How does a person’s childhood socioeconomic status (SES) affect their chances of participating and succeeding in science? To investigate this question, we use machine learning methods to link scientists in the most comprehensive biographical dictionary, American Men of Science (1921), to their childhood homes in the US Census and literature. First, we show that children from low-SES homes were already significantly underrepresented in the early 1900s. Second, we find that SES influences peer recognition, even conditional on participation: Scientists from high SES families are 38% more likely to be stars, controlling for age, publications, and disciplines. Using live-in servants as an alternative to SES confirms a strong link between childhood SES and stardom. Using textual analysis to assign scientists to subjects, we find that mathematics is the only discipline in which SES influences prominence in the number and quality of scientific publications. Using detailed data on job titles to distinguish academics from industry scientists, we find that industry scientists have a lower probability of becoming stars. Controlling for employment in the industry also strengthens the link between childhood SES and popularity. Elite undergraduate degrees explain more of the association between SES and STAR than any other control. At the same time, controls for birth order, family size, foreign-born parents, mother’s education, ownership rights, and available constellations leave the estimates unchanged, highlighting the importance of SES.
That’s according to a new NBER working paper by Anna Airoldi and Petra Moser.
Posts What predicts success in science? appeared first on Marginal REVOLUTION.
Source link