In the late 1990s and early 2000s, international researchers began to examine the
effect of various classroom-level variables and school-level factors on student
achievement including databases produced by the International Association for the
Evaluation of Educational Achievement (IEA) studies such as TIMSS (e.g., Bos and
Kuiper 1999 ; Keys 1999 ; Koller et al. 1999 ). These databases, however, lacked
detailed information on the knowledge and skills that teachers of high-achieving
pupils possessed and information on what characterized their teaching.^8 The results
of the TIMSS Video Study (Stigler and Hiebert 1997 ; Stigler et al. 2000 ) began to
open up the notion that teachers across cultures exhibited substantially different
practices and depths of knowledge that seemed associated with the achievement
levels of their pupils as measured in the TIMSS tests. Until then, the theoretical
framework to guide studies of teacher education had not been based on assessing the
expected knowledge for teaching that future teachers must acquire to teach well.^9
42.3 The Use of Diverse and Increasingly
Sophisticated Methods
Changing conceptions about knowledge, and how it is learned began to influence
the design of education studies. Great advances were made in the conceptualization
of teaching and learning as imbedded in hierarchical school and social contexts
(e.g., Raudenbush and Bryk 1986 ; Riddell 1989 ), and these novel conceptualiza-
tions lead to changes in research design and analytic techniques. But while social
science methods continued to evolve and provide a richer range of methodological
approaches to address the research questions emerging in thefield, research on
teaching and teacher education was still limited. For instance, in a review of the
international literature on teacher education research Tatto ( 2000 ) concluded that
studies that considered teacher education mostly explored teaching effectiveness as
related to pupil learning and included some information concerning teachers’
preparation (e.g., whether or not they had credentials on specific subject matters and
on pedagogy, or whether they had learned“on the job”) but failed to provide
important details on what teachers learned, how, why, and when. Particularly
problematic were the methods used which used such indicators of teacher prepa-
ration as degrees, years of study, and kinds of courses taken but did not directly
measure the actual knowledge teachers had acquired as a result of the experienced
preparation. Most research on the effects of teacher education on teaching and, thus,
(^8) An important reaction to the gap on the research on teacher education was the development of
comparative descriptive studies of teacher education. The research done by the MUSTER project
out of the University of Sussex in a number of African nations merits special recognition. See the
work of Lewin and Stuart ( 2003 ) and other publications on the MUSTER website (http://www.
sussex.ac.uk/education/research/cie/rprojects/muster/pubs).
(^9) For an exception see Fig.42.1showing the resulting program theory and design guiding the Sri
Lanka research by Tatto et al. ( 1993 ).
42 The Role of Comparative and International Research... 625