iiinova — where your digital self keeps evolving

Louis Lippens

@louislippens.be

241 followers 607 following 30 posts

Postdoc and lecturer. UGent, FWO, IZA. Member Flemish PES Ethics Board. Labour economics, discrimination, experiments, meta-analysis. louislippens.be

Posts Replies Media

Louis Lippens @louislippens.be · 10/10/2025

The short answer: I apply UWLS via {fixest} and then avg_predictions(). Outcomes are CATEs in the form of log risk ratios. Regressors are study variables. Looks something like the code attached. (wlsmra() is a custom function that builds on fixest::feols().)

120

Louis Lippens @louislippens.be · 04/03/2025

Last week, @iza.org World of Labor published our policy article on hiring discrimination across vulnerable groups. In this article, we summarise insights from meta-research on the issue.

Figure 1 from the IZA World of Labor policy article displaying hiring discrimination estimates for the publication period 2005 until 2020. The graph shows differences in employer callbacks for various vulnerable groups compared to majority reference groups. The x-axis represents the percentage difference in callbacks. Negative values indicate discrimination (fewer callbacks for the vulnerable group). Each group is represented by a square (point estimate) and a ribbon (95% confidence interval). People with disabilities are discriminated against most, followed by less physically attractive people, older people, lesbian women and gay men, and people of a minority race, ethnicity, or national origin.

1103