Paradoxical negative spill-over of Catholics’ attitudes on induced abortion

Even though it is rather widely known in sociology that individual actions can have unexpected or seemingly contradictory outcomes on the societal level, I always get that little tingly feeling of discovery when reading about such a seemingly paradoxical mechanism. Interestingly, Jelen et al. have found one regarding the attitudes of Catholics on induced abortion.

How is this possible? Can individual and contextual effects of Catholicism on abortion attitudes run in opposite directions? Jelen et al. hypothesize at the start of the article on three possible ways that Catholicism may influence attitudes. The first is that the Catholic church is able to impose pro-life attitudes on its own members. Secondly, it is hypothesized that the presence of Catholics in a country, along with their inculcated pro-life attitudes, effects non-Catholics in such a way, that their opinion also changes towards the pro-life stance. Finally, it is argued that the opposite might also happen: non-Catholics could take the opposite stance to the Catholics in the presence of many Catholics. A counter-mobilization, so to say.

As it was shown by the authors, the first and third hypotheses go: Catholics object against abortion, but when many Catholics are present in a country, it is shown that individuals, net from the effect of Catholicism, generally have a more positive attitude towards induced abortion. Although the authors do not use these words, I think that what they have found can be referred to as a ‘negative spill-over effect’ (spill under?), in contrast with the second hypothesis that is referred to as a ‘spill-over effect’.

On Method

Although I think that the authors elegantly present some interesting findings, and did their analyses on the data from the World Value Surveys in a meticulous fashion on the whole, some aspects of their methodology deserve some closer attention. Sure, nowadays we would use a mixed-effects model instead of the ‘flat’ regression the authors used, but remember that this is a 1993 article.

In general, one might ask why the authors only focused on Catholicism. Clearly, this is perhaps the church that must influentially instills its many members with pro-life attitudes, but nevertheless other denominations do so as well. They do however find that Catholics in a predominant Protestant country object against induced abortion the strongest, so on a contextual level attention is paid to other denominations.

What I missed in this analysis, is an estimation of the impact that legislation has on the attitudes people have. Reason for this is the detailed description of the differences between countries on account of whether or not induced abortion is legalized, and under which conditions women can choose to have an abortion. It would be interesting to see whether this has any effect, and whether it interacts with religious conviction.

More difficult to understand is the way they achieved uncorrelated individual and contextual variables. They performed a kind of standardization by subtracting individual level Catholicism from the country mean of Catholicism. Since you are either a Catholic or you’re not, this means subtracting a value 1 or 0 from the proportion of Catholics in a country. This would indeed lead to uncorrelated variables, but I’m wondering if this leads to a methodological artifact. Their procedure resulted in different values for a Catholic in a non-Catholic country than for a Catholic in a predominantly Catholic country. I worry, but am not completely sure, that this might have affected their outcomes to some extent.

To Conclude

The seemingly paradoxical finding has been solved: individual and contextual effects of Catholicism on attitudes toward induced abortion run in the opposite direction, caused by a counter-mobilization amongst non-Catholics. Interestingly, the authors discuss this by arguing that the net effect of Catholicism is difficult to assess. I wonder if the individual Catholic, expressing the pro-life stance he or she wholeheartedly beliefs in, realizes that these efforts may indeed unexpected, and unintended, consequences by instilling pro-choice attitudes amongst non-Catholics.

Reference

Jelen, T.G., O’Donnell, J., Wilcox, C. (1993). A Contextual Analysis of Catholicism and Abortion Attitudes in Western Europe . Sociology of Religion, 54(4), 375-383.