuals seem stronger than the opportunities for individuals to influence their communities. Therefore we hypothesize that the effect of professional learning communities on teacher learning outweighs the effect in the other direction. We expect that the data analysis will reveal a relatively strong effect of professional learning communities on teacher learning, whereas the effect in the opposite direction will be relatively weak. This would imply that professional learning communities affect teaching practices both directly and indirectly via their effects on teacher learning. Parise and Spillane (2010) point to the importance of teacher learning for realizing changes in teaching practices. They also note that collaborative efforts can stimulate teacher learning and changes in teaching practices. As a result, we expect that the effect of professional learning communities on change toward more learner centred teaching is partly mediated by teacher learning. Apart from this indirect effect of professional learning communities, our model also postulates a direct effect of professional learning communities. In addition to a stimulating environment that promotes teacher learning, a professional learning community also includes aspects that may enhance changes in teaching practices directly, e.g. feedback on instruction, shared decision making and (self-) evaluation (Lomos, Hofman, & Bosker, 2011). True transformational leaders can create the conditions that enable employees to develop themselves and stimulate collaborative learning (Rolls, 1996, cited in Shukla, 1999). In educational settings, transformational leadership is strongly linked with school development, learning and vision. In contrast, transactional leadership, which is based on the exchange of rewards in return for work effort, primarily focuses on maintenance. According to Geijsel, Sleegers, Stoel, and Kruger, (2009), professional development is stimulated when teachers experience transformational leadership. The important role of school leaders, when it comes to bringing about changes in teaching practices is widely acknowledged by educational scholars (Marsh & Willis, 1999; Silins, Mulford, & Zarins, 2002; Sleegers, Bolhuis, & Geijsel, 2005; Toole & Louis, 2002). In particular, these authors point to the importance of professional development and collaborative learning among colleagues as intermediate factors in the trajectory from (transformational) leadership to improved teaching practices. Our conceptual model postulates that the impact of transformational leadership on new teaching practices is indirect. It is assumed that transformational leadership stimulates both professional learning communities and teacher learning, which in turn promote changes in teaching practices. Transformational leadership is generally conceived as a multidimensional concept. In this paper we follow a conceptualization that specifies six underlying dimensions. It has been adopted in a number of previous studies (Silins & Mulford, 2002; Mulford, 2003; Silins et al., 2002; Yu, Leithwood, & Jantzi, 2002), although other scholars have applied somewhat different conceptualizations (e.g., Hopkins, Ainscow, & West, 1998; Leithwood, Tomlinson, & Genge, 1996 Leithwood, Jantzi & Steinbach, 1999; Leithwood & Sun, 2012). The following six dimensions of transformational leadership will be discerned (cf. Bazo, 2011):

• • • •

of caring and trust among staff, sets a respectful tone for interaction with students and demonstrates a willingness to change his or her practices in the light of new understandings Structure: the extent to which the principal establishes a school structure that promotes participative decision making, supports delegation and distributive leadership and encourages teacher autonomy for making decisions Intellectual stimulation: the extent to which the principal encourages staff to reflect on what they are trying to achieve with students and how they are doing it; facilitates opportunities for staff to learn from each other and models continual learning in his or her own practice Individualized support: the extent to which the principal provides moral support, shows appreciation for the work of individual staff and takes their opinion into account when making decisions Performance expectation: the extent to which the principal has high expectations for teachers and students and expects staff to be effective and innovative

Silins and Mulford (2002) assert that transformational leadership stimulates both organizational learning and teacher learning: “The principal’s role is a significant one in facilitating school restructuring in general and, in particular, the reframing of schools as learning organizations” (p. 430). Silins et al. (2002) consider the capacity for collaborative and continuous learning throughout all levels of its organization as the defining characteristic of learning organizations. The notion of schools as learning organizations bears a close relation to the concept of professional learning communities, which exhibit practices like reflective dialogue, classroom observation, and providing feedback on each teacher’s work (Visscher & Witziers, 2004). Toole and Louis (2002) consider professional learning communities as a type of school culture particularly conducive to educational change and improvement. Scheerens, Glas, and Thomas, (2003) state that monitoring and evaluation is a crucial ingredient for learning processes in general, including collaborative learning. Visscher and Witziers (2004) list the following characteristics as the main features of professional learning communities (cf. Bazo, 2011).

• Policy and evaluation: the extent to which student achievement is monitored • Consultation and cooperation among teachers: the extent to which • •

school staff adjust work activities through mutual consultation and the informal exchange of information Consensus: the extent to which teachers have common views on matters like the tasks and function of the school, subject matter goals, teaching, and the teaching content Decision-making: the extent to which individual teachers can decide on their teaching independently, and the extent to which the subject’s group or grade as a “collective” plays a role

More recently, novel conceptualizations and operationalizations have been developed, which to our knowledge all agree that transformational leadership and professional learning communities are complex concepts that include several dimensions. The various conceptualizations differ to some extent on exactly which dimensions should be discerned, but several scholars have still relied on these dimensions or at least very similar dimensions (e.g., Lomos, 2012; Lomos et al., 2011; Moolenaar, Daly, & Sleegers, 2010; Moynihan, Pandey, & Wright, 2012; Saleh & Khine, 2014; Sun, Chen, & Zhang, 2017; Valckx, Devos, & Vanderlinde, 2018; Vanblaere & Devos, 2018). Fig. 1 summarizes our hypotheses regarding the relations between transformational leadership, professional learning communities, teacher learning and learner centred teaching practices. We expect that our data will show indirect effects of transformational leadership on learner centred teaching practices. We expect both a direct and indirect

• Vision and goals: the extent to which the principal works toward whole staff consensus in establishing school priorities and communicates these priorities and goals to students and staff, giving a sense of overall purpose

16

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

Fig. 1. Conceptual Model Transformational Leadership, Professional Learning Communities, Teacher Learning and Learner Centred Teaching Practices.

2. Method: sample, variables and data analysis

effect (via teacher learning) of professional learning communities on learner centred teaching practices. In our analyses, we will test if there is evidence for a reciprocal relationship between professional learning communities and teacher learning. It should be noted, though, that the present study is based on cross-sectional data. A detailed analysis of reciprocal relations should ideally draw on longitudinal data that include repeated measures. Therefore the findings on reciprocity between professional learning communities and teacher learning should be considered tentative. Finally, the model takes into account that school context, school leader and teacher characteristics may affect transformational leadership, professional learning communities, teacher learning and learner centred teaching practices.

2.1. Sample The analyses relate to a sample of 518 teachers from 95 primary education schools in Mozambique. The selected schools were located in five provinces (Maputo-City, Maputo Province, Zambézia, Nampula and Cabo Delgado). The population of these five provinces, which are evenly spread across the country (North, Central and South), represent more than half the entire Mozambique population. As many areas in Mozambique are difficult to access, random selection of schools was not feasible. The schools were selected by the authorities of the provincial or district Directorates of Education and Culture, taking into account the school level (either grades 1–5, grades 6–7 or grades 1–7), school location (urban vs. rural), and school accessibility. Although the local authorities selected the schools, they did not know which respondents participated in the survey. Names of participants were not recorded. The researcher (second author of this paper) personally guaranteed the participants confidentiality of their responses. Out of the 101 schools that were invited, 95 participated in the study, which implies a 94.1% response rate. In schools with six teachers or less, all teachers were involved in the study. In larger schools, six teachers participated. It was assumed that in addition to the school leader responses, six teachers per school would suffice to produce a complete picture of the school. With regard to selecting the teachers, a fairly even spread across grades was realized with a slight overrepresentation of grades 6 and 7. The data were collected in 2008-2009. Pilot studies were conducted in 2006 and 2007 in order to check whether filling out the questionnaires presented any problems to teachers and school leaders. The applicability of the central concepts of the study in the Mozambican context received special attention. Most respondents had no difficulty in filling out the questionnaires.

1.3. Research questions The main question this study aims to answer is whether the empirical data corroborate the conceptual model (see Fig. 1) with regard to the relations between transformational leadership, professional learning communities, teacher learning and learner centred teaching practices. In addition the following questions are addressed: 1 Do the data support the hypothesis of an indirect effect of transformational leadership on learner centred teaching practices via teacher learning and professional learning communities? 2 To what extent is the effect of professional learning communities on learner centred teaching practices mediated by teacher learning? 3 What is the relative weight of the effect of professional learning communities on learner centred teaching practices in comparison to the effect of teacher learning? 4 Is there evidence for a reciprocal relationship between professional learning communities and teacher learning? 5 Does the impact of antecedent variables account for relationships between the core variables?

17

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

2.2. Variables

Table 2 Categorical antecedent variables: frequency distributions.

The findings reported are primarily based on teacher responses, although the survey from which the data are drawn also included a school leader questionnaire. For further details, we refer to Bazo (2011). The core variables included in our analyses are all based on teacher reports. The information on school context, school leader and teacher characteristics is based on both teacher and school leader responses. Transformational leadership and learner centred teaching practices were addressed in both the teacher and school leader questionnaires. Correlations between teacher and school leader responses are close to zero and by no means statistically significant (Bazo, 2011; pp. 86, 111). Although this is an important finding in itself that deserves close attention in future research, this paper focuses on the teacher perspective. In the discussion section we will return to the issue of near zero-correlations between school leader and teacher responses. The concept transformational leadership captures six underlying dimensions (vision and goals, culture, structure, intellectual stimulation, individualised support and performance expectations) and professional learning communities four dimensions (policy and evaluation, consultation and cooperation among teachers, decision making and consensus). The remaining core variables (teacher learning and learner centred teaching practices) are conceived as unidimensional concepts. In order to measure the core variables in our model (including the underlying dimensions), a set of 5-point Likert-scale items was developed with responses ranging from strongly disagree to strongly agree (1–5). The items were drawn from previous work on transformational leadership (Mulford, Silins, & Leithwood, 2004; pp. 153-155), professional learning communities (Geijsel, 2001; p. 67, p. 100) and teacher learning (Geijsel, 2001; pp. 65-66). Appendix A lists the items used in this study (in Portuguese and their English translations). The items make up a number of Likert scales with high internal consistencies (Cronbach’s alpha). The measurement of teacher learning mostly relates to individual learning, as five out of six teacher learning items relate to individual activities. However, as Kwakman (2003) points out, collaborative activities are part of teacher learning as well. All these items were presented to the respondents under the heading “learning activities” (atividades de aprendizagem). The items that relate to learner centred teaching practices are primarily aimed at measuring change in teaching practices. Table 1 reports for the scales that capture the six dimensions of transformational leadership, the four

Percentage School location (Province) Maputo city* Maputo province Nampula Zambézia Cabo Delgado Training school leader School leader did not receive training in the area of Educational Administration and Management (EAM)* School leader received training in the area of Educational Administration and Management (EAM) Training teacher Teacher did not receive training that leads to an academic degree* Teacher received training that leads to an academic degree Teaching shifts Teacher works in one shift* Teacher works in more than one shift Teacher gender Female* Male

Transformational Leadership Vision and goals Culture Structure Individualized support Performance expectations Intellectual stimulation Professional Learning Communities Policy and evaluation Consultation and cooperation Consensus Decision making Teacher Learning Learner Centred Teaching Practices

Number of items

Cronbach’s alpha

Mean

Standard deviation

6 6 6 12 3

.88 .85 .86 .84 .78

4.13 4.13 3.95 3.90 3.92

.87 .87 .89 .83 .72

14

.91

4.07

.90

11 16

.84 .86

4.27 3.82

.64 .91

10 5

.78 .78

3.89 3.64

.72 .74

6 8

.84 .85

4.21 3.64

.84 .94

19.5% 80.5% 67.0% 33.0% 78.4% 21.6% 50.4% 49.6%

* Baseline category.

dimensions of professional learning communities, teacher learning and learner centred teaching practices the means, standard deviations and Cronbach’s alpha. The conceptual model takes into account that antecedent variables may account (in part) for observed relations between the core variables. Prior analyses on the same dataset by Bazo (2011) have shown that the nine antecedent variables listed in Tables 2 and 3 are possibly confounding variables. A series of multilevel regression analyses have shown a relation between these variables (out of a considerably wider range of variables) and at least one core variable. Three variables are based on school leader responses (whether or not the school leader had received training in the area of Educational Administration and Management, the number of teachers working at the school and the quality of the teachers according to the school leader) and five are based on teacher responses (whether or not the teacher had received training that leads to an academic degree, whether or not the teacher worked more than one shift, teacher gender, the extent to which the teacher liked to teach according to tried and tested ideas or methods, the extent to which the teacher did not like to deviate from the prescribed working method). Most teachers in the sample (74.1%) had received training to deal with the new curriculum. Prior analyses (Bazo, 2011) have shown that this training to deal with the new curriculum was not significantly related to any of the core variables. Whether teachers had received training leading to an academic degree, however, was significant for some of the core variables. Other types of training resulted in lower qualifications (certificates) or no qualification at all (in-service courses or seminars and workshops). School location (province) was already known when schools were selected for the survey. Table 2 lists the frequency distributions for the five categorical antecedent variables and Table 3 reports means and standard deviations for the four numerical ones. Four of the categorical variables are dichotomies. In these cases, a single dummy variable suffices to capture all relevant information. Table 2 also indicates which category served as the baseline in the analyses. For example in the case of gender, female is the baseline category. This implies that a positive effect of gender in our analyses denotes a higher score for male teachers. Province is the only categorical variable with more than two categories, namely five. Therefore four dummy variables were created with Maputo city (the country capital) as the baseline category. This implies that any effect of a province dummy expresses a higher or lower score compared to Maputo city.

Table 1 Transformational Leadership, Professional Learning Communities, Teacher Learning and Learner Centred Teaching Practices - number of items, Cronbach’s alpha and descriptive statistics. Scale

23.2% 23.6% 18.7% 18.9% 15.6%

18

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

Table 3 Numerical antecedent variables: descriptive statistics.

Number of teachers working at the school Quality of the teachers at the school as perceived by the school leader (1-5; very poor-very good) Teacher likes to teach according to tried and tested ideas or methods (1-5; strongly disagree-strongly agree) Teacher does not like to deviate from the prescribed working method (1-5; strongly disagree-strongly agree)

Mean

Standard deviation

39.2 3.53 3.84 3.14

25.9 0.71 1.22 1.28

3B). The correlations between the variables involved are reported in Appendices B–D.

2.3. Data analysis In total four structural equation models (SEM) were fitted to the data using the Mplus software (Muthén & Muthén, 2007). In each model, both transformational leadership and professional learning communities are treated as latent variables. The Likert scales “vision and goals”, “culture”, “structure”, “individualized support”, “performance expectation” and “intellectual stimulation” are the observed variables that together make up the theoretical construct transformational leadership (also see Table 1 and Appendix A). The scale scores are simply the averages of the responses to the questionnaire items that make up the scales. In the case of professional learning communities the observed variables are the Likert scales “policy and evaluation”, “consultation and cooperation”, “consensus” and “decision making”. The variables teacher learning and learner centred teaching practices each were measured by means of a single Likert scale. The models present a combination of path and measurement models. Such models are also called LISREL models or hybrid models (Kline, 2005). Our analyses also take into account the nesting of teachers within schools. The standard errors are corrected for the clustered sampling design, which creates non-independence of observations (Asparouhov & Muthén, 2006; Muthén and Muthén, 2007). We did not fit multilevel SEM models, which would involve estimating models with random intercepts or even random slopes. Such models are inherently more complex. In the case of random intercepts the residual variance can be attributed to either the teacher level or the school level. Fitting a random slopes model could reveal if certain relationships vary across schools. For our research questions this would hardly be relevant and therefore it was decided to fit less complex models. However, if the nesting of teachers within schools is ignored, the statistical significance of the results may be overestimated. This is avoided if the estimation of standard errors takes the sample design into account (i.e. the nesting of teachers within schools). Finally, our analyses take into account that school context, school leader and teacher characteristics may affect the core variables. It is conceivable that school context, teacher or school leader characteristics produce spurious correlations between the variables of interest (e.g. if both transformational leadership and teacher learning are both closely related to school location, this may produce a strong correlation between both variables; the correlation will decrease substantially, if one controls for school location in the statistical analysis). The first model that was fitted only includes the four core variables and ignores the possibly confounding effects of antecedent variables. In this case, only the hypothesized paths between the core variables are estimated. In the second model, the antecedent variables are included in the model as well. In this case, all effects of each antecedent variable on every core variable are estimated. In the third model all non-significant effects of antecedent variables on core variables are removed, which results in a more parsimonious model. Both model 2 and 3 show to what extent antecedent variables account for spurious correlations between the core variables. Finally an additional model (model 3B) is fitted in order to check whether the relationship between professional learning communities and teacher learning is a reciprocal one (model

3. Results This section describes the basic findings of the structural equation modelling. The next section discusses in more detail how these findings relate to the five research questions mentioned in section 1.3. Table 4 provides numerical details with regard to models 1-3. Fig. 2 shows a graphical display of the main outcomes with regard to model 3. Table 4 only reports information on the paths between the core variables and on model fit. Information with regard to path coefficients between antecedent variables and the core variables is provided in Appendix E. See Appendix F for information on the factor loadings of the observed variables on the latent variables. Only the results for model 3 are reported. The loadings are virtually identical for the other models. The scores on the fit indices in Table 4 indicate a good fit, although the χ2-values indicate a significant deviation between the observed data and the models fitted. In general, χ2 is considered a somewhat problematic index of model fit. It tends to suggest poor model fit in case of large samples or when correlations are high (Kline, 2005;). Still nearly all SEM reports provide information on χ2, as it is a key ingredient in most formulas of other fit indexes. Three frequently reported indices of model fit are the CFI (Comparative Fit Index), the RMSEA (Root Mean Square Error of Approximation) and the SRMR (Standardized Root Mean Square Residual). Values over .90 on CFI are generally considered indications of good model fit. SRMR values below .10 are considered favourable as well. RMSEA values below .05 indicate a good fit and values between .05 and .08 indicate a reasonable fit. Moreover, the upper bound of the 90% confidence interval of RMSEA should not exceed .10 and its lower bound should be less than .05. An important feature of the RMSEA index is that it corrects for model complexity and favours simple models (Kline, 2005). All three models in Table 4 show favourable scores on these three fit indices. The only exception is the RMSEA score for model 1. The .055 score suggests a “reasonable” rather than a “good” fit. These fit indices only indicate whether the model fits the data on average. Even with favourable scores on these indices some parts of the model may not fit the data very well. Therefore a closer inspection of the results for the distinct path coefficients is required. All hypothesized path coefficients are statistically significant (at least at the .05 level, but in most cases at the .01 level and far beyond). The significance levels of the path coefficients in Table 4 all relate to one-tailed tests, as all effects between the core variables are expected to be positive. The coefficients are somewhat stronger for model 1, but the decrease in models 2 and 3 is quite modest. This indicates that the antecedent variables can only account for the observed correlations to a limited extent. The highest coefficient relates to the effect of transformational leadership on professional learning communities (the standardized path coefficient in model 3 is .682). The effect of transformational leadership on teacher learning is relatively modest (the standardized path coefficient in model 3 equals .165), while the

19

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

Table 4 SEM analyses; standardized path coefficients (including standard errors) and goodness of fit indices. PATHS

From Transformational leadership Transformational leadership Professional learning communities Professional learning communities Teacher learning MODEL FIT χ2-value Degrees of freedom p-value CFI RSMEA value RSMEA 90% confidence interval SRMR value

To Professional learning communities Teacher learning Teacher learning Learner centred teaching practices Learner centred teaching practices

Model 1 (core variables only)

Model 2 (incl. antecedent variables, all paths)1

Model 3 (only significant paths of antecedent variables)1

Coeff. ** .807

S.E. .042

Coeff. ** .696

S.E. .056

Coeff. ** .682

S.E. .056

**

.217 .570 ** .481

.079 .078 .050

.134 .542 ** .469 **

.071 .080 .053

**

**

.165 .488 ** .478

.066 .074 .049

**

.051

**

.053

**

.050

.352

131.366 51 .0000 .971 .055 .044 - .067 .031

*

.274

284.015 155 .0000 .965 .040 .033 - .047 .023

**

.318

286.614 174 .0000 .969 .035 .028 - .043 .026

The outcomes as displayed in Fig. 2 correspond to model 3. 1 See Appendix E for details regarding paths from antecedent variables to the core variables. * path coefficient is significant at the .05 level (one-tailed). ** path coefficient is significant at the .01 level (one-tailed).

Fig. 2. Structural Model Transformational Leadership, Professional Learning Communities, Teacher Learning and Learner Centred Teaching Practices.

analyses reveal strong effects of professional learning communities on teacher learning and learner centred teaching practices (.488 and .478 respectively). The effect of teacher learning on learner centred teaching practices is substantial as well (.318). Table 5 lists the direct, indirect and total effects that can be inferred from the model 3 path coefficients. Even though there is no direct effect of transformational leadership on learner centred teaching practices, its total effect (via different routes) amounts to .484. Its total effect (direct and indirect) on teacher learning is approximately the same as the direct effect of professional learning communities on teacher learning (.498 vs. .488).

Table 6 shows that the additional analysis fails to produce empirical support for a reciprocal effect between professional learning communities and teacher learning. Adding a path from teacher learning to professional learning communities only produces a weak and non-significant coefficient. The model fit does not improve either. 4. Conclusion and discussion Before discussing the findings in more detail an important limitation of this study will be addressed, namely its strong reliance on teacher reports. This could present a threat to the validity of the conclusions. 20

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

Table 5 Direct, indirect and total effects based on model 3. EFFECTS From

To

Transformational leadership Transformational leadership Professional learning communities Transformational leadership Professional learning communities Teacher learning

Professional learning communities Teacher learning Teacher learning Learner centred teaching practices Learner centred teaching practices Learner centred teaching practices

Direct

—

Indirect

Total

—

.682 .498 .488 .484 .633 .318

.682 .165 .488

—

.478 .318

—

.333 .484 .155

The indirect effect of transformational leadership on learner centred teaching practices runs via three different routes (via teacher learning, via professional learning communities and via both teacher learning and professional learning communities). The sum of these indirect effects amounts to a total effect of .484.

schools the teachers showed substantial agreement on their responses with regard to transformational leadership and professional learning communities (Bazo, 2011). Even though the perceptions of school leaders and teachers within schools fail to show a clear relation, the perceptions of the teachers show a substantial degree of consistency with their colleagues in school. At the same time, it seems as though school leaders and their teachers are living in different worlds, as their perceptions of transformational leadership and professional learning communities seem unrelated. We consider both causes and consequences of (the lack of) agreement between teacher perceptions and school leader perceptions as important topics for future research, especially because the risk of negative consequences seems high in situations when positive perceptions of school leaders are not shared by the teachers (Devos et al., 2013). This lack of agreement deserves further study both in the Mozambique context and other contexts. Future research could explore the relation between the functioning of schools as professional learning communities and the extent of agreement between teacher and school leader reports. It may be the case that teacher and school leader reports align more closely if schools function as professional learning communities. The positive scale means in Table 1 may look like an indication of acquiescence bias, but this is disconfirmed by the teacher responses to some other survey questions, which were not included in the analyses reported here. This involves questions regarding concerns about teaching competencies and the rate of curriculum changes. The teachers responded clearly negative to statements on these matters (Bazo, 2011). Our general impression is that the questionnaires were completed with great care. The response rate among schools was very high (94.1%) and the respondents hardly ever skipped questions. The teachers in the current study were probably asked to fill in questionnaires much less frequently than the typical respondents to surveys in western countries. Whereas “survey fatigue” may be a serious problem in some parts of the world, the respondents in Mozambique showed much willingness to fill in the questionnaires. It also seems plausible that some hierarchy due to past centralization is still present. This may account as well for the high response rates in the present study. Even though education reforms started in 1992 through decentralization and reorganization of teacher training, the relation between schools and the ministry of education is more hierarchical than in most western countries. It should also be noted that from the time the data were collected (2008 and 2009) novel conceptualizations and operationalizations of transformational leadership and professional learning communities have been developed. The findings of the present study might have been different in some respects, had we chosen alternative conceptualizations and operationalizations. However, it seems unlikely that somewhat different operationalizations would have led to substantially different outcomes. In any case, we are not aware of any studies on the degree of divergence/concurrence between alternative operationalizations of the core concepts of the present study (i.e. to what extent does the same school score differently in one vs. another operationalization of transformational leadership or professional learning communities?).

Table 6 Model 3b: reciprocal effects of professional learning communities and teacher learning. EFFECTS From Transformational leadership Transformational leadership Professional learning communities Teacher learning Professional learning communities Teacher learning MODEL FIT χ2-value Degrees of freedom p-value CFI RSMEA value RSMEA 90% confidence interval SRMR value

To

Coeff.

S.E.

Professional learning communities Teacher learning Teacher learning

**

.083

*

.235 .388

.139 .189

.083

.134

.636

*

Professional learning communities Learner centred teaching practices Learner centred teaching practices

**

.050

**

.052

.481 .315

286.711 173 .0000 .969 .036 .028 - .043 .026

* path coefficient is significant at the .05 level (one-tailed). ** path coefficient is significant at the .01 level (one-tailed).

Especially the near zero correlations of teacher and school leader reports may cast doubt on the credibility of the teacher responses. Moreover, the mean scale scores with regard to teacher learning, learner centred teaching practices, transformational leadership and professional learning communities are all quite positive and show a clear tendency of agreement with the statements in the questionnaire (see Table 1). This may point to bias due to social desirability and/or acquiescence (i.e. a tendency among respondents to agree with any statement). Although this cannot be ruled out completely, it is also important to point to the arguments in favor of the credibility of the teacher responses. First of all, it needs to be mentioned that social desirability and acquiescence mainly affect the level of the scores and not necessarily the correlations between the variables. Social desirability and acquiescence may lead to unrealistically positive responses, but if this applies to all questions to the same extent, the correlations between variables are not affected. The only bias that occurs in that case relates to the level of the means scores, whereas the main focus of this study is on the relations between learner centred teaching practices, teacher learning, professional learning communities and transformational leadership. It is also important to note that near zero correlations between teacher reports and school leader reports are certainly not a unique feature of the present study. On the contrary, such results have been reported frequently in similar studies (Devos, Hulpia, Tuytens, & Sinnaeve, 2013; Moolenaar & Thoonen, 2009). Moreover, within 21

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

As the findings presented are based on teacher responses (except for a few antecedent variables), the empirical validity of the model only relates to the perceptions of the teachers. School leader perceptions were not included in the present study. The near-zero correlations of school leader perceptions with teacher perceptions present an important topic for closer study, but the result that the model is empirically supported with regard to teacher perceptions is important in its own right. The relations as specified in our model appear to be a valid representation of the relations between the core variables as perceived by teachers. If teachers experience leadership practices like intellectual stimulation and individualized support, they report more professional learning community and learning activities, which subsequently coincide with more learner centred teaching practices (as reported by the teachers themselves). A useful line of further inquiry could focus on student perceptions of teaching practices. Especially the correlation between teacher reports of classroom instruction and student perceptions is highly relevant. It seems hardly plausible that changes in teaching practices will have an impact on student learning, if the students do not perceive any difference. The ultimate goal of improved classroom instruction should be enhanced student learning. Several studies (Marks & Printy, 2003; Robinson, Lloyd, & Rowe, 2008) have shown that an orientation on teaching quality and student learning would ask for instructional leadership to guide the transformational leadership. Transformational leadership implies a focus on student motivation to learn more deeply and teacher motivation to learn and innovate. Research that focuses on the impact of transformational leadership on teaching quality and student learning should therefore also take into account the impact of instructional leadership. Our analyses provide empirical support for the conceptual model outlined in the first part of this paper (see Fig. 1). In accordance with prevalent theoretical notions on the impact of school leadership (Bossert et al., 1982; Geijsel et al., 2009; Hallinger & Heck, 1996; Krüger et al., 2007; Leithwood & Levin, 2005; Sebastian & Allensworth, 2012), an indirect effect on learner centred teaching practices was hypothesized. Its impact was assumed to run via professional learning communities and teacher learning. The analyses show that especially the effects of professional learning communities and the indirect effects of transformational leadership on learner centred teaching practices must be considered large (see Table 5). In accordance with the general consensus among educational experts teacher learning was found to have a substantial impact as well (Darling-Hammond et al., 2009; Kwakman, 2003; Parise & Spillane, 2010), but the present study particularly points to the importance of professional learning communities as levers for educational improvement. This clearly supports arguments by various authors that emphasize the potential of professional learning communities (Darling-Hammond et al., 2009; Lomos et al., 2011; Stoll et al., 2006; Visscher & Witziers, 2004). In the present study the total effect of professional communities (both direct and indirect) on teaching practices was found to be considerably stronger than the effect of teacher learning. This study therefore shows considerable potential for transformational leadership, professional learning communities and teacher learning in realizing educational change in Mozambique primary education. More specifically, with regard to the research questions in Section 1.3. the following conclusions can be drawn:

In addition, there is also an indirect effect of professional learning communities via teacher learning. 4 The analyses presented here show no empirical support for a reciprocal relationship between professional learning communities and teacher learning (see Table 6, model 3B). It should be emphasized, however, that the current study is based on cross-sectional data. It is conceivable that analyses based on repeated measures would present a more nuanced picture. 5 The findings suggest a moderate impact of the antecedent variables on the relations between the core variables in the conceptual model. The differences between model 1 vs. model 2 or 3 (see Table 4) are moderate in size. Although the estimated effects change to some extent, the results are largely similar with or without controlling for the antecedent variables. Moreover, the effects of the antecedent variables on the core variables are quite modest (see Appendix E). The findings point to a central role of professional learning communities in educational change. Stimulating the development of professional learning communities that focus on practices like mutual consultation, information exchange and collective decision making, may be the most promising approach to realize changes in teaching practices. This expectation is further supported by the finding that the relation between transformational leadership and professional learning communities is considerably stronger than the relation between transformational leadership and teacher learning. This indicates that there is considerably more potential for school leaders to promote professional learning communities in their schools than for the stimulation of teacher learning. It is conceivable that this finding applies exclusively to Mozambique schools. For example, if Mozambique teachers have more opportunity for teacher learning as compared to other countries. However, it seems more likely that the central role of professional learning communities found in the present study represents a general trend. As professional learning communities are a collective endeavour between teachers at the school level, it seems unsurprising that it is easier for school leaders to impact this, rather than teacher learning that mainly consists of activities that teachers undertake by themselves. The notion that transformational leaders may both stimulate teacher learning and promote professional learning communities is generally accepted (Geijsel et al., 2009; Rolls, 1996, cited in Shukla, 1999). The present study provides quantitative findings to compare the strength of both relations. In the end, the capacity of teachers to adopt changes in classroom instruction determines the success of any attempt to realize educational reform. Learning efforts by teachers aimed at individual development definitely stimulate the adoption of new teaching practices, but this study indicates that the effect of professional learning communities outweighs the effect of teacher learning. Through mutual consultation, collective decision making and information exchange teachers can profit from their colleagues’ skills and knowledge. Clearly, it is essential that teachers acquire new skills in order to realize educational change, but the impact of teacher learning can be greatly enhanced through professional learning communities. Teacher learning focuses on development of individual skills, while professional learning communities relate to mutual consultation, collective decision making and information exchange among teachers within schools. Thus teachers can learn from each other and reinforce the impact of individual knowledge and skills. It is also important to note that professional learning communities are unlikely to be effective if all teachers within a school lack basic teaching skills. Professional learning communities can only make a difference if the individuals involved are able to help and support each other. Therefore teacher learning that focuses on development of individual skills remains an essential factor in improving the quality of education. The current study stands out from the bulk of research on effects of leadership in education, as it addresses a non-western context. Concepts like transformational leadership, professional learning communities

1 The findings indicate that the effect of transformational leadership on learner centred teaching practices is indirect, mediated by professional learning communities and teacher learning. 2 The effect of professional learning communities on learner centred teaching practices is mediated by teacher learning to some extent. The direct effect accounts for about three quarters of the total effect (.478 out of .633, see Table 5). 3 All in all, the total effect of professional learning communities was found to be twice as large as the effect of teacher learning (.633 vs. .318, see Table 5). The direct effect of professional learning communities already outweighs the direct of effect of teacher learning. 22

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

and learner centred teaching practices were developed in a western context. Still, our analyses provide corroboration for the hypothesized relations between leadership, professional learning communities, teacher learning and changes towards more learner centred teaching practices in Mozambican primary education. Moreover, the six dimensions that are believed to underlie the concept of transformational leadership could be discerned in the data. The same goes for the four dimensions that underlie the concept of professional learning communities. Assuming that the relation between teacher and school leaders in Mozambique (and other non-western contexts) at present is still more hierarchical than in most western countries (due to centralization in the past) and taking into account that instruction traditionally focused on teacher centred drill and practice, it seems plausible that the potential

impact of changes with regard to transformational leadership and professional learning communities will be stronger in Mozambique (and other non-western contexts) than in most western countries. In comparison to western contexts there seems to be more room for change towards more learner centred teaching practices. In addition, there may also be considerable room for enhancing transformational leadership and professional communities. As the present study has shown that the relationships of transformational leadership, professional learning communities and teacher learning with learner centred teaching practices are quite strong, it seems justified to expect that substantial improvements can be accomplished, especially by promoting transformational leadership and professional learning communities in nonwestern contexts.

Appendix A. List of scale items (in Portuguese and English translations) In the present study, the items in Portuguese were administered. Response categories range from 1 to 5 (strongly disagree to strongly agree). Transformational Leadership – Vision and goals English 1. 2. 3. 4. 5. 6.

Português

My principal and/or vice-principal give(s) staff a sense of overall purpose

O meu Director e/ou seu Adjunto Pedagogico comunica (m) os propósitos das prioridades e metas da escola My principal and/or vice-principal help(s) clarify the specific meaning of the O meu Director e/ou seu Adjunto Pedagogico ajuda (m) a esclarecer o significado school’s mission in terms of its practical implications for programs and instruction especifico da missão da escola em termos de suas implicações praticas para programas e ensino My principal and/or vice-principal communicate(s) school mission to staff O meu Director e/ou seu Adjunto Pedagogico comunicam a missão da escola ao corpo docente My principal and/or vice-principal encourage(s) the development of school culture O meu Director e/ou Adjunto Pedagogico encoraja (m) o desenvolvimento da cultura da supporting openness to change escola apoiando a abertura a mudança My principal and/or vice-principal help(s) the staff understand the relationship O meu Director e/ou Adjunto Pedagogico ajuda (m) o corpo docente a entender o between their school’s mission and the subject’s/grade’s groups initiatives and relacionamento entre a missao da escola e iniciativas e politicas de grupos de disciplina e policies de classe My principal and/or vice-principal work(s) toward whole staff consensus in O meu Director e/ou seu Adjunto-Pedagogico trabalha (m) em prol do consenso do corpo establishing priorities for school goals docente em estabelecer as prioridades pra os objectivos da escola

Transformational Leadership – Culture

1. 2. 3. 4. 5. 6.

English

Português

My principal and/or vice-principal show(s) respect for staff by treating them as professionals My principal and/or vice-principal set(s) a respectful tone for interaction with students My principal and/or vice-principal demonstrate(s) willingness to change his or her (their) own practices in light of new understandings My principal and/or vice-principal model(s) problem-solving techniques that each teacher can readily adapt for work with colleagues and students My principal and/or vice-principal promote(s) an atmosphere of caring and trust among staff My principal and/or vice-principal symbolise(s) success and accomplishment within teacher’s profession

O meu Director e/ou seu Adjunto-Pedagogico mostra (m) respeito pelo corpo docente tratando-os como profissionais O meu Director e/ou seu Adjunto-Pedagogico estabelece (m) um tom de interacção com os alunos O meu Director e/ou seu Adjunto-Pedagogico demonstra (m) vontade para mudar suas praticas a luz de novos entendimentos O meu Director e/ou seu Adjunto-Pedagogico modela (m) tecnicas de solucao de problemas que cada professor pode adaptar para o trabalho com os colegas e com os alunos O meu Director e/ou seu Adjunto-Pedagogico promove (m) um clima de harmonia e confianca no seio do corpo docente O meu Director e/ou seu Adjunto-Pedagogico simbolizam sucesso e realização na profissão de professor

Transformational Leadership – Structure

1. 2. 3. 4. 5.

English

Português

My principal and/or vice-principal delegate(s) leadership for activities critical for achieving school goals My principal and/or vice-principal distribute(s) leadership broadly among the staff representing various viewpoints in leadership positions My principal and/or vice-principal ensure(s) that teachers have adequate involvement in decision making related to programs and instructions

O meu Director e/ou seu Adjunto-Pedagogico delega (m) a lideranca para actividades importantes para alcancar os objectivos da escola O meu Director e/ou seu Adjunto-Pedagogico distribuem a lideranca de forma alargada entre os professors com opinioes diferentes em varias posicoes de lideranca O meu Director e/ou seu Adjunto-Pedagogico certificam-se que os professores tem um envolvimento adequado na tomada de decisao em relacao a programas e o processo de ensino e aprendizagem O meu Director e/seu Adjunto-Pedagogico apoia (m) uma estrutura eficaz para a tomada de decisao O meu Director e/ou seu Adjunto-Pedagogico facilitam uma comunicacao eficaz entre os professores

My principal and/or vice-principal support(s) an effective committee structure for decision making My principal and/or vice-principal facilitate(s) effective communication among staff

23

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo 6.

My principal and/or vice-principal provide(s) an appropriate level of autonomy for teachers in their own decision making

O meu Director e/ou seu Adjunto-Pedagogico providencia (m) um nivel apropriado de autonomia a professores na sua tomada de decisao

Transformational Leadership – Performance expectations

1. 2. 3.

English

Português

My principal and/or vice-principal has (have) high expectations for teachers as professionals My principal and/or vice-principal has (have) high expectations for students My principal and/or vice-principal expect(s) teachers to be effective innovators

O meu Director e/ou seu Adjunto Pedagogico tem expectativas altas em relação aos professores como profissionais O meu Director e/ou seu Adjunto Pedagogico tem expectativas altas em relação aos alunos O meu Director e/ou seu Adjunto Pedagogico espera (m) que os professores sejam inovadores efectivos

Transformational Leadership – Individualized support

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

English

Português

My principal and/or vice-principal give(s) regular feedback to teachers in curriculum implementation and teaching My principal and/or vice-principal visit(s) each teacher’s classes to provide each teacher with feedback My principal and/or vice-principal encourage(s) each teacher to try new practices consistent with his or her own interests My principal and/or vice-principal stimulate(s) teachers to try out new didactic methods especially student centred approach My principal and/or vice-principal encourage(s) each teacher to pursue his or her own goals for professional learning My principal and/or vice-principal give (s) constructive criticism to teachers in dealing with the new curriculum and student centred teaching approach My principal and/or vice-principal take(s) opinion of each teacher into consideration when initiating actions that affect his or her work My principal and/or vice-principal is (are) aware of each teacher’s unique needs and expertise My principal and/or vice-principal is (are) inclusive; he or she (they) do not show (s) favouritism toward individuals or groups My principal and/or vice-principal provide (s) moral support by making each teacher feel appreciated for his or her contribution to the school When hiring new teachers, my principal and/or vice-principal let them know he or she (they) want(s) them actively involved in school decision-making on new curriculum and teaching methods My principal and/or vice-principal let teachers know they are responsible for all students, not just their own classes

O meu Director e/ou seu Adjunto Pedagogico da (o) feedback regular a cada professor na implementação do novo currículo e no ensino O meu Director e/ou seu Adjunto Pedagogico visita (m) as aulas de cada professor para fornecer feedback individual Ele (s) encoraja (m) cada professor a tentar novas praticas consistentes com seus próprios interesses O meu Director e/ou seu Adjunto Pedagogico estimula (m) os professores a tentar novos métodos didácticos especialmente a abordagem centrada no aluno O meu Director e/ou seu Adjunto Pedagogico encoraja (m) cada professor a definir seus próprios objectivos para a aprendizagem profissional O meu Director e/ou seu Adjunto Pedagogico faz (em) uma critica construtiva aos seus professores na abordagem do novo currículo e na abordagem de ensino centrado no aluno O meu Director e/ou seu Adjunto Pedagogico leva (m) em consideração a opinião de cada professor ao iniciar acções que afectem o seu trabalho O meu Director e/ou seu Adjunto Pedagogico esta (o) consciente das necessidades e perícia de cada professor O meu Director e/ou seu Adjunto Pedagogico e (sao) inclusivo (s); não mostra (m) favoritismo a um indivíduo ou a um grupo O meu Director e/ou seu Adjunto Pedagogico da (o) apoio moral fazendo com que cada professor se sinta apreciado pela contribuição que dá à escola Quando recebe (m) novos professores O meu Director e/ou seu Adjunto Pedagogico informa (m)-lhes que quer (m) que se envolvam activamente na tomada de decisão em relação ao novo currículo e métodos de ensino O meu Director e/ou seu Adjunto Pedagogico faz (em) com que os seus professores saibam que são responsáveis por todos os alunos e não somente pelas suas turmas

Transformational Leadership – Intellectual stimulation English 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

Português

My principal and/or vice-principal is (are) a source of new ideas for each teacher’s professional learning My principal and/or vice-principal value (s) the professional development of teachers

O meu Director e/ou seu Adjunto Pedagogico é (sao) uma fonte de novas ideias para aprendizagem profissional de cada professor O meu Director e/ou seu Adjunto Pedagogico valoriza (m) o desenvolvimento profissional dos meus professores My principal and/or vice-principal stimulate(s) each teacher to think about what he or O meu Director e/ou seu Adjunto Pedagogico estimula (m) cada professor a pensar she is doing for his or her students sobre o que ele/ela faz para seus alunos My principal and/or vice-principal create(s) opportunities for teachers to develop O meu Director e/ou seu Adjunto Pedagogico cria (m) oportunidades para o professionally in dealing with the new curriculum desenvolvimento profissional dos professores em lidar com o novo currículo My principal and/or vice-principal see(s) that teacher training activities are in harmony O meu Director e/ou seu Adjunto Pedagogico ve (m) que as actividades de formação with school development activities dos professores são em harmonia com as actividades de desenvolvimento da escola My principal and/or vice-principal encourage(s) staff to develop/review individual O meu Director e/ou seu Adjunto Pedagogico encoraja (m) os seus professores a professional growth goals consistent with school goals and priorities desenvolver/rever objectivos de crescimento profissional individual consistentes com os objectivos, metas e prioridades da escola My principal and/or vice-principal encourage(s) staff to evaluate their practices and O meu Director e/ou seu Adjunto Pedagogico encoraja (m) o corpo docente a avaliar refine them as needed suas praticas e refina-las como o desejado My principal and/or vice-principal encourage(s) and stimulate(s) teachers to visit each O meu Director e/ou seu Adjunto Pedagogico encoraja (m) e estimula os seus other’s classes to provide each other with feedback professores a se assistirem mutuamente para se fornecerem feedback My principal and/or vice-principal create (s) plans for the professional development O meu Director e/ou seu Adjunto Pedagogico elabora (m) planos para o desenvolvimento profissional My principal and/or vice-principal see (s) that training programs are followed up O meu Director e/ou seu Adjunto Pedagogico ve (em) se os programas de formação within the school itself estão a ser seguidos dentro da escola em si My principal and/or vice-principal stimulate(s) teachers to exchange knowledge and O meu Director e/ou seu Adjunto Pedagogico estimula (m) os seus professores a skills in order to help each other trocarem conhecimentos e habilidades por forma a se ajudarem mutuamente My principal and/or vice-principal facilitate(s) opportunities for staff to learn from O meu Director e/ou seu Adjunto Pedagogico facilita (m) oportunidades para os each other professores aprenderem uns dos outros My principal and/or vice-principal bring(s) workshops about new curriculum and O meu Director e/ou seu Adjunto Pedagogico traz (em) workshops sobre o novo student centred approach to school where it’s comfortable for teachers to participate currículo e sobre a abordagem de ensino centrado no aluno para a sua escola onde é confortável para os seus professores participarem

24

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo 14.

My principal and/or vice-principal give(s) workshops him or herself (themselves) and O meu Director e/ou seu Adjunto Pedagogico da (o) pessoalmente workshops e he or her (they) share(s) information with teachers on conferences that he or her (they) partilha (m) informação com os seus professores sobre conferencias nas quais attend(s) related to new curriculum and student centred teaching approach participa (m), referentes ao novo currículo e abordagem de ensino centrado no aluno

Professional Learning Communities – Policy and evaluation English 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

Português

In this school we have agreement upon rules with respect to our teaching activities Nesta escola temos acordos nos regulamentos e regras a respeito das nossas actividades de ensino We agreed upon rules about the number of “common tests” each subject group and Concordamos sobre as regras no número de provas gerais que cada grupo de disciplina ou grade takes per school year classe deve administrar por ano In this school we plan our teaching and assessment activities Nesta escola planificamos as nossas actividades de ensino e de avaliação In this school we have agreement on how to assess students Nesta escola acordamos nas formas/modos como avaliamos os nossos alunos In this school we have “common tests” to standardise the goal and content of Nesta escola temos provas gerais para para padronizar os objectivos e os conteúdos de teaching ensino In this school results on “common” tests are feedback to us in order to improve Nesta escola os resultados das provas gerais são feedback para nós para melhorar o teaching and learning processo de ensino e aprendizagem In this school we have evaluation rules Nesta escola temos regulamentos e regras de avaliação In this school we have school policy document Nesta escola temos documentos de politica educacional In this school we have agreement upon rules on the nature and content of testing Nesta escola temos acordos nas regras sobre a natureza e o conteudo das provas In this school we agreed upon rules on the pace of teaching and the teaching Nesta escola concordamos sobre o periodo de ensino e o conteudo a leccionar content In this school we have agreement upon rules related to aspects of didactics of Nesta escola acordamos sobre as regras relacionadas com aspectos das didácticas de ensino teaching

Professional Learning Communities – Consensus

1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

English

Português

Teachers share the same views about tasks and functions of the school It is easy to reach consensus about the goals of teaching Within this school we have the same ideas about how to teach Ideas about good teaching are shared by most teachers We agree with the use of mother tongue instruction We agree with the goals of inclusion of local content in the new curriculum We agree with the way of teaching local content If we want to change the teaching methods we agree easily In this school we have similar goals regarding how to deal with students In this school there are no conflicts about new curriculum and new teaching approach

Os professores partilham mesmas opiniões sobre tarefas e funções da escola É simples e fácil chegar a consenso sobre objectivos do ensino Nesta escola temos as mesmas ideias sobre os objectivos e metodos de ensino As ideias sobre bom ensino sao partilhadas por muitos professores Concordamos com o uso das linguas maternas no processo de ensino e aprendizagem Concordamos com os objectivos da inclusão de conteudos locais no novo curriculo Concordamos com a forma de ensinar conteudos locais Se quisermos mudar os metodos de ensino concordamos facilmente Nesta escola temos os mesmos objectivos sobre como lidar com os alunos Nesta escola existem não conflitos sobre o novo curriculo e sobre a nova abordagem de ensino

Professional Learning Communities – Consultation and cooperation

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

English

Português

I can always turn to my colleagues with problems and questions My colleagues are willing to discuss new didactical methods with me My colleagues are willing to give feedback on the way I teach if requested My colleagues support me in trying out new methods My colleagues tell me about their problems and solutions My colleagues pass on their learning experiences after participation in training programs I have to consult colleagues a lot of my work I consult my colleagues about planning teaching activities I have to contact with my colleagues about the pace of teaching and the teaching content I consult my colleagues about the nature and content of testing We exchange experiences with schools that are facing similar problems We spent time on reflecting jointly on our goals We jointly reflect on what has to be unlearned In this school we review what has gone well and what has not regarding to teaching approach In this school we look for the causes of success and failure in implementing the new curriculum In this school time is reserved to make plans for improvements

Conto sempre com os meus na solução de meus problemas Os meus colegas desejam/gostam de discutir novos métodos didácticos comigo Os meus colegas desejam/gostam de dar feedback sobre as minhas aulas quando solicitados Os meus colegas apoiam-me na tentativa do uso de novos metodos de ensino e aprendizagem Os meus colegas falam-me sobre seus problemas e soluções Os meus colegas transmitem suas experiencias de aprendizagem depois de participarem em programas de formação Devo consultar muito aos meus colegas sobre o meu trabalho Consulto aos meus colegas sobre as actividades de planificacao do ensino Devo contactar os meus colegas sobre o periodo e conteudo de ensino Consulto os meus colegas sobre a natureza e o conteúdo de das provas Trocamos experiencias com as escolas que enfrentam os mesmos problemas Levamos tempo reflectindo em conjunto sobre os nossos objectivos Juntos pensamos sobre o que nao deve ser aprendido Nesta escola revemos o que foi correcto e o que foi errado a respeito da abordagem de ensino Nesta escola procuramos as causas do sucesso e do fracasso na implementacao do novo curriculo Nesta escola é reservado tempo para elaborar planos de aperfeicoamento

25

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

Professional Learning Communities – Decision making

1. 2. 3. 4. 5.

English

Português

Plans for the new curriculum are discussed with the entire teaching staff at our school I can influence the plans being made at our school for the implementation of the new curriculum I have the space to implement the student centred methods in my classroom according to my own judgement and insights At our school, the experiences of teachers play a role in the formulation of a plan for the implementation of the new curriculum During the implementation of the new curriculum at our school, the problems that I may experience in doing this are taken into consideration

Os planos para o novo curriculo sao discutidos com todo o corpo docente da nossa escola Posso influenciar os planos a serem elaborados na nossa escola para a implementação do novo curriculo Tenho espaco para implementar o metodo de ensino centrado no aluno nas minhas aulas, de acordo com os meus criterios e meu juizo de valor Na nossa escola as experiencias dos professores jogam um papel para a formulacao de um plano para a implementacao do novo curriculo Durante a implementação do novo curriculo na nossa escola, os problemas que enfrento neste processo são tomados em conta

Teacher Learning

1. 2. 3. 4. 5. 6.

English

Português

I keep myself informed on developments within educational science Even if participation is not obligatory, I participate in training programs In my lessons I experiment with various didactic methods I use new knowledge and skills in my lessons I use the reactions of pupils to improve my teaching practices I discuss the problems that I experience at work with my colleagues

Mantenho-me informado sobre os desenvolvimentos na ciência educacional Participo em programas de formação, mesmo quando a participação nao e obrigatoria Nas minhas aulas experimento vários metodos de ensino Nas minhas aulas uso novos conhecimentos e habilidades Uso as reacções dos alunos para melhorar as minhas práticas de ensino Discuto, com os meus colegas, os problemas que experimento no trabalho

Learner Centred Teaching Practices English 1. 2. 3. 4. 5. 6. 7. 8.

Português

My pupils have to work on tasks more frequently (instead of listening to the teacher)

Os meus alunos trabalham frequentemente em tarefas do que ouvindo somente o professor I more frequently give pupils tasks involving small-scale research Dou tarefas com frequencia, envolvendo pesquisas de pequena escala I place more emphasis on the way problems should be dealt with instead of the problem Enfatiso mais a forma como os problemas devem ser resolvidos do que os problemas itself em si The examples that I give are more practical Os exemplos que dou sao mais práticos My pupils work more autonomously Os meus alunos trabalham mais independentemente I use more varied teaching methods and instructional formats Uso metodos e formas de ensino variados I use more varied teaching materials Uso material de ensino variado My teaching is more about practical matters O meu ensino é mais sobre assuntos práticos

Appendix B Table B1 Table B1 Correlation matrix of core variables. Vision & goals Vision & goals Culture Structure Individualized support Performance expectations Intellectual stimulation Policy & evaluation Consultation & cooperation Consensus Decision making Teacher learning Learner centred teaching **

Culture

Struct.

Ind. supp.

.720 **.701 **.641 **.552

1 **.736 **.640 **.557

1 **.737 **.617

1 **.647

**.731 **.567 **.576

**.715 **.465 **.457

**.787 **.551 **.539

**.806 **.540 **.561

**.655 **.432 **.535

1 **.553 **.583

1 **.675

**.546 **.512 **.617 **.569

**.473 **.422 **.490 **.450

**.519 **.496 **.583 **.545

**.558 **.521 **.557 **.567

**.503 **.476 **.550 **.547

**.610 **.559 **.606 **.606

**.567 **.550 **.604 **.520

1

**

Perf. exp.

Intell. stim.

Policy & eval.

Consult. & coop.

Cons.

Dec. making

Teacher learning

Learner centred

1 **.535 **.556

1 **.711

1

1

correlation is significant at the .01 level (two-tailed).

26

1 **.640 **.641 **.607 **.608

1 **.617 **.595 **.619

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

Appendix C Table C1 Table C1 Correlation matrix of antecedent variables with core variables.

Maputo province Nampula Zambézia Cabo Delgado School leader received EAM training Teacher training leading to an academic degree Teacher works more than one shift Male teacher Number of teachers working at the school Perceived quality of teachers at the school Like to teach according to tried and tested ideas Don’t like to deviate from prescribed working method

Maputo province Nampula Zambézia Cabo Delgado School leader received EAM training Teacher training leading to an academic degree Teacher works more than one shift Male teacher Number of teachers working at the school Perceived quality of teachers at the school Like to teach according to tried and tested ideas Don’t like to deviate from prescribed working method

Vision & goals

Culture

Struct.

Ind. supp.

Perf. exp.

Intell. stim.

Policy & eval.

**-.258 -.035 **.195 **.123 -.073 **-.323 .038 -.025 .028 **.156 **.401 **.122

**-.213 -.025 **.179 *.091 .017 **-.266 .064 .022 *.096 .075 **.305 *.093

**-.219 -.085 **.229 *.095 .001 **-.303 .081 -.009 *.108 *.088 **.351 *.100

-.233 -.046 **.228 *.098 -.013 **-.279 .071 .005 .048 *.099 **.380 *.111

**-.216 -.078 **.198 **.114 .024 **-.291 *.102 -.014 **.130 *.101 **.363 **.202

**-.239 -.071 **.291 .068 .007 **-.333 .046 -.022 .018 *.094 **.379 **.122

**-.177 -.048 **.131 **.135 -.076 **-.158 .022 -.085 .059 .051 **.412 **.171

Consult. & coop.

Cons.

Dec. making

Teacher learning

Learner centred

**-.211 .001 **.196 **.145 *-.094 **-.231 .010 *-.101 -.009 .082 **.445 **.213

**-.289 -.001 **.222 **.177 -.062 **-.308 .080 -.068 .024 .063 **.425 **.194

**-.263 .007 **.199 **.196 -.054 **-.269 .035 -.073 .026 .060 **.376 **.229

**-.243 -.071 **.193 **.151 **-.120 **-.399 **.167 -.044 **.140 *.104 **.497 **.165

**-.220 -.059 **.175 **.207 **-.114 **-.370 **.135 -.035 *.089 **.115 **.480 **.156

**correlation is significant at the .01 level (two-tailed). *correlation is significant at the .05 level (two-tailed).

Appendix D Table D1

27

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

Table D1 Correlation matrix of antecedent variables.

Maputo province Nampula Zambézia Cabo Delgado School leader received EAM training Teacher training leading to an academic degree Teacher works more than one shift Male teacher Number of teachers working at the school Perceived quality of teachers at the school Like to teach according to tried and tested ideas Don’t like to deviate from prescribed working method

Maputo province Nampula Zambézia Cabo Delgado School leader received EAM training Teacher training leading to an academic degree Teacher works more than one shift Male teacher Number of teachers working at the school Perceived quality of teachers at the school Like to teach according to tried and tested ideas Don’t like to deviate from prescribed working method

Map. prov.

Namp.

Zamb.

Cabo Delgado

1 **-.266 **-.268 **-.239 **.147 **.171 -.048 .041 .243 .064 **-.166 -.041

1 **-.232 **-.207 .074 **.137 -.072 .078 *-.105 **-.214 .014 .011

1 **-.208 -.011 **-.150 **.117 -.075 .056 *-.106 **.140 -.047

1 **-.231 **-.121 **.135 .072 *-.095 **.119 **.131 **.129

EAM Training

1

.004 *-.097 -.009 -.004 **-.180 **-.136 *-.104

Ac. Degree

1 *-.089 .075 *-.090 -.070 **-.204 -.016

>1 shift

1 **.201 **.186 -.064 .027 -.059

Male Teacher

Nr. of Teachers

Qual. Teachers

Tried & tested

Not to deviate

1 .040 -.021 .007 *-.108

1 -.075 .033 .029

1 -.004 .061

1 **.269

1

**correlation is significant at the .01 level (two-tailed). *correlation is significant at the .05 level (two-tailed).

Appendix E Table E1

28

29

**-.192 -.010 **.153 .040 *.105 **-.202 .018 .024 *.095 **.139 **.325 .043

.026 *.084 .086 *.125 -.031 -.011

-.010 **-.100 -.022

.008

**.182

**.088

**coefficient is significant at the .01 level (two-tailed). *coefficient is significant at the .05 level (two-tailed).

Maputo province Nampula Zambézia Cabo Delgado School leader received EAM training Teacher training leading to an academic degree Teacher works more than one shift Male teacher Number of teachers working at the school Perceived quality of teachers at the school Like to teach according to tried and tested ideas Don’t like to deviate from prescribed working method

Learner centred teaching

-.028

**.136

.034

**.108 .002 **.093

-.072 -.058 -.074 -.062 -.038 **-.139

-.041

.078

.035

.047 .015 .029

.014 .010 -.011 .050 -.014 *-.086

**.079

**.197

—

— **-.095 —

-.001 *.077 .060 *.137 — —

—

**.324

**.154

— — *.089

**-.179 -.013 **.166 .041 *.102 **-.208

Professional learning communities

Trans-formational leadership

Teacher learning

Trans-formational leadership

Professional learning communities

Model 3 (only significant paths of antecedent variables)

Model 2 (all paths of antecedent variables)

Table E1 Standardized path coefficients from antecedent variables to core variables in models 2 and 3.

—

**.128

—

**.099 — *.076

— — — — — **-.148

Teacher learning

—

—

—

— — —

— — — — — **-.096

Learner centred teaching

H. Luyten, M. Bazo

Studies in Educational Evaluation 60 (2019) 14–31

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo

Appendix F Table F1 Table F1 Factor loadings measurement model (model 3).

Transformational leadership Vision & goals Culture Structure Individualized support Performance expectations Intellectual stimulation Professional learning communities Policy & evaluation Consultation & cooperation Consensus Decision making

Loading

S.E.

**.812 **.795 **.867 **.851 **.725 **.914

.025 .031 .020 .016 .031 .016

**.760 **.833 **.788 **.755

.027 .021 .023 .027

**loading is significant at the .01 level (two-tailed).

achievement – A meta-analysis. School Effectiveness and School Improvement, 22(2), 121–148. Marks, H. M., & Printy, S. M. (2003). Principal leadership and school performance: An integration of transformational and instructional leadership. Educational Administration Quarterly, 39(3), 370–397. Marsh, C. J., & Willis, G. (1999). Curriculum: Alternative approaches, ongoing issues (Second edition). Upper Saddle River, N.J: Prentice-Hall International. Middlewood, D., & Burton, N. (2001). Managing the curriculum. London: Sage Publications. Ministério da Educação (1998). Education sector strategic plan 1999-2003 “Reviving schools and expanding opportunities”. [Online] Available:http://www.mined.gov.mz. Ministério da Educação (2004). Education sector strategic plan II (ESSP II). 2005–2009. Ministério da Educação e Cultura (2006). Plano Estratégico de Educação e Cultura 20062010/11 “Fazer da escola um polo de desenvolvimento consolidando a Moçambicanidade”. Maputo: MEC. Moolenaar, N. M., Daly, A. J., & Sleegers, P. J. C. (2010). Occupying the principal position: Examining relationships between transformational leadership, social network position, and schools’ innovative climate. Educational Administration Quarterly, 46(5), 623–670. Moolenaar, N. M., & Thoonen, E. E. J. (2009). Teachers and principals (dis)agree: Differences in perceptions of transformational leadership in relation to organizational functioning. Paper Presented at the 13th Biennial Conference of the European Association for Research on Learning and Instruction (EARLI). Moynihan, D. P., Pandey, S. K., & Wright, B. E. (2012). Setting the table: How transformational leadership fosters performance information use. Journal of Public Administration Research and Theory, 22(1), 143–164. Mulford, B. (2003). School leaders: Changing roles and impact on teacher and school effectiveness. Paper Commissioned by the Educational and Training Policy Division, OECD, for the Activity Attracting, Developing and Retaining Effective Teachers. Mulford, W., Silins, H., & Leithwood, K. A. (2004). Educational leadership for organizational learning and improved student outcomes. Dordrecht/Boston/London: Kluwer Academic Publishers. Muthén, L. K., & Muthén, B. O. (2007). Mplus user's guide (Sixth edition). Los Angeles, CA: Muthén & Muthén. Nguni, S. C. (2005). Transformational leadership in Tanzanian education. A study of the effects of transformational leadership on teachers’ job satisfaction, organizational commitment and organizational citizenship in tanzanian primary and secondary schools. Nijmegen: Drukkerij Quickprint BV. Nguni, S., Sleegers, P., & Denessen, E. (2006). Transformational and transactional leadership effects on teachers’ job satisfaction, organisational commitment, and organisational citizenship behavior in primary schools: The Tanzanian case. School Effectiveness and School Improvement, 17(2), 145–147. Nhavoto, A., Buendia, M., & Bazo, M. (2009). Direcção E Gestão De Escolas: Promovendo Processos De Mudança E Formação De Direcções De Escola. Maputo: MEC. Parise, L. M., & Spillane, J. P. (2010). Teacher learning and instructional change: How formal and on-the-Job learning opportunities predict change in elementary school teachers’ practice. The Elementary School Journal, 110(3), 323–346. Robinson, V. M. J., Lloyd, C. A., & Rowe, K. J. (2008). The impact of leadership on student outcomes: And analysis of the differential effects of leadership types. Educational Administration Quarterly, 44(5), 635–674. Reframing transformational leadership. In I. M. Saleh, & M. S. Khine (Eds.). New school culture and effectiveness. Rotterdam/Boston/Taipei: Sense Publishers. Scheerens, J., Glas, C., & Thomas, S. M. (2003). Educational evaluation, assessment and monitoring. A systematic approach. Lisse: Swets & Zeitlinger publishers.

References Asparouhov, T., & Muthén, B. (2006). Multilevel modeling of complex survey data. Proceedings of the Joint Statistical Meeting in Seattle (pp. 2718–2726). Bazo, M. (2011). Transformational leadership in Mozambican primary schoolsThesis University of Twente. Enschede. Netherlands: University of Twente Thesis. Beck, T. (1999). A handbook for an Education and Management Development Programme: Guidelines for running and sustaining a cascade model programme for developing the management capability of school leaders. Paris, France: International Institute for Educational Planning. Bossert, S. T., Dwyer, D. C., Rowan, R., & Lee, G. V. (1982). The instructional management role of the principal. Educational Administration Quarterly, 18(3), 34–64. Castaneda, D. I., & Rios, M. F. (2007). From individual learning to organizational learning. The Electronic Journal of Knowledge Management, 5(4), 363–372. Darling-Hammond, L., Chung Wei, R., Andree, A., Richardson, N., & Orphanos, S. (2009). Professional learning in the learning profession: A status report on teacher development in the United States and abroadDallas: National Staff Development Council. Davies, B., & Ellison, L. (Eds.). (1997). School leadership for the 21st century: A competency and knowledge approach. London and New York: Routledge. Devos, G., Hulpia, H., Tuytens, M., & Sinnaeve, I. (2013). Self-other agreement as an alternative perspective of school leadership analysis: An exploratory study. School Effectiveness and School Improvement, 24(3), 296–315. Geijsel, F. P. (2001). School and Innovations: Conditions fostering the implementation of educational innovations. Nijmegen: Nijmegen University Press. Geijsel, F. P., Sleegers, P. J. C., Stoel, R. D., & Kruger, M. L. (2009). The effect of teacher psychological and school organizational and leadership factors on teachers’ professional learning in Dutch schools. The Elementary School Journal, 109(4), 406–427. Hallinger, P., & Heck, R. H. (1996). Reassessing the principal’s role in school effectiveness: A review of empirical research, 1980-1995. Educational Administration Quarterly, 32, 5–44. Hopkins, D., Ainscow, M., & West, M. (1998). School improvement in era of change. School development series. London: Trowbridge: Redwood Books, Wiltshire. Kline, R. B. (2005). Principles and practice of structural equation modelling (second edition). New York: The Guildford Press. Krüger, M. L., Witziers, B., & Sleegers, P. J. C. (2007). The impact of school leadership on school level factors: Validation of a causal model. School Effectiveness and School Improvement, 18(1), 1–20. Kwakman, K. (2003). Factors affecting teachers’ participation in professional learning activities. Teaching and Teacher Education, 19, 149–170. Leithwood, K. A., Jantzi, D., & Steinbach, R. (1999). Changing leadership for changing times. Buckingham: Open University Press. Leithwood, K., & Levin, B. (2005). Assessing school leader and leadership programmes effects on pupil learning (Research Report 662)London, UK: Department for Education. Leithwood, K., Tomlinson, D., & Genge, M. (1996). Transformational school leadership. In K. Leithwood, J. Chapman, D. Corson, P. Hallinger, & A. Hart (Eds.). International handbook of educational leadership (pp. 785–840). Dordrecht: Kluwer Academic Publishers. Leithwood, K., & Sun, J. (2012). The nature and effects of transformational school leadership: A meta-analytic review of unpublished research. Educational Administration Quarterly, 48(3), 378–423. Lomos, C. (2012). Professional community and student achievement. Doctoral DissertationGroningen: University of Groningen. Lomos, C., Hofman, R., & Bosker, R. J. (2011). Professional communities and student

30

Studies in Educational Evaluation 60 (2019) 14–31

H. Luyten, M. Bazo Sebastian, J., & Allensworth, E. (2012). The influence of principal leadership on classroom instruction and student learning: A study of mediated pathways to learning. Educational Administration Quarterly, 48(4), 626–663. Shukla, M. (1999). Competing through knowledge: Building organisation: Architecture of a

Ten Bruggencate, G., Luyten, H., Scheerens, J., & Sleegers, P. (2012). < CT > Modeling the influence of school leaders on student achievement: How can school leaders make a difference? < /CT > Educational Administration Quarterly, 48(4), 699–732. Toole, J. C., & Louis, K. S. (2000). The role of professional learning communities in in-

Learning Organisation. New Delhi: Response (Sage). Silins, H., & Mulford, B. (2002). Schools as learning organisations. The case for system, teacher and student learning. Journal of Educational Administration, 40(5), 425–446. Silins, H. C., Mulford, W. R., & Zarins, S. (2002). Organisational learning and school change. Educational Administration Quarterly, 38(5), 613–642. Sleegers, P., Bolhuis, S., & Geijsel, F. (2005). School improvement within a knowledge economy: Fostering professional learning from a multidimensional perspective. In N. Bascia, A. Cumming, A. Dathow, K. Leithwood, & D. Linvingstone (Eds.). International handbook of educational policy (pp. 527–543). Dordrecht, the Netherlands: Kluwer. Stoll, L., Bolam, R., McMahon, A., Wallace, M., & Thomas, S. (2006). Professional learning communities: A review of the literature. Journal of Educational Change, 7, 221–258. Sun, J., Chen, X., & Zhang, S. (2017). A review of research evidence on the antecedents of transformational leadership. Education Sciences, 7, 15.

ternational education. In K. Leithwood, & P. Hallinger (Eds.). The second international handbook of educational leadership (pp. 245–279). Dordrecht, Netherlands: Kluwer. Valckx, J., Devos, G., & Vanderlinde, R. (2018). Exploring the relationship between professional learning community characteristics in departments, teachers’ professional development, and leadership. Pedagogische Studiën, 95(1), 34–55. Vanblaere, B., & Devos, G. (2018). The role of departmental leadership for professional learning communities. Educational Administration Quarterly, 54(1), 85–114. Visscher, A. J., & Witziers, B. (2004). Subject department as professional communities? British Educational Research Journal, 30(6), 785–800. Yu, H., Leithwood, K., & Jantzi, D. (2002). The effects of transformational leadership on teachers’ commitment to change in Hong Kong. Journal of Educational Administration, 40(4), 368–389.

31