ABSTRACT
Conducting research on the support for extreme-right political parties that uses urban subarea data has considerable potential as a methodology for offering certain types of explanation. However, such research is often faced by real-life limitations of data availability; these may include lack of voting data for extreme-right parties at a particular relevant time, that the population sizes of urban subareas for which voting data and aggregate social and economic data may be available often differ significantly between cities in different countries or within countries, that social and economic data may be available but not voting data at the same subarea level, and even that the sets of social and economic data that are available and suitable for comparative analyses may differ in their inclusions or in the times to which they refer. Other complications exist, such as the frequent lack of a theoretical basis for deciding on appropriate measures of change over time or on time lags between an explanatory variable and subsequent aggregate voting behaviour. Notwithstanding these drawbacks and with appropriate caveats, an analysis of voting for extreme-right parties in nine cities in four countries, variously between 1975 and 1989, shows that there are differences in the intra-urban variation of their support (reflecting the different relevance of neighbourhood or community factors as opposed to city-wide ones), and also differences in the strength of subarea correlations with resident foreigner populations (reflecting the different extent of proximity-based racism). Such analyses show that, though there are indeed factors that predispose uniformly towards extreme-right support, there are also strong elements of specificity in different environments.
