Predictive validity evidence for a no-stakes, untimed, machine-scored diagnostic writing assessment

Authors: Elie ChingYen Yu, Oxana Rosca, Heidi L. Andrade, Angela M. Lui, and Jason Bryer

Abstract

The Diagnostic Assessment and Achievement of College Skills (DAACS) is a free, online system for assessing and promoting newly enrolled college students’ readiness in reading, writing, mathematics, and self-regulated learning. The system provides instant, personalized feedback and resources to promote success in college. In contrast with pre-entry, high-stakes, time-limited standardized testing, DAACS is a diagnostic assessment allowing unlimited time to write about a personally relevant, information-based topic in a free, online, no-stakes context. In search of predictive validity evidence, this study examines whether writing assessment scores predict students’ future performance in college academic writing courses. Data from 2719 non-traditional students at an online university support the hypothesis that writing assessment scores predict future performance. The study contributes to the validity evidence for the interpretations of students’ DAACS writing assessment results. In addition, it indicates that a writing assessment does not have to be expensive, high-stakes, proctored, or timed to be useful.

Download full article here: https://www.sciencedirect.com/science/article/abs/pii/S1075293525000650