Sep 9, 2014

Measuring religious identities / interpreting religious identities

There's some truth to the perception that Europe is generally less religious than the United States, but the claim is often exaggerated.

Europeans, by some accounts, have completely abandoned religion. They've given up faith in droves, leaving churches as empty monuments to another time. The modern European landscape is agnostic and atheist and looks like IKEA crossed with The Girl with the Dragon Tattoo. Of course there are ways in which that's true, but it's not that true.

These are exaggerations.

These are broad characterizations bolstered, largely, by anecdotal accounts of impressions made in casual cultural exchanges.

Such exaggerations are further bolstered by the fact European religious identities are often measured differently than religious identities in the United States.

When asking about religious affiliation, for example, the European Social Survey -- a respected organization -- first asks a yes-or-no question. It asks, "Do you consider yourself as belonging to any particular religion or denomination?" If the answer is "yes," then there is a follow-up, "Which one?" Studies of US religious identity, by comparison, ask for the respondent's religious identity and offer a list of choices, including "none."

The European way of asking the question, according to some comparisons, results in as many as twice the number of people saying they have no religion.

It would be too simple to say that non-affiliation is overrepresented in the European studies. But comparisons between Europe and America need to be complicated: differences can be exaggerated by different ways of asking the question.

"In Europe, many surveys measure religious identity with a two-step question," explains Conrad Hackett, a demographer for the Pew Research Center's Religion & Public Life Project. "In many cases, two-step questions seem to filter out respondents who might otherwise claim a religious affiliation but who do not consider themselves as having a significant level of religious belonging."

Hackett offers this as one example of the complexity of measuring religious identity.

In an excellent article recently published by the journal Religion, Hackett argues that religious measurements are often not straightforward, and study results need to be considered with care.

He makes seven suggestions for those considering survey results:
(1) Definitions and measures of religious identity shape knowledge about religious groups;
(2) Variation in question wording leads to variation in responses;
(3) Comparing results across surveys provides valuable perspective;
(4) Incentives shape how respondents report their religious identity;
(5) Religious identity may be liminal;
(6) Salient identity categories are often unmeasured; and
(7) Religious identity and religious practice may not seem congruent.
Another example of the complications of religious measurements that Hackett dicusses is the wildly divergent measures of American evangelicals.

He writes,
Evangelicals may be considered a small, thriving minority or about half of the United States, depending on how they are measured. In 1998, Christian Smith and his colleagues described evangelical Protestants in the United States as an embattled and thriving minority constituting about 7 percent of adults (Smith et al. 1998). In the same year, the Gallup organization reported 47 percent of US adults to be evangelical (Gallup and Lindsay 1999). The difference between these estimates -- 40 percent of the adult population in the US -- was equivalent to over 80 million people in 1998. Since these estimates are from the same time period, it is safe to assume that the differences are due to measurement and conceptualization issues rather than change in the size of the evangelical population. Nevertheless, the gulf between estimates of the evangelical population based on Gallup's measure and on Smith's estimate of the evangelical population has been cited to explain the putative 'decline of evangelical America' (Dickerson 2012).
Gallup measures evangelicals based on positive response to the double-barreled question: 'Do you consider yourself a born-again or evangelical Christian?' Gallup counts all respondents saying yes to this question as evangelical, including Jews, Catholics, and other religious groups that scholars may consider to be outside what is presumed to be the Protestant world of evangelicalism. By contrast, Christian Smith and colleagues operationally define evangelicals as Protestants who either worship at least twice monthly or say their faith is extremely important to them (to include those unable to attend worship services for health reasons) and who describe themselves as evangelical when asked to pick which identity best describes them from a list including evangelical, fundamentalist, mainline Protestant, liberal, or something else.
There is, when one studies these things, always the fact of the matter. The problem is, there are too many facts of the matter, and they all always require interpretation. The facts require some careful thought, as Hackett shows.

You can't just measure religious identities. You have to interpret religious identities.

The essay, "Seven Things to Consider When Measuring Religious Identity," can be read here.