Selecting and Improving Quasi-Experimental Designs in Effectiveness and Implementation Research

By M. Handley, C. Lyles, C. McCulloch, A. Cattamanchi. Annual Review of Public Health, 2018

359 citations recorded by the source corpus.

Interventional researchers face many design challenges when assessing intervention implementation in real-world settings. Intervention implementation requires ‘holding fast’ on internal validity needs while incorporating external validity considerations (such as uptake by diverse sub-populations, acceptability, cost, sustainability). Quasi-experimental designs (QEDs) are increasingly employed to achieve a better balance between internal and external validity. Although these designs are often referred to and summarized in terms of logistical benefits versus threats to internal validity, there is still uncertainty about: (1) how to select from among various QEDs, and (2) strategies to strengthen their internal and external validity. We focus on commonly used QEDs (pre-post designs with non-equivalent control groups, interrupted time series, and stepped wedge designs) and discuss several variants that maximize internal and external validity at the design, execution, and analysis stages.

Subfields: Developmental and Educational Psychology, Economics and Econometrics, General Health Professions

View Paper: https://www.semanticscholar.org/paper/d157a3c8915985c6e1d249fb9987baa1e949109f