Presentation Title

Conflicting Beliefs about Social Categories

Faculty Mentor

Andrew Shtulman, Andrew Young

Start Date

17-11-2018 10:15 AM

End Date

17-11-2018 10:30 AM

Location

C151

Session

Oral 2

Type of Presentation

Oral Talk

Subject Area

behavioral_social_sciences

Abstract

This study investigated the coexistence of generic and statistical beliefs about social categories. Our research is important for identifying conflict between generic and statistical beliefs and its relationship to stereotype endorsement. The participants were Occidental College students/faculty staff (N=100; 65 females, 34 males, 1 other). A computer-based survey was utilized to assess generic beliefs, statistical beliefs, and social judgements through “true - false” questions, percentage estimates, and likelihood ratings about stereotypes. Stereotypes about four social categories (race, gender, sexual orientation, and political orientation) were used to create various statements. Several of the questions contained statements that created conflict between participants’ generic and statistical beliefs (conflict statements). Results showed that participants were more accurate in their responses to the no conflict questions than the conflict questions in the categories of race, gender, and political orientation, which was used to define conflict score. Results only demonstrated a significant difference in response time between no conflict and conflict questions in the sexual orientation category. Results also showed a correlation between conflict scores and stereotype endorsement, especially on the item level. Therefore, the more conflict participants demonstrated between generic and statistical beliefs about a social group, the more they endorsed stereotypes relevant to that group.

Summary of research results to be presented

Results demonstrated that in the categories of gender, race, and political orientation, a significant difference was found in accuracy between no conflict and conflict statements. Participants were generally more accurate in their responses to the no conflict questions than the conflict questions. Results also demonstrated a significant difference in response time between no conflict and conflict questions in the sexual orientation category, meaning that participants took longer to answer conflict questions than no conflict questions. All four categories demonstrated a significant difference in both mean percentage estimates and mean likelihood ratings between target and nontarget questions. Therefore, participants differed in their response to questions that corresponded with the stereotype (target) and questions that did not (nontarget).

For gender, race, political orientation, and sexual orientation, a higher conflict score was correlated with a higher difference in both percent estimates and likelihood ratings. Conflict score was defined as the difference in accuracy between statements in which generic and statistical beliefs converge and those in which they diverge. As a result, the more conflict participants demonstrated between generic and statistical beliefs about a social group, the more they endorsed stereotypes relevant to that group. Within each category, conflict scores served as a predictor of percentage estimates and likelihood ratings. Right down to the item level, our measure of conflict between generic and statistical beliefs was a powerful item-specific predictor of participants' percentage estimates and likelihood ratings in each category. Correlations between conflict scores and the measures of stereotype endorsement were consistently stronger for the same item than for different items. No effects of one's own social group (gender, race, political orientation, sexual orientation) on any of the outcome variables were found.

This document is currently not available here.

Share

COinS
 
Nov 17th, 10:15 AM Nov 17th, 10:30 AM

Conflicting Beliefs about Social Categories

C151

This study investigated the coexistence of generic and statistical beliefs about social categories. Our research is important for identifying conflict between generic and statistical beliefs and its relationship to stereotype endorsement. The participants were Occidental College students/faculty staff (N=100; 65 females, 34 males, 1 other). A computer-based survey was utilized to assess generic beliefs, statistical beliefs, and social judgements through “true - false” questions, percentage estimates, and likelihood ratings about stereotypes. Stereotypes about four social categories (race, gender, sexual orientation, and political orientation) were used to create various statements. Several of the questions contained statements that created conflict between participants’ generic and statistical beliefs (conflict statements). Results showed that participants were more accurate in their responses to the no conflict questions than the conflict questions in the categories of race, gender, and political orientation, which was used to define conflict score. Results only demonstrated a significant difference in response time between no conflict and conflict questions in the sexual orientation category. Results also showed a correlation between conflict scores and stereotype endorsement, especially on the item level. Therefore, the more conflict participants demonstrated between generic and statistical beliefs about a social group, the more they endorsed stereotypes relevant to that group.