Abstract
The Althusserian notion of a problematic refers to the underlying connections among the elements of a body of thought that shape its orientation. It is employed to examine the relationship between the practical decisions governing the actual conduct of attitude surveys and their sociopolitical context. Two sets of surveys, on preferences for political leaders and constitutional arrangements and on Black attitudes to sanctions against apartheid, are analyzed that featured in the transition to democracy in South Africa. Decisions regarding the phases of questionnaire design, fieldwork, analysis, interpretation, and dissemination are more or less explicit. It is found that in conjunction, they may predispose the survey to confirming specific presuppositions. Such a problematic is considered to be ideological. Based on further concrete examples of alternative survey practice, regulative commitments are proposed for each phase to ensure that the problematic of attitude surveys, even in such polarized contexts, may instead be scientific.
Get full access to this article
View all access options for this article.
