In some Sub-Saharan African countries where wife beating is widely accepted as a response to women's transgressing gender norms, men find less justification for the practice than do women.1 An analysis of Demographic and Health Survey data from seven countries found that 36-89% of women justified wife beating in at least one of five specified situations; among men, who were included in six of the surveys, the proportions ranged from 25% to 75%. In multivariate analyses, both women's and men's acceptance of wife beating generally declined as income rose and was reduced among those with the most education; other results were less consistent across countries.

The data were collected between 1999 and 2001 in Benin, Ethiopia, Malawi, Mali, Rwanda, Uganda and Zimbabwe, low-income countries with strong patriarchal societies. All seven country surveys included nationally representative samples of 15-49-year-old women; all but Zimbabwe's also included men aged 15-59. Each survey asked respondents if a husband is justified in hitting or beating his wife if she burns the food, neglects the children, argues with him, goes out without telling him or refuses to have sex with him. After assessing levels of acceptance of wife beating in each of these situations, the analysts used multivariate logistic regression to identify background characteristics that were associated with acceptance; separate analyses were conducted for men and women.

Among women, the proportion of respondents who considered wife beating justifiable in at least one situation ranged from 36% (in Malawi) to 89% (in Mali); among men, the range was from 25% (in Malawi) to 75% (in Ethiopia). In general, women and men were least supportive of wife beating as a response to a woman's having burned food or refused sexual relations with her husband; they most often considered the practice acceptable as chastisement for a woman who had neglected her children or gone out without informing her husband. In the six countries where both women and men were interviewed, women tended to consider wife beating justifiable in more circumstances than did men.

In the multivariate analysis, income (as measured by the number of household assets) and level of schooling had the most consistent associations with women's acceptance of wife beating. With each increase in household assets, the odds of accepting wife beating were significantly reduced, by up to 15% in six countries and by 47% in Rwanda. Similarly, in each country, the likelihood of support was significantly, and often dramatically, lower among women with postsecondary education than among those with no schooling (odds ratios, 0.1-0.4); findings for women with primary or secondary schooling, however, did not show a clear pattern across countries.

Other variables were significant in only some countries. Urban residence was associated with reduced odds of accepting wife beating among women in Benin and Malawi; results for Rwanda, Uganda and Zimbabwe were in the same direction but only marginally significant. In four countries, adult women in two or all three 10-year age-groups studied were less likely than teenagers to justify wife beating, but in the other three, relationships were at best marginally significant. Never-married women in all countries except Malawi and Uganda had lower odds of accepting wife beating than did their counterparts in monogamous marriages, and the less decision-making power a woman in Rwanda, Uganda or Zimbabwe had at home, the greater her odds of accepting wife beating. In Benin, Malawi, Rwanda and Zimbabwe, paid employment, unpaid employment or both were associated with an elevated likelihood of support. Women in Uganda had elevated odds of accepting wife beating if they held a nonpaying job, and reduced odds if they were employed for pay.

Two variables—religion and contribution to household expenditures—yielded conflicting results across countries. In Benin and Mali, Muslims had twice the odds of Catholics of accepting wife beating; in Malawi, they had a one-third reduction in odds. And whereas Ugandan and Zimbabwean women who paid more than half of the household's expenses were more likely than those who paid less than half to justify wife beating, their counterparts in Benin had reduced odds of doing so.

For men, as for women, income and education had the strongest associations with acceptance of wife beating. In every country except Rwanda, acceptance declined with increasing household assets (odds ratios, 0.8-0.9); in all but Malawi, support was reduced among those with the highest level of education (0.1-0.6). Compared with 15-9-year-olds, adult Rwandan men in each 10-year age-group had reduced odds of accepting wife beating, as did 30-59-year-olds in Malawi, those aged 40-59 in Ethiopia, and those in their 50s in Benin.

Other variables showed no clear patterns of association. In Malawi, Muslim men had reduced odds of accepting wife beating, but in Benin, they had elevated odds. In Ethiopia, Muslims and Protestants had marginally elevated odds, and adherents of other religions had significantly increased odds. Marital status was significant only in Malawi (odds ratio, 1.8 for polygamous men) and Ethiopia (3.4 for widowed or separated men), and employment was significant only in Uganda (0.5 for those engaged in unpaid work). Men who contributed half of household expenses had elevated odds of justifying wife beating in Benin (1.4) and reduced odds of doing so in Uganda (0.6); the odds of support were reduced for Malian and Ugandan men who contributed more than half of expenses (0.5 for each).

The analysts contend that women's acceptance of wife beating "may be explained only by entrenched social and cultural learning processes that subjugate the position of women in the society, socially and collectively undermine their self-esteem and facilitate romanticisation of the 'ideal' gender role of women." They further argue that the first step toward eliminating this practice is to "build up a substantial amount of momentum" in opposition to the use of violence in conflict resolution and that, given its widespread acceptance in these societies, the development of a "new social consensus," albeit a slow process, will be crucial.—D. Hollander


1. Rani M, Bonu S and Diop-Sidibé N, An empirical investigation of attitudes towards wife beating among men and women in seven Sub-Saharan African countries, African Journal of Reproductive Health, 2004, 8(3):116-136.