Outcome Measurement in Evaluation of Educational Interventions: State Assessments and the Concept of Alignment

• In Event: 70.066 - Roundtable Session 28
  In Roundtable Session: 70.066-8 - Evaluating Teachers and Students

Abstract

How does one select an outcome measure for impact evaluation of a math/science intervention designed to improve achievement? Evidence shows that investigators use large-scale assessments for experimental and quasi-experimental evaluation of interventions. However, scholars have questioned the validity of large-scale assessments for impact evaluation of educational interventions. We apply a research method based on alignment to uncover problems with the use of large-scale assessments as outcome measures in IES-funded math/science education program research. Analysis of the IES database shows that 53% of projects list a large-scale assessment as a key measure; 47% of those are potentially problematic. We present three examples that illustrate misalignment. The results raise questions about the validity of IES-funded evaluations that require greater attention from leaders.

Author

• Josh Sussman, University of California - Berkeley