| The primary purpose of the study was to examine the concurrent validation of the Texas Higher Education Assessment (THEA) and Computerized Adaptive Placement Assessment and Support System (COMPASS). Specifically, relationships between remedial placements on the basis of the COMPASS and THEA reading, English, and mathematics scores were examined, using a series of Kappa Coefficient analysis. The on-probability sample consisted of students who had taken both the THEA and COMPASS within 90 days from each other between 2000 and 2006 (n = 253). Results showed lack of agreements between the two sets of placements. Generally speaking, it was found that for the purpose of placement, the COMPASS is a more conservative test than is the THEA and refers more students to remedial classes than does THEA. Thus, concurrent validation could not be established. Additionally, there was no statistically significant relationship between students' high school GPA and their placement status on the basis of any of the THEA and COMPASS reading, English, and mathematics scores. Qualitative data were also gathered and analyzed to complement the quantitative data. Theoretical and practical implications are discussed. |
