Unfounded confidence? Gender differences in overconfidence across Toronto adolescents.

Abstract

Most people are overconfident and miscalibrated in their probability estimates. Miscalibration can have severe implications, especially in fields like medicine and finance where professionals make consequential probability judgments, and thus studying this topic has useful applications in the real world. This study examines confidence and calibration across gender and age and is guided by two inquiries: do previous findings in the gender and confidence literature apply in the adolescent age group, and does the effectiveness of internal calibration vary across Toronto adolescent males and females? A one-group pretest-posttest experiment was conducted on a sample of 70 Toronto high school students aged 14-18, with the treatment being an internal calibration method that corrects for miscalibration. There are three central findings: both male and female adolescents were under-confident, there is no statistically significant difference between male and female confidence, and internal calibration was effective for male but not female participants.

Keywords

• confidence
• calibration
• gender
• adolescents