Transfer as the Test: Rethinking Program Evaluation for Workforce Retraining Initiatives

Abstract

Workforce retraining programs are routinely evaluated on completion rates, satisfaction surveys, and immediate skill assessments — metrics that are easy to collect and poor predictors of whether participants can perform the target work in actual conditions six months later. This paper argues that the field's dominant evaluation framework is systematically misaligned with the stated purpose of retraining: durable behavior change that transfers to workplace performance. Drawing on Kirkpatrick's four-level model and subsequent critical extensions, the paper proposes a streamlined transfer-focused evaluation protocol designed for programs with limited evaluation budgets and community-based research partnerships. The protocol centers longitudinal outcomes (6- and 12-month supervisor ratings, employment retention, and observed task performance) over proximal outcomes (post-training assessments and self-reported confidence). A pilot application of the protocol in a regional aviation workforce context is described, with attention to the methodological challenges of longitudinal data collection in low-resource settings. ---