Monday, April 25, 2011

Christians Are...What, Exactly?

An old and well-worn saying, attributed to British Prime Minister Benjamin Disraeli via Mark Twain, suggests that the three kinds of lies are "lies, damned lies and statistics." University of Connecticut sociology professor Bradley Wright takes a look at what we supposedly know based on news stories and other reports of surveys, opinion polls and the statistics they produce in his book Christians Are Hate-Filled Hypocrites...and Other Lies You've Been Told.

For example, many people inside and outside the church believe that the percentage of younger people attending church has diminished over time, and that the church is "losing young people." He notes a widely-quoted statistic that suggests only four percent of today's teens will be evangelical Christians by the time they reach adulthood, but then points out that this statistic comes from a study that's not very reliable because of limited sample size and confusingly-worded questions.

Surveyors prefer larger sample sizes because they have a wider variety of respondents. Calvin used to poll himself and suggest to his dad that parenting techniques change based on the poll results, which favored the commencement of driving lessons and the abolition of bedtimes. His small sample size skewed the results his way.

Survey takers also prefer questions to be as clear and specific as possible. "What's the best color for a vehicle?" seems like a simple question, but the answers may differ if people have different pictures of what a "vehicle" is. Asking, "What's the best color for a pickup truck?" for example, would offer clearer results.

So, Wright says, the particular survey that produced the 4 percent number has a lot of problems and shouldn't be counted on without references to other statistics that would back it up. When he analyzes data from a couple of better-conducted surveys, the General Social Survey and the Pew U.S. Religious Landscape Survey, he finds that there is a dropoff in numbers of young people attending church over the last half-century or so. But it's not nearly what the much doomier forecast says it is and it tracks pretty well with the percentage of all people who've stopped attending church.

Wright also points out that surveys like this may provide some good information about what things are like when it was done, but they're not nearly as useful in extrapolating future events. Since the future hasn't happened, we have no way of knowing what factors may affect things like religious participation. Wright notes that in 1822, Thomas Jefferson predicted that "there is not a young man now living in the United States who will not die a Unitarian." He based it on the increase in Unitarianism in the U.S. in the early 19th century, but that increase didn't continue and today's Unitarians number .5% of the U.S. population. He also cites this xkcd webcomic, which uses the trending increase in the number of a woman's husbands (day before yesterday she had none, but as of yesterday she had one) to predict that by next month she'll have four dozen husbands and thus she needs to invest in bulk wedding cake.

The first chapters of the book, which take apart the methodology of surveys and point out why people should question survey-based information just as thoroughly as they question any other information, are the most interesting. The later ones, in which Wright takes on some of the myths about Christians that are supposedly "common knowledge," are also interesting at first, but begin to repeat themselves after awhile.

On the whole, though, Wright has written a clever brief introduction to the ins and outs of surveying and its good and not-so-good uses, as well as providing some reasons to re-think some of the things we Christian folk might believe about ourselves -- as well as what we believe others think about us.

No comments: