January 10th, 2018
Article Review: Methodology
The study uses a meta-analytical approach to review literature to test its hypotheses. The literature sources included online databases and those obtained from researchers who had attended either of the international conferences held in 2007 and 2009 on social dilemmas. The inclusion criteria limited for only studies with adult participants (18 years and above) and those that “compared a reward or punishment condition to a control condition”, but not those in which rewards and punishment were simultaneous incentives (Balliet, Mulder & Van Lange, 2011, p. 597). From such criteria and further exclusion of studies that were devoid of critical statistics to enable calculation of effect size, 76 papers representing 187 effect sizes were included in the meta-analysis.
The researchers conducted various analyses to provide information for evaluating the hypotheses. These included d statistic, t-test and chi-square tests. The d statistic is a measure of effect size that examines “the effect of a manipulated dichotomous variable (incentives vs. control) on a continuous dependent variable (degree of cooperation)” (Balliet, Mulder & Van Lange, 2011, p. 598). Apart from cost and source of incentive (centralized or decentralized), other variables that could influence the effect of the incentive such as cost-to-fine ratio in the studies that used punishment as the incentive, design of study (partner versus stranger), type of dilemma and repetitions of dilemma were included in the analysis. Go to the results and conclusion.