Predictive Validity Evidence for Machine-Scored Diagnostic Writing Assessment

Saturday, August 6th, Minneapolis, MN

Elie ChingYen Yu, Oxana Rosca, Heidi L. Andrade

Abstract The Diagnostic Assessment and Achievement of College Skills (DAACS) is a free, online system that assesses incoming college student’s readiness in reading, writing, mathematics, and self-regulated learning and provides feedback and resources. The purpose of this study is to examine evidence of predictive validity for the DAACS writing assessment by determining whether the assessment that students took during orientation predicts their future writing course performance. We collected data from 2722 non-traditional college students (mean age is 33.7, SD= 9.0) at an online institution and 196 traditional college students (mean age is 17.7, SD = 0.5) from a brick-and-mortar university in the U.S.A. Our results suggest there is predictive validity evidence for the DAACS writing assessment. The contribution of this study lies in the uniqueness of the writing assessment and the inclusion of a diverse college population.

View the paper and poster here: https://www.sciencedirect.com/science/article/abs/pii/S1075293525000650