Electoral geography: From mapping votes to representing power

Abstract

In some ways, electoral geography has never been more popular. From the detailed, online maps of the Brexit vote to discussions of the electoral college versus the popular vote in the 2016 Trump–Clinton U.S. presidential contest, the relationships among geography, voting, and political power have seldom been more visible. The popularity of electoral geography in social and news media, however, does not necessarily reflect its presence in scholarly discussions, and indeed, in some ways, the former has replaced the latter. Digital technology and the burgeoning availability of electronic data mean that it is easier than ever to create maps of votes, often in near real time. Yet the academic field of electoral geography encompasses more than just mapping votes, including the study of election campaigns, political parties, electoral systems, and gerrymandering. The 3 major approaches are the geography of voting (mapping and visualizing votes), geographic influences on voting (the effect of place on political preferences and behavior), and the geography of representation (the analysis of electoral systems). Indeed, the structure of the electoral system, including gerrymandering, is often the key to understanding how political and racial/ethnic minorities can (or cannot) wield power and influence. This article examines each approach after a brief review of the historical origins of the subfield.

Who are the voters and where are they? Using spatial statistics to analyse voting patterns in the parliamentary elections of the Czech Republic

“…Although Ghitza and Gelman (2013) showed that by means of multilevel models and post-stratification (MPR) we can quite precisely model and map the different behaviour of voters group across 50 American states, this method is not viable in countries with a smaller number of secondlevel units, such as the Czech Republic. Generally, the practice of using polling data is relatively limited among electoral geographers (Forest, 2018).…”

“…The lower the level of the aggregation of data provided by the Statistical Office the better (e.g. precinct data would be superior to municipal level data), but the ultimate disadvantage of this is that we are unable to understand voter behaviour at the individual level, commonly known as the ecological fallacy (see Forest, 2018). We are completely unable to inspect hypotheses that are attitudinal or emotional (Kouba & Lysek, 2019).…”

The People and The Nation: The “Thick” and the “Thin” of Right‐Wing Populism in Canada

“…Forward Sortation Areas (FSA), derived from Canadian postal codes, are used as a proxy for neighborhoods, and Statistics Canada provides census data for these units. Model 2 has been included because, within political geography, scholars have argued that contextual effects have an impact on voter behaviour and opinions (Forest, 2018; Walks, 2006). Model 3 adds opinion‐based variables: self‐reported ideology on a left–right ideological scale, level of dissatisfaction with democracy, belief that the economy has become worse in the last year, and feelings toward racial minorities in general, immigrants, and feminists.…”

Inter‐person Separation: A New Objective Standard for Evaluating the Spatial Fairness of Political Redistricting Plans

“…where k = 1,2, … ,q = number of districts A k = PartyAswasted votes in districtk B k = Party Bs wasted votes in k T = Total votes cast in all district races. The increasing sophistication of partisan districting plans and rising political polarization have been among the reasons for significantly revived academic attention now being paid to electoral geography (Warf and Leib 2011;Forest 2017). The efficiency gap approach is distinctly different from most academic literature on spatial fairness in electoral districting.…”