Academic journal article Population

Collecting Community Histories to Study the Determinants of Demographic Behaviour: A Survey in Burkina Faso

Academic journal article Population

Collecting Community Histories to Study the Determinants of Demographic Behaviour: A Survey in Burkina Faso

Article excerpt

Over the past thirty years, many authors have stressed the importance of taking the characteristics of the local environment, or context(1), into account when explaining demographic behaviour. The general idea is that demographic behaviour is not only influenced by individual-level and household factors, but also depends upon the social and economic environment, sanitary conditions, etc. The demographic literature contains numerous examples of potential ways in which the local environment may influence fertility (Schoumaker, 1999; Casterline, 1987; Freedman, 1974), mortality (Frankenberg, 1995; DaVanzo, 1985) or, of greater interest here, migration (Lucas, 1997; Hugo, 1985; Bilsborrow, 1984).

In parallel with this growing interest in the influence of the local context on demographic behaviour, the World Fertility Survey (WFS), from the 1970s onwards, and Demographic and Health Surveys (DHS), from the mid-1980s, started to collect community data in localities where survey clusters for surveys of individuals were located. In total, around one third of the surveys carried out under these programmes involved the collection of community data, resulting in numerous analyses of the determinants of fertility and mortality, incorporating explanatory variables at different levels (Schoumaker, 1999; Tuo, 2002)(2).

However, the community data gathered in these surveys have several weaknesses. Often considered to be of poor quality(3) and primarily concerning health and family planning services, they include practically no retrospective data. While individual-level data on fertility and mortality collected in the WFS and DHS are based on the full birth histories of women, the community data generally only concern the characteristics of the localities at the time of the survey. This considerably limits the scope for carrying out longitudinal analyses incorporating both individual-level and community variables. However, since the mid-1990s, a number of operations to collect retrospective community data have been conducted as part of demographic surveys on developing countries (for Indonesia and Malaysia, see Frankenberg, 2000; for Nepal, see Axinn et al., 1997), paving the way for multi-level longitudinal analyses of demographic behaviour.

Yet even when retrospective community data are collected in these surveys, they only relate to the sample localities selected for the survey of individuals. This means that for individuals who once lived outside the sample area (migrants), there are no suitable community data to explain their behaviour prior to arrival in the survey locality. For analyses concerning mortality or fertility, this problem can to some extent be overcome by restricting analyses to those periods after the last migration for example (Schoumaker, 2001). However, this lack of retrospective data on nonsample areas has more serious consequences for the study of the determinants of migration. If one is interested in the contextual characteristics that prompted individuals to leave their former place of residence, some knowledge of the features of this place in the past is essential. Yet such data are not available in a standard retrospective community-level survey.

One solution to this problem is to conduct a retrospective survey covering not only the localities sampled in a retrospective survey of individuals, but also the former places of residence of the persons surveyed. This potentially involves a much more extensive data collection operation, to the extent that some authors have ruled out an approach of this type (Findley, 1982). Such a survey was nevertheless carried out recently (2002) in Burkina Faso, as part of a research project on migration and urban integration, involving researchers from the Department of Demography at the Université de Montréal, the Institut supérieur des sciences de la population (ISSP, former UERD) at the University of Ouagadougou and the Center for Study and Research on Population and Development (CERPOD) in Bamako. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.