AP Workshop – Rocio Titiunik
DateNovember 23, 2015
Time3:00pm
Location
4357 Bunche Hall
Contact
The widespread availability of voter files has changed the study of participation in American politics. However, many studies calculate voter turnout among registrants, which may lead to invalid causal inferences if registration rates differ between treatment and control groups. We introduce a sensitivity analysis approach that can be used to assess the potential bias induced by this problem, which we call differential registration bias, and illustrate its use with two empirical studies of the effects of voting eligibility on subsequent turnout among young voters. In both cases, we find that eligibility decreases turnout as a proportion of total registration. However, these negative effects are highly sensitive to differential registration bias and, in one study, change signs when we instead calculate turnout as a proportion of the population. These results suggest the need for scholars to provide rigorous sensitivity analysis as part of any study of voter file data.Paper: PDF
Event Details:
Parking | Directions
Please register here: