July 4, 2011
Latent Variable Regression 4-Level Hierarchical Model Using Multisite Multiple-Cohorts Longitudinal Data
Authors:
Kilchan Choi
This report explores a new value-added model for monitoring school performance over time. The model estimates three performance indicators: initial status, growth rate, and educational gap parameters across different cohorts. The model shows where each cohort of students within a school starts, how much it gains/grows within a specific time period, and how much the initial gap between initially-low performing students and initially-high performing students is magnified or diminished. Furthermore, these three performance indicators of different cohorts are used to examine the extent to which differences or fluctuations across different cohorts within schools are related to the differences in each schools’ characteristics, which is important as it reveals the possible impact that it may have on school-wide or particular grade level achievement over the years. The model is distinguished from other value-added models by providing a more comprehensive picture of student growth over time and the distribution of student growth across cohorts within a school.
Choi, K. (2011). Latent variable regression 4- level hierarchical model using multisite multiple-cohorts longitudinal data (CRESST Report 801). Los Angeles: University of California, Los Angeles, National Center for Research on Evaluation, Standards, and Student Testing (CRESST).|Choi, K. (2011). Latent variable regression 4- level hierarchical model using multisite multiple-cohorts longitudinal data (CRESST Report 801). Los Angeles: University of California, Los Angeles, National Center for Research on Evaluation, Standards, and Student Testing (CRESST).