Monday, May 16, 2011

Of Wild, Feral, and Domestic UUs...

UPDATE: I've made a correction based on Chris Walton catching an error I made while moving data from the World's pdf to Excel. I've corrected the data in the columns so that it's accurate, but am striking through the erroneous data in the text and replacing it with the corrected numbers. Thank you, Chris!

Also, Christine asked about how I extracted numbers from the Pew survey on the unchurched, which made me realize I'd forgotten to write about assumptions and method, as I'd intended. The basic assumptions; the two surveys are accurate and reliable, and the Pew survey includes both churched (member) and unchurched UUs in proper proportions, as well as that the two data sets, taken four years apart, represent pretty much the same population. I think all of those are reasonable, but any one of them might be questioned.

It's that second assumption that the actual math rests on. If member UUs are represented in Pew's sample in proper proportion to the overall population, and non-member UUs are as well, then we can take the Pew number, subtract the UUA's adult membership data, and discover how many unchurched UUs there are (the "wild" and "feral" populations). This, in fact, is what was done in the post I linked to in Transient and Permanent. When I saw that post, and the data from UU World, I realized that one could do some fairly simple algebra; we had the Pew numbers (UUs at large) and the UU World data (member UUs). Backing out the UU World numbers from the Pew numbers would give the numbers for the free range UUs. If the two sub-population were the same size, I'd simply have subtracted (Pew - UU World = unknown). Transient and Permanent's insight that there were three free range UUs for every member UU (actually, just over three--the numbers are 76% free range and 24% members--since I don't believe that the numbers we're working with are that precise, I worked with the 3:1 ratio) just required weighting the numbers. Thus the actual formula looks like 4(Pew) = UU World + 3(unknown) -- the need to multiply Pew's number is because we're working in percentages, not absolute numbers.

-------------------------

On Facebook this morning, Chris Walton posted a link that I'd previously missed at UU World, a comparison of data from two polls, one by UU World in 2004 and one by Pew Research in 2008. Both look at reported data from UUs. What's fascinating are the differences....

I am indebted as well to this posting at Transient and Permanent, which mulled over some of the Pew data three years ago.

Here's the basic data:


UU World

Pew Research

Income



< $30,000

14%

19%

$30k–$49,999

13%

25%

$50k–$74,999

19%

16%

$75k–$99,999

16%

13%

$100,000>

25%

26%

No response

13%





Education



< high school

0%

3%

High school

2%

16%

Some college

11%

30%

College grad.

20%

22%

Post-graduate

65%

29%

No response






Gender



Male

31%

54%

Female

65%

46%

No answer

4%





Age



Under 30*

4%

18%

Over 65

29%

16%


The asterisk notes that UU World used the UU range of 18-35 for Young Adults, when surveying.

Pew surveyed some large number of Americans and calculated that there were, as of 2008, some 683,000 UUs. Adult UUs. Given the data from the UUA at the time, Transient and Permanent observed that this meant that there were just over three UUs out there, unchurched (but familiar enough with us to claim to be UUs) for every one of us who's on the books as a member. That's a startling number.

The Pew data alone is striking. In the population at large, slightly over half of those affirming they're UUs are male. But no more than a third of those who are members are male. Given the imperfect overlap on the age ranges, it looks like there are five times as many under 30s who say they're UUs as are members. Etc., etc. The numbers clearly indicate that there's a very real difference between the different UU populations, the unchurched--what worker referred to as "free range," recently--and the churched.

And then it hit me. The Pew survey wasn't of the unchurched. It's of those who responded that they were UUs; members and not, both. The difference between the churched and unchurched would be even more extreme. So I did some number crunching (after having sufficient coffee, because uncaffeinated this morning, I managed the trick of multiplying billions by two (I think that's what I did, anyway; it's irreproducible) and came up with trillions). What I did was to figure out what the unchurched population would have to have responded with in order for Pew to get its numbers after adding in the churched part of its sample. (Note to self: Take this back to the kids to prove that there are occasional adult uses for algebra.)

The numbers are, of course, imperfect; there are rounding errors, and such. So columns do *not* add up to 100%, etc.


Churched

Unchurched

Total UU pop

24%

76%




Income



< $30,000

14%

20%

$30k–$49,999

13%

29%

$50k–$74,999

19%

15%

$75k–$99,999

16%

12%

$100,000>

25%

26%

No response

13%





Education



< high school

0%

4%

High school

2%

21%

Some college

11%

36%

College grad.

20%

23%

Post-graduate

65%

17%

No response

3%





Gender



Male

31%

62%

Female

65%

40%

No answer 4

4%





Age



Under 30*

4%

23%

Over 65

29%

20%


The classism that some of us have talked about becomes more apparent when we look at what economic class member UUs are likely to be in, versus non-member UUs. Ouch. Interestingly, the people in the upper middle class-to-wealthy range are the only group that seems unaffected. They're equally likely to be UUs.

Education is even more striking. That's where the real class boundary is stunningly apparent. Mark Morrison-Reed's analysis that education is the single best marker for whether one is likely to be a (member) UU is clearly on the money. Ouch, ouch.

Gender... ok, so it's not news to anyone that the outwardly religious have been predominately female for well over a century. It thus isn't surprising to see that UU members are two-thirds women. But if it was a surprise that over half of those who say they're UUs are male, to find that three out of five people who are unchurched UUs are male was startling.

And then there's the age data. 65% 29% of those who are members are over 65. Only 17% 20% of those who are unchurched UUs are over 65--and while at most 4% of those who are UU members are under 30, 23% of those who aren't members are under 30.

So what's the picture look like--in very crude, simplistic averages?

Church-going UUs are relatively well-off, extremely well-educated older white women. Indeed, that's what we see in congregations. "Free range" UUs are probably middle class, but are pretty likely to be lower middle class, if not poor; they're much less educated (perhaps struggling still to get educations?), they're young, and male. Simply taking US demographics--young, male, lower education and income... they're also far, far more likely to be mixed race or people of color.

Now that's a divide that's going to take a lot of work to bridge.

5 comments:

Chris Walton said...

Ogre, you've grabbed the wrong number for the percentage of UU members who are over 65 years old. The UU World survey found that 29 percent are over 65, *not* 65 percent.

Christine Leigh said...

Fascinating!

Math geeks among us (or, uh, just me) are curious about your methods. When you said you calculated how many unchurched Pew would need to find in order to come up with the numbers, exactly how did you separate out the churched from unchurched? It's not that I don't believe you, it is that I want to know how you did it incase I come across similar data, I could know how to do this. I'm not afraid of algebraic equations :D

ogre said...

Thanks, Chris! The risk in moving data.... I wish I'd been able to copy and paste it into Excel.

I'll go and update accordingly.

And Christine, I'd meant to include a comment about my assumptions and method; I just got ahead of myself. As long as I'm updating, I'll do that, too.

Transient and Permanent said...

Glad to see someone continuing the discussion. I've been too busy with teaching and writing duties to update the blog very much.

This is a good preliminary analysis, and I appreciate the handy numbers. Let's remember too that there is a lot more going on here than classism. I'll provide my own biography as an example of one common pattern that partially accounts for these numbers.

As a young adult, I was poorly paid, highly mobile (had to move multiple times to various parts of the country, and eventually to a foreign country, to pursue work), and did not yet have graduate education. I also happened to be male. I was sometimes but most often not a UU church member during this time, though I was at all times a UU. Being young, I was poor and mobile. Being young and mobile, I didn't spend enough time in many places to really connect to a local church. Being poor and especially being young, I didn't connect too well with some of the congregations I did encounter. Being young, I was a lot more into my friends and going out than I was to attending church regularly.

As I aged I realized that my income situation would never improve dramatically without more education. Thus being older, I turned from the pleasurable pursuits of youth to the task of gaining additional education. That made my income go up. Acquiring a well-paying job allowed me to finally put down roots. Staying in one place made me more likely to join a local church, and being older I wasn't in to going out with friends as often anymore. And when kids came along, that was basically over with--and it was the necessity of providing the kids with religious education, as much as anything, that caused me to finally join my current UU church. Soon we hope to buy a house, which I suppose will complete our decade-plus climb up into the solid middle class.

This is just one story, but it's not an unusual one. The takeaway from this is that churches will almost never be a full representation of the public at large's demographics because external factors keep many otherwise committed UUs from officially joining even the most welcoming and cool congregations. This is of course true for virtually all denominations, it is not a UU situation specifically.

Of course, issues of classism, ageism, and other internal factors do come into play, especially as reinforcing elements that increase disparities that already exist for unrelated reasons. Regionalism is also important, as UU churches are not at all evenly distributed, and there are some overlaps here. For instance, people in rural areas of the United States tend to be poorer and less educated, just like unchurched UUs, and there aren't many rural UU churches left. Of course, we see here again the interlinked nature of all of this, as we can argue that classism contributes to UU churches being located in affluent suburbs or residential urban areas. It's not the sole key to explaining everything, but it can never be removed from the equation.

Tom said...

The Pew people surveyed 35,000 people of whom .3 percent said they were UUs. That is only 100 individuals. Assuming that 25 were actual UUs your data on the unchurched is coming from 75 responses.

What are the odds that 75 out of 35,000 people would simply answer the question incorrectly? If one assumes that people who are male, less educated and poorer are also worse at responding to questionaires that could explain the data.