Quarterly Journal of Political Science > Vol 7 > Issue 4

Movers, Stayers, and Registration: Why Age is Correlated with Registration in the U.S.

Stephen Ansolabehere, Professor of Government, Harvard University, USA, sda@gov.harvard.edu , Eitan Hersh, Assistant Professor of Political Science, Yale University, USA, eitan.hersh@yale.edu , Kenneth Shepsle, George D. Markham Professor of Government, Harvard University, USA, kshepsle@iq.harvard.edu
Suggested Citation
Stephen Ansolabehere, Eitan Hersh and Kenneth Shepsle (2012), "Movers, Stayers, and Registration: Why Age is Correlated with Registration in the U.S.", Quarterly Journal of Political Science: Vol. 7: No. 4, pp 333-363. http://dx.doi.org/10.1561/100.00011112

Publication Date: 17 Oct 2012
© 2012 S. Ansolabehere, E. Hersh, and K. Shepsle
Political participation,  Formal modelling,  Electoral institutions,  Voting behavior


Download article
In this article:
Age, Mobility, and Participation 
A Stayer–Mover Model 
Empirical Predictions of the Model 


Age is among the strongest predictors of political participation, yet it is also among the least well understood. We offer a probability model of participation in the U.S. voter registration system — the first step in the voting process. In this model, people have a constant probability of registering to vote at any given time and a constant probability of moving. A strong relationship between age and participation arises simply as a byproduct of the rules of the registration system, namely that participation is voluntary and that it is residentially based. Specifically, the probability that someone is registered increases over time (and thus with age) even when the probability of becoming registered is constant. A new, national random sample of 1.8 million voter registration records is employed to test the model. The model provides a theoretical foundation for the relationship between age and participation, identifies the functional form of that relationship, and solves a puzzle about the nature of participatory bias.