Many sociologists point out that the data compiled by Durkheim and his followers may not be ideal representations of religious integration. For instance, Rodney Stark, a sociologist of religion, opposes the use of church membership or attendance as the only measure of religious integration. He points out that the data on church membership has only existed in a few countries for only a limited number of years. Although the data has become more accessible in recent years, the actual number of church membership has always been under-reported because of difficulties in data collection (Stark, 1987: p.74). Additionally, even Breault and Barkey themselves admit the limitations in their study design of using the number of religious books as an indicator of religious integration. For instance, Eastern European and Islamic countries in particular have a lower number of religious publications, but this does not necessary represent low religious sentiment in these regions (Breault and Barkey, 1983: p.631). In addition to the cross-national differences, focusing on the number of religious publications is unrelated to the power of religious integration in Durkheim’s fundamental argument (Breault, 1986, p.639). Therefore, the empirical evidence provided by Durkheim and his followers becomes less persuasive since researchers cannot reach an agreement about the how to …show more content…
In order to examine Durkheim’s prediction, Breault and Barkey analyzed family integration and suicide rates across 42 countries in 1970 (Breault and Barkey, 1982: p.325). They measured the level of family integration by the rate of marriage minus divorce rates, divided by marriage plus divorce rates. They found that, as predicted by Durkheim, suicide rates are reduced significantly when the family integration is strong.
Despite the evidence provided by Durkheim and his followers, many sociologists have remained suspicious about whether marital-status differences in suicide rates are caused by family integration or the marital selection mechanism. J.E. Veevers argues that one possible selective factor for a stable and satisfying marriage is mental health, such that neurotic people with high risk of suicide are more likely to get divorced (Veevers, 1973: p.139). Therefore, the correlation between family integration and suicide rates can be spurious since people with suicidal tendencies are more likely to divorce in the first