Ghanaian women's fertility preferences are generally stable in the short term, according to a study that repeatedly assessed such preferences over a five-year period.¹ Among fecund women, 81% of those who said in a given survey round that they did not want any more children reported the same preference in the next round, typically 7–10 months later; similarly, 79% of those who did want a child in the future felt the same at the subsequent survey. However, the stability of women's preferences varied according to whether they had attained their ideal family size.

Data on fertility preferences are used for such purposes as estimating unmet need, and these preferences are often assumed to be stable. However, recent work suggests that fertility desires may be influenced by temporary factors, such as economic conditions. To examine the extent to which preferences fluctuate, researchers asked women in southern Ghana about their fertility preferences up to eight times from 1998 to 2003. A total of 1,219 women aged 15–50 were interviewed in the study's first round, and an additional 209 were added to the sample in the second round. The vast majority of respondents remained in the study during the subsequent six rounds: The attrition rate between the first and last rounds was 15%. Survey rounds were irregularly spaced, but most intervals were 7–10 months.

In each round, women were asked, "Would you like to have a(nother) child with your husband/partner, or would you prefer not to have any more children with him?" Women who wanted a child in the future were asked about their preferred timing; on the basis of their answers, they were classified as wanting a child soon (within two years), wanting a child later, wanting no more children, undecided or infecund. To assess women's ideal family size, the researchers asked women at study entry, "If you could go back to the time when you did not have any children and could choose exactly the number of children to have in your whole life, how many would that be?" Respondents also provided information on their marital status, contraceptive use, pregnancies and demographic characteristics.

The researchers conducted two analyses. They first examined, for each woman, changes in fertility preference categories between consecutive survey rounds; thus, information was available for up to seven survey intervals per respondent. The second analysis, ideally, would have examined the preference patterns of individual women across the study's eight rounds, but doing so posed statistical challenges because of the high number of possible patterns. Instead, the researchers performed a latent class analysis using data from rounds 2, 4, 6 and 7. In this approach, which simplified the statistical analysis and provided intersurvey intervals of about one year, fecund women whose preference patterns were broadly similar were grouped into classes.

Overall, 1,428 women participated in the study. Mean age at study entry was 31, and about four-fifths of the women were married. More than a third (35%) had never attended school, and another one-fourth (24%) had no more than an elementary school education. Most women were Christian (60%) or Muslim (21%). During the course of the study, the mean lifetime number of births among study participants increased from 3.5 to 4.2, and the mean number of live children rose from 2.8 to 3.5. At baseline, mean ideal family size was 4.4.

Between rounds one and eight, the proportion of women who wanted no more children increased from 30% to 36%, and the proportion who reported being infecund rose from 2% to 9%. Conversely, decreases occurred in the proportions of women who wanted a child within two years (from 28% to 22%) or later (from 34% to 22%). Consistency of fertility preferences between consecutive survey rounds was highest for women who wanted no more children (76%) and lowest for women who were undecided (38%); stability of preferences was intermediate among women who wanted a child soon (55%) or later (62%).

In analyses that omitted infecund respondents, 81% of women who wanted no more children in a given round expressed the same preference in the next round; the proportion was greater among those who had attained their ideal family size than among those who had not (86% vs. 75%). However, only 58% of women who reported in any of the first seven rounds that they wanted no more children expressed the same preference in the final round.

Preferences were also generally stable among women who wanted more children: Seventy-nine percent reported in the next round that they wanted a child in the future. Here, consistency was greater if women had not yet reached their ideal family size. For example, among women who wanted a pregnancy soon, nearly two-thirds (64%) of those who had not attained their ideal family size reported at the next survey that they wanted a pregnancy within two years, compared with only 39% of those who had attained their ideal family size.

The latent class analysis revealed five broad groups of women. The largest, comprising 31% of respondents, consistently reported wanting a child, but not in the next two years. These women were generally young (mean age at baseline, 24) and hence relatively early in their reproductive careers; they had an average of one birth during the study. A similar proportion of women (29%) tended to report not wanting to have a child in the future; these women were typically nearing the end of their reproductive years. The third largest group, representing 16% of women, typically first reported wanting a child after two years but then switched to wanting no more children. The fourth group was generally characterized by women who repeatedly said they wanted a child soon, though most did not in fact give birth during the study, while women in the fifth and smallest group tended to be undecided about their childbearing preferences, in part because they were less likely than their peers to be married.

Overall, the findings show that "approximately one in five women changes her mind about whether to have an additional child within the next several months," indicating that "considerable regularity and stability exist in stated fertility preferences over time," according to the researchers. The results suggest "that most women's preferences are reasonably strongly held and are not likely to be driv-en by everyday circumstantial factors." In a separate analysis, the researchers found that most women who wanted no more children said their partner felt otherwise. Because perceived spousal disagreement about family size is a substantial barrier to adoption of modern contraceptives, the authors emphasize that "empowering women to exercise control over their fertility is clearly central to the achievement of their fertility preferences," and family planning policies should "promote contraceptive methods suitable to [women's] familial and social contexts."—P. Doskoch

REFERENCE

1. Kodzi IA, Casterline JB and Aglobitse P, The time dynamics of individual fertility preferences among rural Ghanaian women, Studies in Family Planning, 2010, 41(1):45–54.