Meta-Analysis of Seven Standard Experimental Paradigms Comparing Student to Non-student

Abstract

This study embarked on the very challenging proposition of systematically organizing and classifying an assortment of experimental economics essays pertinent to seven experiments performed with both non-student and student populations. The experiments were the Dictator game, Stag Hunt – Coordination game, Risk Aversion Measurement (as measured by the players type of lottery choice), Trust game, Guessing game, Prudence Measurement, and the Guessing game. This meta-analysis reviewed 126 published and unpublished papers collected from several journals and papers provided by several authors via the Google Groups "Economic Science Association - Experimental Methods Discussion" group. Ultimately, only 39 studies were utilized due to methodological alignment. While some studies showed statistically significant differences between non-students and students as indicated by their respective 95% confidence intervals, the overall random-effects model of each of the seven games showed not to be statistically significant. This study contributes to the literature in three important ways. First, the study generates a comprehensive inventory and review of experiments comparing student to non-student populations for the last four decades;  second, the study points out a possible limitation when combining several studies of the same game, despite following similar protocols, suggesting that compounded contextual complexities might diminish aggregate effects of the individuals’ behavioral responses to the financial incentives, and third, the study indicates that generalizations from one experimental economic study, may not render a solid base for extending statistical extrapolations applicable to the total population since the aggregate effects do not indicate substantial differences.