Beyond Predictive Validity: A Mixed Method Study of Self-directed Developmental Education Placement at a Small Community College

Abstract

ABSTRACT Colleges across the country struggle with determining the most appropriate means of placing students into math and English coursework. Most colleges utilize a single high-stakes standardized placement exam to assess incoming students and use strict cut-scores to determine students’ placement. However, the ability of these tests to accurately place students has been called into question. A small, but growing, number of colleges allow students to have more input into their placement through self-directed placement (SDP) policies, which are typically based on a self-assessment tool that asks students to reflect on their content-related practices, abilities, and beliefs or attitudes. This mixed methods research study explored the experiences of a small community college that implemented SDP as part of a larger multiple measures placement process. Data were examined through the lens of Messick’s conceptualization of validity and revealed evidence that both supported and challenged the validity of placement decisions made from SDP. When designing SDP instruments, we recommend that colleges ensure that SDP tools reflect the skills and knowledge that students need to be ready for the college-level courses at each particular institution, rather than relying heavily on SDP instruments developed by other institutions. We also recommend that colleges adopt a holistic approach to students’ self-assessment. Our findings indicate a need to consider a shift in the dominant conversation on placement methods from a focus almost exclusively on predictive validity toward a more comprehensive, and potentially more robust, holistic consideration of qualitative and quantitative validity evidence.