Fig. 1.
Regression coefficients: GPT-3.5 Turbo score differences across social groups. This figure presents the regression coefficients that compare the score differences across different social groups by GPT-3.5 Turbo, following the specification in Eq. (1). The bold dashed lines divide the results into three sets depending on different regression specifications. In the first part, we compare the difference in scores between candidates from any minority group (female OR black) and white male candidates (the benchmark “majority” group). In the second part, the comparison is between all females versus all males or between all black versus all white males. In the third part, each minority group is compared with the white male group. The diamonds mark the estimated coefficients, and the dashed lines show the 95% CIs.

Regression coefficients: GPT-3.5 Turbo score differences across social groups. This figure presents the regression coefficients that compare the score differences across different social groups by GPT-3.5 Turbo, following the specification in Eq. (1). The bold dashed lines divide the results into three sets depending on different regression specifications. In the first part, we compare the difference in scores between candidates from any minority group (female OR black) and white male candidates (the benchmark “majority” group). In the second part, the comparison is between all females versus all males or between all black versus all white males. In the third part, each minority group is compared with the white male group. The diamonds mark the estimated coefficients, and the dashed lines show the 95% CIs.

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close