Replicating Experimental Studies: Moving Beyond Undergraduate Student Samples

University Of Maryland, College Park, College Park MD

Investigators

Abstract

Social scientists have in recent years become increasingly concerned about the extent to which results of experimental studies can be replicated. Although replication is a hallmark of science, social scientists devote relatively little time and attention to replication efforts. Additionally, in the cases when social scientists do attempt strict methodological replications of experimental studies, it is nearly always a case of replicating studies carried out with undergraduate students as participants with studies that have new samples of undergraduate students. Experimentalists rarely pursue methodological replications with samples not comprised of undergraduates. The investigator will attempt to replicate findings in three widely cited experimental studies using a sample of older adults, but also makes new predictions for each study. This research proposes to replicate three important experimental studies from social psychology, each with a focus on gender norms and/or gendered behavior. The replications will be completed with a sample of older adults in an age-restricted retirement community as participants. One of the studies to be replicated is Shelley Correll et al.'s (2007) study on the motherhood penalty, which found that undergraduate students evaluated mothers who were job candidates more negatively than women without children, whereas they evaluated fathers more positively than men without children. A second is Brent Simpson?s (2003) paper which found that women responded relatively more to fear, and men relatively more to greed, when participating in a social dilemma. The third is Rob Willer et al.'s (2013) paper which found that men who had their masculinity threated responded with extreme displays of masculinity, whereas women who had their femininity threatened did not respond with femininity displays. There are a number of reasons to expect that older adults might behave differently in these studies than did college students, although we expect them to behave in some of the same ways (e.g., penalizing mothers relative to non-mothers in the labor market) as well. The results of the studies should advance science in fundamental ways, pursuing an activity that is frequently acknowledged as fundamental to the advancement of science but rarely pursued. Additionally, each replication has the capacity to produce results that are of theoretical and intellectual interest beyond simply determining whether the original results hold in a new sample. The proposed studies also have a significant training component, leading to intensive training in research methods in an innovative setting for three graduate students.