| Session Title: Psychometric Analysis of Student Performance Comparisons in Evaluation Using Item Response Theory: An Illustration With National Longitudinal Study Data |
| Demonstration Session 752 to be held in Panzacola Section H2 on Saturday, Nov 14, 10:55 AM to 11:40 AM |
| Sponsored by the Quantitative Methods: Theory and Design TIG |
| Presenter(s): |
| Raymond Hart, Georgia State University, rhart@gsu.edu |
| Abstract: Measuring change in variables is common in social science research. Experimental and quasi-experimental studies often employ pretest-posttest or longitudinal time-series designs to compare effects on dependent variables. A number of statistical methods are available for measuring change including: 1) Analysis of Variance (ANOVA) on gain scores, 2) Analysis of Covariance (ANCOVA), 3) ANOVA on residual scores, and 4) Repeated measures ANOVA. These methods are often dependent on raw score differences and statistical comparisons on sample based performance on standardized tests. Traditional sample comparisons capitalize on measurement error and sampling error. The purpose of this demonstration is to illustrate the practical application of a method that uses item parameter estimates from Item Response Theory (IRT) assessments to measure student or group progress longitudinally without the need for comparison groups. Evaluators can use this research to compare student performance and growth across time and across subgroups of students to population performance. |