Religious Preference, Religious Participation, and Sterilization Decisions: Findings from the National Survey of Family Growth, Cycle II

Abstract

Data from a nationally representative sample of ever married women interviewed in 1976 were used to access the findings of earlier studies regarding the relationship of religion to contraceptive sterilization. The National Survey of Family Growth Cycle II provides information about contraceptive behavior birth intentions fertility and other aspects of maternal and child health that are closely related to childbearing. For purposes of investigating the hypotheses about religion and sterilization a subsample of respondents was selected from the main sample. To examine the effects of religious participation and religious preference on contraceptive sterilization recently developed linear probability modeling procedures were used. The sampling design was very complex with sampling ratios ranging from a low of 1 in 43000 to a high of 1 in 647. In overall figures and within groups of the same race education and levels of religious participation contraceptive education was much less prevalent among Catholics than Protestants. Within groups of the same education religious participation and religious preference there was a tendency for contraceptive sterilization to be less prevalent among blacks than whites. Within groups of the same race religious participation and religious preference there was a tendency for the proportion contraceptively sterile to decline with education. Religious preference was both a substantively and statistically important determinant of contraceptive sterilization. Being Protestant raised a couples probability of being contraceptively sterile by 3.7% above the average while being Catholic reduced the probability by a similar percent. Religious participation had virtually no effect on the probability of being contraceptively sterile and was not statistically significant. Race and education were the 2 most important determinants in the model. Being white raised the probability of being contraceptively sterile by 6.9%; being black lowered it 6.9%. Overall education was the single most important determinant. Having less than a high school education raised the probability of a couple being contraceptively sterile by 8.6% while having more than a high school education lowered the probability by 7.8%. Among the interactive terms the interaction of religious preference and education was the most powerful determinant. In general for Protestants education decreased the likelihood of a female sterilization operation but for Catholics education increased the likelihood. Overall there is only slightly more than a 10% difference between Catholics and Protestants in the use of sterilization and the data suggest that this difference is due primarily to differences in social characteristics rather that doctrinal differences.