WP 2021-01 Gender, Confidence, and the Mismeasure of Intelligence, Competitiveness and Literacy

AUTHORS: Glenn W. Harrison, Don Ross and J. Todd Swarthout

ABSTRACT: The measurement of intelligence should identify and measure an individual’s subjective confidence that a response to a test question is correct. Existing measures that use multiple-choice answers are constrained in measuring responses that reflect, at best, only modal beliefs, due to the use of surveys with no extrinsic financial incentive for a truthful response and by not eliciting measures of confidence. We rectify both issues, and show that each matters for the measurement of intelligence. We use a canonical measure of fluid intelligence, the Raven Advanced Progressive Matrices test. We show that awareness of the confidence of responses, and being able to express them in a properly incentivized manner, plays a critical role when interpreting intelligence measures. Despite evidence that women tend to do worse than men in historically cited intelligence measures using the Raven test, they do much better than men when responding to incentivized measures of confidence. We also show that our results on gender and confidence in the face of risk have wider applications in terms of “competitiveness” and financial literacy. It is not the case that women lack the confidence to take on the risks of competition: they respond exactly as any risk averse agent should. When confidence in response to questions about financial literacy is measured correctly, “do not know” responses are not evidence that women are afraid of answering questions about financial knowledge.