A rasch analysis of three measures of teacher perception generated from the school and staffing survey

Abstract

The National Center for Educational Statistics' 1999-2000 Schools and Staffing Survey data are used extensively by researchers conducting secondary analysis on a variety of issues including teacher quality, teacher preparation, and the use of technology in the classroom. Researchers frequently combine the data from several related survey questions to create scaled variables, but little is known about the psychometric quality of these created variables. In this article, a Rasch analysis is conducted on three types of teacher perception: perception of influence, perception of students, and school climate. Measures for each type of teacher perception are scaled using a Rasch Rating Scale Model. The quality of these measures for the three scales is then examined. Evaluations focus on six characteristics of each instrument: dimensionality, internal consistency, rating category effectiveness, item quality, item hierarchy, and measure quality. Overall, the three scaled variables exhibit acceptable quality and are appropriate for use in secondary data analysis.