3. Methods 5. Discussion

4. Results

Table 2 shows the gender difference in selection of marriage partner. Never married grooms married never married brides in 93 per cent of cases (949 out of 1020) whereas never married

(Table 2 here)

brides married never married grooms in 78 per cent of cases (949 out of 1220). Likewise, the remarriage rate per 100 marriages was 42 (17 + 5+ 21) for grooms and 31 (3 + 28) for brides. The incidence of polygyny was 17 per 100 marriages of grooms. Nearly one-third of the brides of polygynous unions was never married (86 out of 293), 60 per cent were divorced women (180 out of 293) and the rest were widowed. Grooms chose brides of similar marital status (homogamous marriages) in 64 per cent of all marriages.

The divorce rate (per 100 marriage-years) showed an inverse relationship with duration of marriage; the rate was especially high during the first two years of marriage (Table 3). Marriages experiencing higher than average divorce rates were polygynous

(Table 3 here)

marriages followed by remarriages of brides who were divorced or widowed before. The divorce rate was found to be somewhat higher if grooms were illiterate. The divorce rate during 12-59 months of marital life was lower if brides had had a live birth in the preceding six months than if brides had not had a live birth.

Spouses' age at marriage may have a bearing on the divorce rate. Since bride's and groom's age at marriage was found to be strongly correlated (r = 0.74), and bride's age was not significantly related to the divorce risk when controlling for groom's age (data not shown), the bride' s age was not included in multiple logistic regression analysis of divorce.

Table 4 shows that after adjustment of bride's and groom's predisposing factors

(Table 4 here)

duration of marriage was inversely related to the odds of divorce. The odds were 2.47 times higher if grooms were polygynous men than if grooms were first-time married men. The odds were 1.6 times higher if brides were widowed or divorced before than if brides were never married prior to the current marriage except when marriage duration was 24-35 months. The odds of divorce were higher, but not statistically significant, if brides were widowed than if brides were divorced before (data not shown). This result is based on a small number of remarriages of widowed women. Groom's age at marriage and kinship between spouses were negatively related to divorce risks, although weakly. Groom's schooling and high household socioeconomic status led to lower odds of divorce except when marriage duration was 12-23 months.

Because of the high probabilities of divorce of polygynous unions and bride's remarriages, the relationship between fertility and divorce was, therefore, examined using first marriages of both partners lasting more than one year (Table 5). The result shows that

(Table 5 here)

the odds of divorce for women who had had a live birth in the preceding six-months period were one-third compared to women who had not had a live birth. Among women having a live birth, the odds of divorce were lower, but not significant, for women who had a son than for women who had a daughter (data not shown).


3. Methods 5. Discussion

Determinants of Divorce in a Traditional Muslim Community in Bangladesh
Nurul Alam, Sajal K. Saha, Jeroen K. van Ginneken
© 2000 Max-Planck-Gesellschaft ISSN 1435-9871