Tuesday, November 28, 2023
2:00 - 4:00 PM (PST) at Berkeley Way West 1204 and via Zoom
Open to GSE faculty, students, and community.
Request a zoom link from convenors@bear.berkeley.edu
Abstract:
MI is a prerequisite for meaningful comparison of group means and standard deviations of latent constructs; however, studies in developing economies rarely test for this psychometric property (Bertling, 2021). To examine the consistency between approaches, we test for MI using Multigroup Confirmatory Factor Analysis (MGCFA) and IRT-modeling. We report analyses based on indices of the magnitude (i.e. modification indices from the MGCFA and group-specific item fit statistics in IRT) and direction of local misfit.
This project is part of an ongoing research on an innovative secondary education program in rural Honduras designed to provide professional coaching for teachers in grades 7-12. Working with research teams both in Berkeley and NYU, we piloted the effects of an intervention to improve teacher quality through classroom observations and professional coaching. Baseline data was collected in March 2020 (pre-pandemic) and March 2022 (post-pandemic). Endline data was collected in Fall 2022.
About the speaker:
Enrique Valencia López is a PhD Candidate in the Policy, Politics, and Leadership cluster at the School of Education. Enrique’s work combines sociological theory, econometrics, natural language processing tools, and educational policy frameworks to examine how educational institutions shape social identities of ethnic minorities and the extent to which these social identities, and their measurement, affect estimates of labor market integration. He has previously been a Deputy Director in Mexico’s educational evaluation agency, a research assistant for UNDP China, and an UNESCO GEM Fellow.