This report investigates the links between son preference and differential treatment of children by analyzing data from the Demographic and Health Surveys of Morocco and Tunisia, two countries that are thought to vary considerably regarding indicators of gender. The analyses find no significant differences in either country in the duration and intensity of breastfeeding and small differences in favor of boys in Tunisia regarding immunization and the treatment of diarrhea. These findings, which are, to some degree, unexpected, are discussed in light of other research relevant to son preference in the two countries.
PIP: An analysis of Demographic and Health Survey data from Morocco (1987) and Tunisia (1988) failed to document the expected link between son preference and differential treatment of children. Although excess female child mortality and morbidity have been declining in the Middle East, gender-based differentials persist in selected health and nutrition variables. The present analysis used data on breast feeding, immunization, and the treatment of diarrhea for the sample of children in both countries born 5 years before the surveys. A slightly higher proportion of boys than girls breast fed for durations exceeding 18 months in Tunisia, but no pattern of difference by gender was observed for Morocco. In both countries, boys were slightly more likely to be fully immunized than girls (67% versus 64% in Morocco and 88% versus 85% in Tunisia). In Morocco, the proportion of children with diarrhea not receiving either home or medical treatment was slightly higher among girls than boys (46% versus 42%); in Tunisia, 35% of girls but only 31% of boys were untreated. The unexpectedly small magnitude of the sex differences found in this analysis contradicts the social science literature, which emphasizes a pattern of sex discrimination in both these countries. Also surprising was the finding that Tunisia, considered the more egalitarian of the two countries, had stronger son preference than Morocco for indicators of both fertility and health behavior. This calls into question simplistic explanations of the effect of women's status on demographic behavior.