The College of St. Scholastica

Political polls reporting voters� intentions and beliefs are often skewed by the participant�s desire to be seen as having socially acceptable opinions, otherwise known as the social desirability effect (Streb, Burrell, Frederick, & Genovese, in press). This study will use the list experiment technique to investigate this bias when political polls question support for a female presidential candidate. Participants (N = 80; M = 40, F = 40) will fill out an electronic survey, which will include (a) demographic questions, (b) the likelihood that they would vote for a female presidential candidate, (c) the Marlowe-Crowne Social Desirability Scale (M-C SDS), and (d) the list experiment. The list experiment will unobtrusively determine anger towards a woman serving as president by comparing those asked about the issue (test group) and those not (baseline group). It is predicted that the test group will have a higher anger level than the baseline group. The main hypothesis is that those in the test group with a high anger estimate but respond with a high likelihood of voting for a female candidate will also score high on social desirability. In a three-way ANOVA analysis, this effect is expected to be stronger in males since the topic relates to females in power. These results will demonstrate the effects of social desirability on expressed beliefs; the predicted suppression of genuine opinions, to be seen as socially acceptable, will present implications for interpreting self-report responses, specifically in political polls.