To be sure, the most visible red vs. blue differences in party support and voter turnout can largely be explained by variations in the composition of local populations. Income, race and ethnicity, education, generation or age cohort, and religious confession, have all been leading contenders in recent years. But the geographic clustering of like-minded voters may indicate far more than the presence or absence of particular social groups at these locations.
Two groups of blue-collar workers with similar background traits, but from two different geographic settings, may vary widely in their political behavior. These differences may be accounted for by local effects, sometimes called “neighborhood effects,” whereby otherwise disparate voters act similarly because they live at the same location and have therefore been influenced by the same local forces, including each other.
In the practical world of campaign politics, much is at stake in the clear understanding of the geographic distribution of voter support.
Both federalism and the Electoral College turn states and regions into distinct electoral battlegrounds, each with their own political, economic and demographic characteristics, or “cultures.” To understand how to win South Carolina is not to understand the key to victory in Minnesota. Campaign strategists understand this, of course, and spend considerable effort analyzing geographic variations in party support in order to target their resources to best effect.
The electoral of the 2004 election followed the pattern of 2000. The Southern states went heavily Republican. President Bush did poorly on the Pacific Coast, losing Washington, Oregon and California. Kerry won in the Upper Midwest: Minnesota, Illinois, Michigan and Wisconsin, and also across New England. Kerry slightly edged Bush in Pennsylvania, by 144,000 votes. Bush won Florida much more comfortably in 2004, by 381,000 votes.
While states must be targeted for purposes of the Electoral College, major campaigns realize that within states there is critical variation that must be explored and understood. Most observers now recognize that the “red” vs. “blue” Election night maps really mask an urban- rural divide within states — a gap that has increased in recent years according to a range of definitions. In 2004, the difference in Democratic support between the most and least populous counties in the nation exceeded 25 points.
Party strategists are alarmed by this widening urban-rural gap, particularly on the Democratic side. Although on their own, the nation’s tiniest burgs do not amount to much, collectively they do cast enough votes to anchor the Electoral College to the Republican candidate in many states. In the wake of the 2004 election, newly elected DNC chair Howard Dean is said to have urged his party’s elites to study and address their “rural problem.”
Many political geographers are involved in the study of redistricting, and this is often what non-experts think of first when they think about the intersection of political science and geography. Geographic dispersion or concentration is but one characteristic of an electorate, but it is a very important one in a nation where political representation is geographically based. As important as it is, political geographers working within the field today have moved far beyond the study of redistricting.
Social scientists interested in change over time are examining how geographies have morphed, tracing these alterations to a series of sources, including migration, immigration, generational turnover, the surge and decline of who is mobilized to vote, and, less frequently, political conversion.
Survey researchers are busy examining voters’ local interactions with other voters within and across neighborhoods to understand the micro-foundations of the geographic patterns we observe on maps. Students of campaign politics are at work scrutinizing the decisions candidates make in targeting some locations for visits and advertising buys while ignoring others.
There’s an interesting diffusion across nearby jurisdictions that suggest processes of geographic influence are at work. New methods for conducting geographically-related research are being developed across several social science fields, greatly aided by advances in computer and software technology. It is fair to say that the study of political geography is only just beginning, with new insights and understandings mounting.
Author: James Gimpel, University of Maryland-College Park