Objective. The purpose of this research is to examine the effect of place d
ifferences-city, suburban, and nonmetropolitan-on voting behavior in the U.
S. House of Representatives. Because the information necessary to classify
congressional districts according to their settlement pattern is not regula
rly provided by the Census Bureau, studies of the impact of place on congre
ssional voting behavior continue to rely on improper operationalizations of
place. Methods. We reconstructed the data provided by the Census Bureau in
order to classify districts as central city, suburban, nonmetropolitan, or
mixed. Using multivariate analysis, we then tested the hypothesis that, co
ntrolling for a wide range of other factors, representatives' voting behavi
or varies by type of settlement pattern, and, in particular, that the votin
g behavior of central-city representatives is more liberal than that of non
-central-city representatives. Results. We found that, controlling for legi
slator's party, personal characteristics, region, and constituency characte
ristics, the type of place a legislator represents has a significant and no
ntrivial impact on legislator voting behavior, and that increases in the pe
rcentage of central-city residents in a representative's district are assoc
iated with increases in liberal voting by the representative in the House.
Conclusions. Since living in a central city is significantly associated wit
h greater liberalism on roll-call voting, ceteris paribus, the movement of
population from the cities and the decline in the number of majority centra
l-city districts will reduce liberal voting behavior in the House.