Social interaction within religious congregations appears to have had a strong impact on the adoption of contraceptive use in Mozambique, particularly in large cities.¹ An analysis of 1997 Demographic and Health Survey data for large cities and rural areas indicates that city women who belong to less diverse and more socially isolated denominations or have no religious affiliation are less likely than mainstream Christian women to have ever practiced contraception or to have discussed it with friends or relatives. A similar but weaker association exists among women living in rural areas, probably because of the minimal social diversity within and among church congregations in rural Mozambique.

Contraceptive adoption and fertility change are probably affected by social interactions among women; church activities are among the situations in which such interactions may take place. The impact of religious involvement on an individual's attitudes toward and knowledge about contraception, however, is difficult to analyze, because religions have widely divergent views on contraception and fertility, and types of social interaction may differ among religious groups.

To explore this issue, a sociologist studied qualitative and quantitative data collected in Mozambique, a former Portuguese colony with relatively high fertility and a low level of contraceptive use. The investigator's hypothesis was that through shared church membership, individuals establish informational links that help transmit innovative reproductive practices, usually from women of higher social status to those of lower status. However, he expected the extent to which innovative practices are diffused to differ by type of religion: Some religions promote social isolation, in which members interact mostly with other members, and this may limit people's access to new ideas.

Census data from 1997 indicate that 24% of people in Mozambique are Roman Catholics, 8% are "mainstream" Protestants (such as Anglicans, Baptists, Methodists and Presbyterians), and 17% belong to churches that practice an indigenous type of Pentecostalism (often known as Zionist churches).* In addition, one in five are Muslims and one in four practice no religion. Muslims were excluded from the study, as women's participation in Islam and Christianity differs.

A research team collected qualitative data during 1998 and 1999 in the suburban area surrounding Maputo, the nation's capital. Church leaders and leaders of women's church groups were interviewed in three Catholic parishes, two mainstream Protestant churches, several Zionist congregations and two churches with practices similar to those of Zionism. In addition, focus groups were conducted among adherents of the various religious denominations, and religious services and weekly women's group meetings were observed.

The investigator notes that in every observed event, women outnumbered men; for urban African women, he comments, "church offers a vital form of association, social support, and identity that complements--and often replaces--the crumbling institutions of family and kinship." In particular, the Zionist churches focus on family issues and emphasize the important role that women play in maintaining a strong family. One result, the researcher observes, is that in weekly church group meetings women's conversations are "usually devoted to exchanges of ideas on how to keep their husbands satisfied, children fed, and houses clean."

In general, the Christian churches studied did not differ much in their official positions on the use of contraceptives: The mainstream Protestant denominations and the Zionist churches, for example, both took a neutral stance. Moreover, according to the investigator, while Catholicism officially rejects modern contraception, "the local clergy does not actively pursue the anti-contraceptive agenda" of the church. Thus, doctrinal differences per se might not have substantial direct effects on members' contraceptive practices.

On the other hand, contraception is generally not formally discussed in the Zionist churches, often because individuals perceive it as something in which church leaders have little interest. In addition, Zionist churches are often quite homogeneous, and members are warned to limit their social contact with the "impure" outside world. In contrast, the Catholic and mainstream Protestant congregations are generally more heterogeneous and are more open to outside influences. According to the researcher, it is not unusual for nurses or others with appropriate training to discuss family planning at church women's meetings. The Protestant and Catholic women also reported having received advice on preventing HIV infection that included information on family planning.

The researcher observes that "at first glance, Zionist women's interaction in and outside the congregations is very similar to that of mainstream church members." Yet whereas Zionist women are "more engaged in the life of their congregations," he comments, they also are more homogeneous in their educational attainment and social class (both most often quite low). In addition, because of the Zionist women's mostly self-imposed isolation from the larger society, they probably have less access than the other women to new ideas. Finally, although older women in Zionist congregations are expected to advise younger women on marital and reproductive matters, they are unlikely to have had much experience with modern contraceptives.

To further analyze the relationship between religious involvement and reproductive behavior, the researcher conducted special analyses of the 1997 Mozambique Demographic and Health Survey. From among the 8,779 women aged 15-49 who participated in the survey, the investigator identified three groups--women who were Roman Catholic or mainstream Protestant, women who were Zionist (or belonged to "quasi-Zionist" congregations), and women who declared no religious affiliation--from which he selected women who lived in major cities and those who lived in rural areas in order to produce a starker contrast. The hypothesis to be tested was that fertility control would be greatest among women affiliated with mainstream churches, less among those in Zionist congregations and least among nonreligious women (who were assumed to be the most isolated socially), and that these differentials in contraceptive knowledge and use would be greater among city residents than among rural women.

Regardless of type of religion, rural women had slightly more children than did urban women, while urban women had attained more education than rural women. Women belonging to mainstream churches tended to have the most years of education (5.3 in cities and 1.6 in rural regions) and women with no formal religious affiliation the fewest (3.3 and 0.8, respectively).

Likewise, levels of contraceptive knowledge were much higher in urban areas, with some religious differentials as well. In cities, 93-96% of Christian women knew of at least one modern contraceptive method, compared with 80% of those with no religious affiliation, while in rural areas, 56-61% of Christian women and 42% of unaffiliated women knew of a method. In cities, the proportion who had ever used a modern method fell steadily, from 48% among members of mainstream congregations to 37% among Zionist women and 27% among unaffiliated women. The relative proportions in rural areas were very much lower and varied less by religion (6%, 9% and 5%, respectively). Finally, the proportion who said they had discussed family planning with a friend or neighbor was slightly higher among urban members of mainstream churches (24%) than among urban Zionist women (22%) or women with no affiliation (18%); similar differentials were seen among rural women (15%, 13% and 9%, respectively).

The investigator then conducted two multivariate logit analyses--one for cities and the other for rural regions--in which he controlled for the effects of cultural, social and demographic characteristics. In the cities, Zionist women and women with no formal religious affiliation were both significantly less likely than women belonging to mainstream churches to have ever used modern contraceptives (odds ratios of 0.66 and 0.63, respectively). In rural areas, women with no religious affiliation had significantly reduced odds of having used a modern method (odds ratio, 0.59), but the contraceptive use of mainstream Christians did not differ significantly from that of Zionists. The researcher comments that "these results conform to the hypothesis that in cities the cultural and informational milieu of mainstream churches would be more propitious for contraceptive experimentation than that of smaller, Zionist-like churches."

In comparable analyses of the likelihood that women had discussed family planning with a friend or neighbor, the researcher again found an impact of religious affiliation: Among city residents, women attending Zionist churches and those not following any formal religion had significantly lower odds of having discussed family planning than did women belonging to mainstream churches (odds ratios of 0.74 and 0.56, respectively). Having no religious affiliation was also linked with a significantly reduced likelihood of family planning discussions among rural women (odds ratio, 0.62).

An examination of educational attainment among the survey subgroups reveals that city women who attended mainstream Protestant churches were much more likely to have completed seven or more years of schooling than were Zionist women. The researcher comments that "the presence of so many women with relatively high levels of education in mainstream congregations helps create a certain innovation-prone cultural environment that may benefit other, less educated members." In rural areas, on the other hand, the proportion of more educated women is so small, he notes, that churches lack the "critical mass of educated innovators capable of influencing the rest of their congregations."

The investigator concludes that in Sub-Saharan Africa, churches are social venues in which women can share their concerns and experiences with other women, although "whether or not this sharing leads to adoption of sociocultural and technological innovations depends largely on the social environment in which it occurs." The researcher also notes that this effect is seen not necessarily in direct communication about contraception, but in individuals' perceptions, through others' examples, that family planning is a desirable and legitimate practice.--M. Klitsch

REFERENCE

1. Agadjanian V, Religion, social milieu, and the contraceptive revolution, Population Studies, 2001, 55(2):135-148.