Don’t Panic: A Statistician’s Guide to Religion Data

As I sat at my desk Tuesday, I felt a great disturbance in my Twitter feed as if tens of thousands of Catholics had suddenly left the faith. The cries of both liberal and conservative Catholic bloggers rang out demanding their bishops act to stem the tide of the mass exodus (“Latin Masses only, ordain women, stop giving sanctuary to immigrants, the death penalty is not that bad, is celibacy really necessary?, be nice to Paul Ryan”… it’s like a match game you figure out who would likely be calling for what).

I had known the Association of Statisticians of American Religious Bodies (ASARB) would soon be releasing their decennial estimates for the number of religious adherents by county. When these were posted on The Association of Religion Data Archives (ARDA) website it did not take long for journalists, commentators, and bloggers to react. The surprise for many was that “The number of Catholics decreased by five percent” in the last decade. As Catholic Culture noted, “The census estimates the number of Catholics in 2010 at 58.9 million; the latest edition of The Official Catholic Directory, using figures provided by dioceses, puts the number at 68.3 million.”

Confusion… did the Catholic Church really get smaller in the last decade? If you believe that I have a wager I’d like to discuss with you! We’ve gone over this topic before. To really understand what is going on some clarification on definitions needs to be made.

The ASARB data counts “adherents” and although it is referred to as a census it really isn’t in the traditional sense. Do you remember filling out your form? No you don’t because the adherents counted in the study are reported by the religious groups studied. A Catholic adherent “is roughly equivalent to those who are known in some way to each parish or mission.” The ASARB study considers the remainder of individuals in any county (as measured by U.S. Census totals), not counted by a religious group, to be “unclaimed.” Many of these people are religious and some are Catholic—the Church is just unaware of their presence. They may attend Mass only occasionally and are likely not registered with a parish. Does this make them non-Catholic? No (and they have not “ex-communicated” themselves either!). In fact these people still self-identify as Catholic and do many Catholic things.

The ASRAB study basically measures church-connected membership. This is not the definition used by the Catholic Church or social scientists more broadly. For the Church, being Catholic is defined by baptism. For social scientists we consider self-identification to be the base definition.

With that said there is a lot of consistency between numbers reported in the The Official Catholic Directory (OCD) and ASARB. Why? Because the Church provides both estimates based on what they know using parish registration numbers, October Mass attendance head counts, giving to Church collections, sacramental numbers, etc.

The OCD and ASARB only differ by an average of 2.4% within states. Why are ASARB’s numbers lower overall? ASARB’s 2010 estimates for Catholics in Georgia and Texas are potentially problematic. ASARB estimates there were 596,384 Catholic adherents in Georgia and 4,673,500 in Texas. By comparison the OCD estimated 977,287 in Georgia and 6,896,051 in Texas. A majority of the disparity between the two national estimates for 2010 is based on different estimates for these two states. What numbers would I go with? Not ASARB’s. Surveys using self-identification indicate the Church’s estimates in the OCD were (and continue to be) closer to the reality for Church-connected Catholics. 

Is it possible that the number of Church-affiliated Catholics declined from 2000 to 2010? Not if the measure being used is Mass attendance. The rate at which Catholics attend has remained unchanged as the number of self-identified Catholics in the U.S. grew by 8%. However, it is the case that parish registration has dropped be about 8 percentage points in the last decade. So if the Church is reporting Catholic numbers from registration rolls alone then technically a drop in adherents is possible but unlikely (…and using registration is neither a valid or accurate measurement of activity!).

So just how many Catholics are there in the United States? Currently there are about 77 million and in 2010 it was estimated to be nearly 75 million. Think I’m just being too optimistic? Here are Pew’s recent estimates for 2010 based on self-identification in surveys and census data.

 

The Catholic population has been about 23% to 25% of the total U.S. population since the end of World War II and is expected to continue to account for this much of the population or perhaps even a bit more in the future. As the population grows, so do the total number of Catholics. I don’t know of any serious social scientist (who cared about their reputation) that would argue that the number of self-identified Catholics in the U.S. declined in the last decade (...even as immigration of many Catholics from Mexico may have shifted negative and the Church has welcomed some “reverts”). It may be possible that the number of Church-connected (registered, weekly attending) Catholics declined, but I again I wouldn’t bet on it.

The Church counts from the OCD and ASARB tend to underestimate the self-identified Catholic population by an average of 26%. There are two recent diocesan examples where I can specifically demonstrate the effects of relying on Church-connected counts. In Albany, CARA recently conducted a telephone survey of more than 3,500 randomly selected adults in the Diocese. Of these 34% self-identified their religion as Catholic. Extrapolating to the population this means there were more than 475,000 Catholics in the Diocese of Albany even as only 334,055 are reported in the OCD and 346,013 are estimated using the ASARB data for 2010. The Diocese of Camden recently commissioned a telephone survey (regrettably not with CARA!). This survey is a bit more unreliable as it only includes 612 randomly selected adults of which 34% self-identified as Catholic (...thus it likely has just over 200 Catholic respondents resulting in a margin of error of +/-6.8 percentage points for this sub-group). Extrapolating from this survey, one would estimate there to be about 474,000 Catholics in the Diocese which is similar to the OCD number of 505,504. But ASARB only estimates there to have been 387,872 Catholic adherents in Camden in 2010.

It's not that Church-connected counts are “wrong.” They simply measure something other than the total number of Catholics. Instead, these estimate only active (regular attending) and parish-connected (registered) Catholics. In fact, when these counts are used in conjunction with the self-identified Catholic estimates provided by surveys one has a very good understanding of the numbers that might be a target for New Evangelization and outreach.

[Note: the difference between ASARB/OCD counts and survey estimates for self-identified are not “former Catholics.” This is an entirely different group... and phenomenon.]