UCLA

The expansion of education systems across Latin America has failed to decrease education inequality. An important mechanism driving educational inequality relates to the distribution of classroom practices. This dissertation studied the distribution of classroom practices in Chile, Colombia, and Mexico, the three Latin American countries taking part in the Teaching and Learning Survey (TALIS) Video Study (TVS). Specifically, I focused on the evidence provided by classroom observations and student survey responses. The overarching research purpose of this study was to further our understanding of how classroom practices are distributed between and within these three Latin American countries, in particular centering on the inequities between the classroom experiences of students from different socioeconomic backgrounds. Specifically, this dissertation centered around the following research questions: first, I analyzed the key similarities and differences between the educational systems, of Chile, Colombia, and Mexico, particularly as it refers to teaching standards and frameworks to evaluate teachers and teaching; second, I explored the factorial structures of measures of classroom practice derived from the TVS observation system and student surveys in Chile, Colombia, and Mexico, and the extent to which this were invariant across contexts; and finally, I investigated the extent to which the distribution of classroom practices (for each measure) related to student, family, teacher, and school characteristics. The findings of this dissertation showed that observation scores of classroom practice were not invariant across countries, but student ratings were, highlighting the complementarity of both measures and fostering the use of multiple measures for the assessment of classroom practices. In addition, observation scores in Latin America were lower than student ratings of classroom practices, especially comparing to averages in the rest of participating jurisdictions in the TVS. Finally, few characteristics were correlated with said scores of classroom practices and these correlations varied by country. However, residual variances in both sets of models remained large, indicating the need for exploring further factors that can explain different scores of classroom practices. The findings of this dissertation provide context for future research that seeks to understand how other measures of student, classroom, and teacher characteristics are related to classroom practices. Additionally, this study provides evidence for the use of both, observation systems and student surveys for the assessment of classroom practices depending on the purpose and intended use of the assessment.