Table of Contents

  • The eagerness to continually improve the educational experience of students has been growing steadily around the world. We are now more aware of how teaching practices help shape the student learning experience and advance motivation and achievement. When teachers work well together they tend to also work well with students. So, it has become important to encourage teachers to share more of their expertise and experience and in ways that go beyond the mere exchange of information.

  • The pressure to increase equity and improve educational outcomes for students is growing around the world. Teaching practices, in contrast to student background variables such as socio-economic status and cultural capital, are factors affecting student learning that are more readily modifiable. Moreover, additional professional practices have received attention, especially those that help transform the school into a professional learning community.

  • This chapter introduces the premises and vocabulary needed to understand and interpret the report. It sets forth what the TALIS 2008 has ascertained about teaching practices and teachers’ participation in professional learning communities. Chapter 1 also states what the TALIS study was unable to measure; for example, the cause-and-effect relationship between teachers’ level of motivation and their participation in extracurricular learning activities. Chapter 1 indicates that country-by-country profiles will further develop the TALIS findings.

  • This chapter reviews Western educational theory from the 20th century, and concludes that teaching practices can be developed through professional development and appraisal from colleagues. Central features of professional learning communities are co-operation, shared vision, a focus on learning, reflective inquiry and de-privatisation of practice. A small school size, high autonomy, a school management that feels responsible for improving instruction, and a constructive feedback culture also help develop a professional learning community. Chapter 2 notes variations within the West and contrasts, for example, Asian countries where Confucian thought has long promoted collective thinking and action.

  • Chapter 3 details the factors that this report studied. Drawing on socioconstructivist approaches to teaching practices and professional learning communities, discussed in Chapter 2, questions drawn from the TALIS 2008 teacher and school questionnaires were chosen for further analyses. The main themes covered by TALIS 2008: the teachers’ professional development, the type of appraisal and feedback they receive, their activities and attitudes, and schools’ leadership and management styles. The aim was to characterise the underlying profiles of teaching practices and participation in learning communities in each of the 23 countries that participated in TALIS 2008. For this purpose, multilevel latent profile analysis was applied.

  • Chapter 4 provides country-specific results for classroom teaching practices. Specifically, scales are developed and compared for structuring activities, student orientation and enhanced activities. At the teacher level, some relevant variables are the level of education, participation in professional development, gender, subject matter taught, and beliefs about teaching and learning. At the school level, variables include school size, average hours of work, autonomy in curriculum and in hiring, parents’ educational level, administrative and academic leadership style and percentage of teachers reporting appraisal for innovative teaching.

  • As in Chapter 4, profile patterns emerge here among the countries studied, but this time with regards to teachers’ participation in professional learning communities. One general pattern is defined mostly by a separation on teachers’ participation in collaborative activities involving joint teaching. Another shows separation in both joint teaching collaboration and the de-privatisation of practice. A third pattern shows a separation in joint teaching collaborations, de-privatisation of practice and, to a lesser degree, reflective inquiry.

  • This chapter summarises the findings and policy implications of this research. Although the patterns of teaching practices and participation in professional practice are strongly influenced by the specific interaction between traditions, culture and educational policy in each education system. Across systems, however, it is clear that high-quality instruction must surpass teacher-centred instruction: its vocation is to stimulate and challenge students. Student motivation is enhanced by both autonomy and social relatedness, as well as structured teaching and good classroom management. This report suggests that the main driver for advancement in teachers’ professional practices lies with developing a large repertoire of classroom teaching practices and granting autonomy and isolation to co-operatively reflect pedagogical practice.

  • Latent profile analysis (LPA) is derived from conventional latent class analysis, originally introduced by Lazarsfeld and Henry (1968) for the purposes of deriving latent attitude variables from responses to dichotomous survey items. Important contributions to latent class analysis have been made by Clogg (1995). For a review, see Magidson and Vermunt (2004) and Kaplan, Kim, and Kim (2009). In a traditional latent class analysis, it is assumed that an individual belongs to one and only one latent class, and that – given an individual’s latent class membership – the observed responses are independent of one another (referred to as the assumption of local independence). The latent classes arise from the patterns of response frequencies to categorical items, where the response frequencies play a role similar to that of the correlation matrix in factor analysis (Lanza, Collins, Lemmon and Schafer, 2007). The analogue of factor loadings are parameters that estimate the probability of a particular response on the manifest indicators given membership in the latent class. Unlike continuous latent variables (i.e. factors), categorical latent variables (latent classes) divide individuals into mutually exclusive and exhaustive groups.

  • For the multinomial regressions, the results are presented in terms of odds ratios. A straight-forward way to interpret the odds ratios is in terms of percentages above or below 1.0 (even odds).1 For example, if the effect of FEMALE on membership in Profile B vs. Profile A has an odds ratio of 1.50, this would mean that females are 50% more likely to be in Profile B vs. Profile A . Similarly, an odds ratio of 0.65 would mean that females are 35% less likely to be in Profile B vs. Profile A. Profile A is always the profile with the lowest mean scores for most items/scales. This profile is compared with Profiles B and C respectively, which show higher means for most items/scales.