Comparisons of State Test Performances of Public and A Charter School System in Old and New Testing in Texas

Abstract

This study examined a particular charter school system’s (CSS) high school state tests performances in mathematics, reading, and science between 2010 and 2013, and compared it with matched traditional public schools’ (TPS) in Texas. After propensity score matching, 12 CSS schools are compared with 32 matched public schools. Independent samples t-tests were performed to compare schools’ TAKS and STAAR scores in each year. One-way ANOVA was conducted to examine differences across years for each school type and Univariate General Linear Model was used to investigate if school type by year interaction has an effect on scores for each subject over the years. T-tests results revealed that CSS’s scores are significantly higher than TPS’s scores in most of the categories. One-way ANOVA results showed that both TPS and CSS scores dropped significantly in most of the categories during the transition from TAKS to STAAR and univariate GLM analyses yielded a significant interaction effects between school type and year variable for all but 10th grade mathematics.