Nature and Impact of Reference Group Effects in Personality Assessment Data

Abstract

People have a tendency to engage in social comparisons when evaluating and reporting on personality. This tendency and variation in who people compare their personality to is known as the reference group effect and has been largely discussed in cross-cultural research. However, reference group effects have implications beyond cross-cultural research and should be considered when collecting and interpreting personality data. In the present study, we examined the nature and impact of reference group comparisons on the Big Five personality traits in a sample of N = 1194 participants. Specifically, we examined what reference groups participants most believed they compared their personality to, and which reference group was actually the most impactful on trait scores. We found that most people believed they compared their personality to people in general. However, the most influential reference group was people the same age as the participants. Moreover, we found that people mostly engaged in between- as opposed to within-person comparisons when evaluating their own personality. Overall, our findings highlight that people have relatively little insight into the comparisons they engage in when make judgments on personality. Discussion focuses on theoretical and practical implications of our findings in light of personality assessment data.

Data availability statement

For supplementary material, data that supports these findings, and its corresponding reproducible code, see https://osf.io/qm5sd/.

Open Scholarship

This article has earned the Center for Open Science badges for Open Data, Open Materials and Preregistered through Open Practices Disclosure. The data and materials are openly accessible at https://osf.io/qm5sd/, https://osf.io/qm5sd/ and https://osf.io/6nav5.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 We additionally examined predictors of primary reference group choice using participants’ unprompted Big Five trait scores, age, and gender. However, only age emerged as a significant predictor (see Tables S1-S4).

2 This is a deviation from our original analysis plan as we did not pre-register it, and as such is exploratory. We thank a reviewer for suggesting this analysis.

Additional information

Funding

This work was supported in part by a Student Research Grant from the Society for Personality Assessment.