It has been a very interesting couple of days with the release of some new data sets from the Presbyterian Church (U.S.A.) Research Services office. I don’t think I had seen the 2011 Comparative Statistics on the web site before today and in the last couple of days a new Presbyterian Panel report was issued on Current Issues in Church and Society. (For the record the summary report was issued and I am interested to see the full report when it is made available.)

There are a number of interesting items in that summary report but the one that got a lot of play yesterday was the attitude of members towards same-sex marriage. This was highlighted by a Christian Post article about one of the findings headlined “Poll: PCUSA Members Increasingly Favoring Same-Sex Marriage.” This article has been tweeted about a number of times in the last day or two.

Here are the details of the findings from the second paragraph of the article:

In research conducted by Presbyterian Research Services and published in
October, the number of PC(USA) members and pastors supporting the denomination
redefining marriage grew. In 2005, 23 percent of PC(USA) members
supported same-sex marriage; in 2012, the number had increased to 34
percent. Among pastors, support for same-sex marriage in 2005 was at 35
percent; in 2012, it is at 49 percent.

The article then presents two interpretations.

From the author of the original research article, Mr. Jack Marcum, coordinator of Research Services the news article has the quote:

This
result indicates a broad-based shift in opinions across the church in
only a few years. Longer term, the effect of generational change will be
felt: 75 percent of young adult advisory delegates at the General
Assembly supported the redefinition of marriage.

The article also quotes Dr. Paul E. Detterman, executive director for Presbyterians for Renewal, with a different perspective:

There
is a substantial number of conservative individuals and congregations …
who have disengaged from these debates within the PC(USA) or who are
now in the process of departure.

This survey is
probably quite accurate in reflecting the views of people responding in
the PC(USA) in 2005 and people responding in the PC(USA) in 2012. As
different from the U.S. demographic analysis, however, these are not
people who have changed their views – these are simply not the same
people.

Without a longer baseline on the study it is difficult to distinguish between these two interpretations since, along the lines of what Dr. Detterman points out, the study group for the Presbyterian Panel is refreshed every three years. But in thinking about this I figured it was possible to put some constraints on these possibilities with a few end-member models. While these do not definitively answer the question I think they point to the interpretation that the answer is somewhere in between these two viewpoints. Probably no surprise there.

OK, now things are going to get geeky – you have been warned.

And what follows comes with this caveat: As you will see these are back of the envelope calculations. I could do more detailed modeling, and someday I might, but for now I think these ballpark models give good enough results to constrain the viewpoints further. And being back of the envelope calculations I am going to assert things in a couple of places that I am not going to chase down the citations for. Also, the survey divides between members, ruling elders, teaching elders in parish ministry and teaching elders in specialized ministry. Since this deals with membership changes I am only going to consider the members.

So here are the data: The statement is “same-sex couples should be allowed to marry.” In the current survey 51% of members answered disagree or strongly disagree. There were 34% of members who said agree or strongly agree and 15% who were neutral or not sure. This compares with the 2005 report where 61% answered disagree of strongly disagree, 17% were neutral or not sure and 23% answered agree or strongly agree. And from here on I will use “agree” to mean agree or strongly agree, “neutral” to mean both neutral and not sure and “disagree” to mean the combined disagree and strongly disagree. (And note that in the summary report they are not further broken down either.)

Since these are membership models we need to state that in 2005 the PC(USA) membership was 2,313,662 and in 2011 it was 1,952,287.

Model 1: The simplist model is just to say that all those that left the PC(USA) in this time period were those that disagreed with the statement. This would mean that if we extrapolate the sample population to the total membership 532,142 members agreed with the statement, 393,323 were neutral and 1,179,167 disagreed. Since 2005 the PC(USA) has lost 361,375 members so if all of those come out of the disagree population that leaves 818,593 members in that group.  That means that out of the new total population the group that agrees is now 27%.

On a number of levels this model is a bit of a stretch, not the least of which is the fact that losses include three categories – transfers, deaths and other – and it is tough to make an argument that those that have gone on to the Church Triumphant all disagreed with that statement. (I personally know several that strongly agreed.) But what this model does do with this particular stretch is to demonstrate that just with departures attaining the 34% is not likely.

Model 2: Let’s try something similar but introduce another factor. This end-member says that every individual that leaves (even those that die) disagrees with the statement and every individual that joins is in favor. We can add up those that joined from the Comparative Statistics and find that is 605,063 members. For the record 966,438 left in that same time period. That means that our group that agrees has grown to 1,137,205 or 58% of the denomination.

So if you want to argue that the shift in the PC(USA) is due to change in people and not attitudes this is clearly your starting point for a more refined model.  The model is underdetermined so you can move a lot of variables around to figure out ways to make the replacement population grow to 34% agree.

But one observation regarding this – the PC(USA) does not have well defined patterns of church growth and decline along the lines of theological viewpoint. (Something else I am working on but that is for another day.) If this replacement model is workable then the replacement is happening across large numbers of churches in the denomination and not in concentrated pockets of congregations.

On a side note, one of the things that always amazes me when I crunch these numbers is seeing that the denominational turnover is about 100,000 a year and so we become a new church every 20 years. Whether this is truly enough replacement with individuals with differing opinions from those they replace is an open question.

Model 3: OK, let’s look at this a little differently and, in my mind, maybe a bit more realistically.  I have a model I have used elsewhere that breaks the PC(USA)’s 3-ish percent decline into a “mainline” component of about 1.5% and a “controversy” component of about 1.5%. So what if the mainline component is across the board and the controversy component is only for those that disagree?

Well, it turns out that it only results in a 2% increase in the number that agree. So this does not get us very far by itself and replacement and changing individual opinions are still necessary.

Discussion: Since these are only back of the envelope calculations I am not going to draw hard conclusions from these numbers.  But I will say that as I look at these models the answer is probably somewhere in between. Sure, you can find a combination of variables that will give you the results of the survey based on turnover alone, and maybe that is Dr. Detterman’s point. I can’t disprove it at the moment but as I look at the model it would seem to require not just some significant turnover in the PC(USA) but significant turnover with one fairly homogeneous group being replaced by another.

On the other hand I think Dr. Detterman is correct that departure of those that disagree needs to be considered in the conclusion. My point is that this is a complex system and that both the departures from the denomination and the change in attitudes of those in the denomination are probably involved.  If I had to guess, and this is purely a guess, I would think it would be in roughly equal parts.

(And it is worth noting that I may be placing too much emphasis on the two different views and maybe the news article just quoted them in a way that makes their explanations look mutually exclusive.)

So maybe I’ll add some more variables into Model 2 and see how realistic the numbers are to make the current number of 34%.

In case you are still with me I thought I would throw in a couple more interesting tidbits from the Panel report.

• When asked if they were familiar with various groups and initiatives in the PC(USA) 17% said they were familiar or very familiar with the Fellowship of Presbyterians, 7% said the same for ECO, 3% for 1001 New Worshiping Communities and 1% for NEXT church.
• Regarding the 1001 New Worshiping communities 32% of members and 40% of pastors said it was a realistic or very realistic goal. (And those are the numbers in the narrative because the quantities shown in Figure 2 look different to me.)
• For members, 17% recall any discussion in the past year about their congregation leaving the PC(USA).
• There is a big difference between members and pastors regarding whether a split is inevitable – only 20% of members think it is while 53% of pastors do.

Well, hopefully that gives you something to think about. I hope to revisit a couple of those with more data.