Forecasting Long-Run Effects of College Interventions Using Short-Term Impacts

• In Event: AERA Roundtable Session Thursday 9:45 am - Gold 1
  In Roundtable Session: Roundtable 1

Abstract

When considering the low transfer and credential attainment rate of community college students, it is imperative that interventions be able to forecast long-run impacts early in order to change intervention components. Interventions may use correlational estimates to forecast students’ long-term outcome based on short-term impacts on indicators like credit accumulation and GPA. Such correlational estimates may under- or over-estimate students’ likelihood of graduating or transferring to a four-year college. We use a meta-analytic dataset of community college RCTs to determine the extent to which short-term indicators predict long-run impacts. Such findings may be useful for intervention developers eager to determine the likelihood their intervention will impact graduation and transfer based on early indicators of student success.

Authors

• Mindy Lachs Rosengarten, Teachers College, Columbia University

• Drew Bailey, University of California - Irvine

• Michael Joseph Weiss, MDRC

• Tyler Watts, Teachers College, Columbia University