AN ANALYSIS OF RESIDENTIAL CHOICE AMONG SELF-SETTLED AFGHAN REFUGEES IN PESHAWAR, PAKISTAN

The majority of the world\u27s refugees are resettled according to government schemes, but many refugees choose their own residence. Previous research on refugee self-settlement indicated that personal characteristics (e.g., ethnicity and geographic origin) are a paramount factor in self-settlement. Other research suggested that factors related to the refugee experience (e.g., departure dates, reasons for leaving, political involvement) play a role in refugee settlement. This research examined the influence of personal attributes and refugee-related experiences of self-settled refugees from Afghanistan on residential choices in Peshawar, Pakistan. Personal attributes included (1) geographic origin, (2) ethnicity, (3) education level, and (4) employment background. Refugee-related experiences included (1) departure dates, (2) reasons for leaving, and (3) political involvement. Data were collected (December 1983-December 1984) from 976 households in Peshawar. The eleven discrete areal units into which the refugees had sorted themselves provided the framework for sampling and comparing refugee attributes. The data were computer-analyzed (SPSS Cross-Tabs) to determine if an association existed between refugee-related experiences/personal attributes and residential choice. The analysis indicated that residential choices were influenced by background attributes and/or refugee-related experiences. Residential choices appeared to be most related to shared geographic origins and political orientation. Common geographic origins (sometimes combined with ethnic or sub-ethnic affiliation) seemed to have influenced clustering of persons from the same district (or village) as well as individuals from provinces underrepresented in the whole sample (e.g., northern provinces). In some cases, shared origins appeared to have overcome differences in ethnicity or political orientation. Political involvement was shown to be related to residential choice in areas where residents shared political objectives and membership in the same political parties and/or in areas where residents demonstrated a relationship to a party leader who lived in the neighborhood. Lack of political involvement was demonstrated by residential choice in several peripheral neighborhoods. Status attributes (i.e., education and employment) and conditions of departure (dates of departure and reasons for leaving) did not appear to have been strongly influential. They were useful, however, in providing a finer differentiation beyond the basic criteria of ethnic affiliations, geographic origins, and political involvement