A Comparison of Modeling Strategies for Student Ratings of Instruction Data: Application of Simple Hierarchical Model and Cross-Classified Model

• In Event: Roundtable Session 10
  In Roundtable Session: Methods for Understanding the College Student Experience

Abstract

The results of student ratings of instruction are widely used as a way of appraising institutional accountability and improving lecturers’ teaching competencies. In recent years, hierarchical linear model (HLM) has been employed as an appropriate method. However, it is less likely that student ratings of instruction data are always strictly nested within courses, given that individual students take and rate several courses. Rather, the data have a cross-classified structure, in which the ratings are nested within courses and concurrently within students. Therefore, the purpose of the current study is to compare the results of fitting two-level HLM and fitting cross-classified model. This study will demonstrate the importance of employing a cross-classified model to student ratings of instruction data.

Authors

• Hyo Jeong Shin, University of California - Berkeley

• In-Hee Choi, University of California - Berkeley