Body Type Matters: Evidence from Machine Learning on the Relationship between Body Type and Income

Release date: July 30, 2021

https://doi.org/10.1371/journal.pone.0254785

Explanation

The relationship between physical appearance and income has received a lot of attention in the social sciences, and most research to date has measured and examined appearance using classical proxies based on survey-based subjective opinions.

In this study, a new data set called CAESAR is used and analyzed.

This new data includes three-dimensional (3D) full-body scans, which have been found to reduce the possibility of reporting and measurement errors.

These allow us to estimate the relationship between body shape and family income. The possibility of endogenous body shape problems was also taken into account and analyzed using proxy variables and control functions.

These estimation results show that there is a statistically significant relationship between physical appearance and family income. We find that these associations vary by gender, supporting our hypotheses about the premium of physical attractiveness in labor market outcomes and its heterogeneity by gender.