Stigma-Induced Compositional Selection

Picture this

Imagine an open invitation to a neighborhood cleanup day. If the flyer adds a note saying "There is no shame in doing manual labor," status-conscious teenagers stop attending, whereas older, wealthier neighbors who already hold high social status feel unaffected and sign up. The total count of volunteers stays identical, but the demographic mix in the room completely changes.

What the evidence says

Addressing social stigma shifted the applicant demographic index by 0.174 standard deviations toward higher socioeconomic status (p < 0.05). Professional stigma messaging increased the proportion of wealthy applicants by 62% relative to control (p < 0.01) and currently employed applicants by 28% (p < 0.05).

Who was studied

N = 1,460 youth (Experiment 1) and N = 768 youth (Experiment 2), aged 18–35 in Cairo, Egypt.

How

Randomized Field Experiment evaluated with Chernozhukov et al. (2022) generic machine learning inference and Lasso-based baseline characteristic index regressions comparing control vs. treatment applicants.

What to do

Evaluate program recruitment pilots by tracking demographic shifts across baseline participant subgroups rather than relying exclusively on total enrollment numbers.

From the source

"In both experiments, the stigma treatments deliver applicants that are richer, older, and more likely to be currently working, relative to applicants from the control group. In many cases the differences are large and significant."

Stigma and Take-Up of Labor Market Assistance: Evidence from Two Field Experiments

Tags

• selection bias
• program composition
• heterogeneous treatment effects