Quasi-experimental study designs series – Paper 10: Synthesizing evidence for effects collected from quasi-experimental studies presents surmountable challenges

Betsy Jane Becker, Ariel M. Aloe, Maren Duvendack, T.D. Stanley, Jeffrey C. Valentine, Atle Fretheim, Peter Tugwell

Research output: Contribution to journalArticle

10 Citations (Scopus)
114 Downloads (Pure)

Abstract

Objective: To outline issues of importance to analytic approaches to the synthesis of quasi-experiments (QEs), and to provide a statistical model for use in analysis.

Study Design and Setting: We drew on the literatures of statistics, epidemiology, and social-science methodology to outline methods for synthesis of QE studies. The design and conduct of quasi-experiments, effect sizes from QEs, and moderator variables for the analysis of those effect sizes were discussed.

Results: Biases, confounding, design complexities and comparisons across designs offer serious challenges to syntheses of QEs. Key components of meta-analyses of QEs were identified, including the aspects of QE study design to be coded and analyzed. Of utmost importance are the design and statistical controls implemented in the QEs. Such controls and any potential sources of bias and confounding must be modeled in analyses, along with aspects of the interventions and populations studied. Because of such controls, effect sizes from QEs are more complex than those from randomized experiments. A statistical meta-regression model that incorporates important features of the QEs under review was presented.

Conclusion: Meta-analyses of quasi-experiments provide particular challenges, but thorough coding of intervention characteristics and study methods, along with careful analysis, should allow for sound inferences.
Original languageEnglish
Pages (from-to)84-91
Number of pages8
JournalJournal of Clinical Epidemiology
Volume89
Early online date30 Mar 2017
DOIs
Publication statusPublished - Sep 2017

Keywords

  • Meta-analysis
  • quasi-experiment
  • effect size
  • risk-of-bias
  • moderator variables
  • confounding

Cite this