Strengths and limitations of the survey approach for measuring and understanding fertility change: alternative possibilities.

Caldwell JC
In: Reproductive change in developing countries: insights from the World Fertility Survey, edited by John Cleland and John Hobcraft, in collaboration with Betzy Dinesen. London, England, Oxford University Press, 1985. 45-63.

The World Fertility Survey's (WFS) greatest achievements were to set high standards for fertility surveys in developing countries and to measure marriazges, pregnancies, births, contraception, and lactation. A major limitation of surveys is that their questions are usually passed down rather than generated by grass roots research. For example, WFS's African surveys initially confused lactation and deliberate postpartum sexual abstinence. Other problems with WFS questions include too few and too quick questions being asked about difficult subjects, and forced answers. The author believes that abstinence and withdrawal could be shown to be known worldwide if the right questions were asked, whereas WFS data indicates that only 13% of Indonesians and 5% of Nepalese have heard of abstinence. WFS and other fertility surveys only question women and do not deal with kinship, family, couples, or community. Interviewing only women in childbearing ages really limits data collection on traditional means of contraception and their former practice by older women. International surveys collect minimal socioeconomic data because of its lack of comparability. The author recommends more studies of the causes of fertility decline in smaller, more limited subnational samples. The author describes some in-depth community surveys, which required residence of several months to a year or more to complete. This micro approach used principal investigators with male-female teams of interviewers and made it much easier to measure change. Large-scale surveys have great difficulty measuring change because older people are not usually interviewed. The author compares micro and macro studies. Micro studies do not improve on macro studies for basic demographic data. The obvious difference appears in studies of attitudes, causes, and processes, especially when larger surveys have asked only a few questions. Surveys should be based on substantial micro fieldwork and need to be carried out by investigators very familiar with the questions being studied. The best micro/macro mix links a large survey with an intensive study for a subsample of villages. This type of research is specifically designed to measure social and demographic change.

Document Number: 
Add to my documents. Add to My Documents