Gendered Misclassification from Surname-Based Race Imputation
Organizations: PhD Candidate, Department of Government, Harvard University.
Abstract
Researchers increasingly estimate political behavior based on race and gender from voter files, where race is imputed from surnames and geography. This only works where surnames suggest group membership, an assumption violated where surnames follow norms around marriage. Imputation is therefore biased against women who adopt their husband's surname, especially those who racially inter-marry, an outcome itself endogenous to partisanship. Using six US states recording self-reported race, I show that imputation is less accurate for Asian and Hispanic women, particularly those sharing the surname of a partner of another race, who differ politically from other women of their race. This violates assumptions behind downstream inference approaches, which nearly double the gender gap in partisanship among Asian voters. I derive an expression showing the bias is largest where a group is small. Using maiden rather than marriage-adopted surnames in North Carolina raises accuracy up to fourfold for women married across racial lines. Recovering maiden surnames, or omitting the married surname, removes much of the bias. In a national replication of a study of gender gaps in turnout, the correction more than triples the gap among Asian voters. In Lebanon, where wives register in their husband's locality, the error enters through geography.
Figures & tables
| State | Women with their partner’s surname | Middle name is a surname (%) | Same rule for men (%) |
|---|---|---|---|
| Alabama | 622,083 | 37 | 12 |
| Florida | 2,150,240 | 15 | 6 |
| Georgia | 1,191,761 | 27 | 10 |
| Louisiana | 450,173 | 42 | 6 |
| North Carolina | 1,219,096 | 42 | 12 |
| South Carolina | 674,931 | – | – |
| Classified correctly (%) | ||||
|---|---|---|---|---|
| Registered voters | Name and locality | Name only | Locality only | (thousands) |
| Men | 86.6 | 69.8 | 84.9 | 1,752 |
| Women | 82.6 | 67.1 | 80.1 | 1,785 |
| Women on their own sect’s list | 88.6 | 68.8 | 87.2 | 1,605 |
| Women on another sect’s list | 29.4 | 52.0 | 16.8 | 180 |
Algorithmic Bias
Algorithmic bias studies how automated systems produce or amplify unfair outcomes across demographic groups, and how to measure and reduce them.
See all 108 papers