Two new reports: Thumma / Bird on Megachurches and Chaves on American Congregations

Below I have highlighted two important new reports on the church by premier academic sociologists.  At the end I have listed a few things to keep in mind while interpreting statistics. 

The National Survey of Megachurch Attenders report "Not Who You Think They Are: The Real Story of People Who Attend America’s Megachurches" by Scott Thumma and Warren Bird

This 40-page PDF was just released.  It is an outstanding example of good research and clear writing. 

Thumma wrote with Dave Travis the excellent book Beyond Megachurch Myths: What We Can Learn from America's Largest Churches (J-B Leadership Network Series) (San Francisco: Jossey-Bass, 2007).  Thumma and Bird work for Hartford Institute of Religious Research and Leadership Network respectively. Thumma has done more research on megachurches than anyone else. 

Probably the biggest difference between megachurches and other churches that they highlight is that "Megachurch attenders are younger and more of them are single . . . Additionally, they are more educated and wealthier" (Not Who You Think They Are, p. 28).  There are positive and negative ways of interpreting this.  The megachurch supporter could say, "Megachurches are doing something right!  They are attracting more youth, single, educated and wealthy people."  The megachurch critic could say, "The megachurch unfortunately probably makes old, married, uneducated, and poor people feel unwelcome." 

The findings will be particularly valuable when critics and defenders of the megachurch declare their personal experiences and opinions as statistical facts. 

This played out in a series of conversations at Leadership Journal's Out of Ur blog and on lots of blogs in December 2008.  During this time, I wrote two posts at Out of Ur:

Megachurch Misinformation Mega or missional? The stats say both are doing well. by Andy Rowell

and

Out of Ur: Missional vs. Attractional: Debating the Research - a post by Andy Rowell and the editors of Leadership Journal

(I tried to chronicle all the discussions at: 
Following Dan Kimball's Missional vs. Megachurch conversation)   

I would encourage the reader of the National Survey of Megachurch Attenders report to note all of the things that megachurches and churches have in common.  There are many common problems that we all need to work on.  For example, Thumma and Bird note, "The Longer People Attend, the Less Likely They Are to Report 'Much Growth' in Their Faith" (p. 27).  Why is that?  There are a number of ways of interpreting that.  In my opinion, this report is what people hoped they might be able to learn from the Willow Creek Reveal and Follow Me reports but unfortunately without sociological expertise and in conjunction with bungled communication, the Reveal reports ended up causing more confusion than anything else.  (I like Willow Creek but think they made some missteps with the Reveal endeavor--I was frustrated because it made them look worse than they are!  See my Willow Creek REVEAL's second book Follow Me tells us very little).

This Thumma / Bird report does not however take a "and this is what we should do about this" approach.  That is up to us in church leadership.

The National Congregations Study report "American Congregations at the Beginning of the 21st Century" by Mark Chaves

This 40-page PDF also came out this week.  In my opinion, Duke sociologist Chaves is the most important sociologist of congregations in the United States.  He is author of Congregations in America (Cambridge: Harvard University Press, 2004), overseas the National Congregations Study, and regularly writes at the Call & Response blog at the Faith & Leadership website.

Here is the summary of findings from page 2. 

"This report highlights some of the National Congregations Study’s most important findings, including:

• Most congregations are small but most people are in large congregations.
• Worship services are becoming more informal.
• Congregational leaders are still overwhelmingly male.
• Predominantly white congregations are more ethnically diverse.
• Congregations embrace technology.
• Congregations and clergy are getting older.
• Congregations’ position in the social class structure remains unchanged.
• Congregations’ involvement in social service activities remains unchanged.
• Only a small minority of congregations describe themselves as theologically “liberal,” even within the Protestant mainline.
• Congregations are more tolerant and inclusive than we might expect them to be, even when it comes to hot-button issues.
• There has been no significant increase in congregational conflict since 1998.
• Congregations’ involvement in political activities is largely unchanged since 1998." (American Congregations at the Beginning of the 21st Century, p. 2). 

The issue that I have referred to repeatedly from Chaves's work is the finding the report begins with.  (For example, see my post How to Read Hybels: Book Review of Axiom by Bill Hybels). It is so important!

• In both 1998 and 2006-07, the average congregation had just 75 regular participants.
• In both 1998 and 2006-07, the average attendee worshiped in a congregation with about 400 regular participants. (American Congregations at the Beginning of the 21st Century, p. 2.  See p. 3 for the explanation of this statistic).

Here is only one of the fascinating implications of this concept. 

"It means that most seminarians come from large churches (since that’s where most people are), but most clergy jobs are in small churches" (American Congregations at the Beginning of the 21st Century, p. 3).

I only have one small quibble with this statistic and I will share it to show the difficulty of interpreting data.  I wish Chaves would have given us the statistics on "weekly attendance" (which the National Congregation Study has also gathered) rather than "regular participants."  It seems to me this is a more accurate way of describing the size of a church than what number the pastor deems are "regular participants."

Here are the two questions.  I like the second question better because it seems less susceptible to bias. 

• Wave II question 13. “How many persons—counting both adults and children—would you say regularly participate in the religious life of your congregation—whether or not they are officially members of your congregation?”
• Wave II question 52. “What was the total attendance, including both adults and children, at all of the worship services that took place this past weekend, including services on Friday, Saturday, and Sunday?”

I did the calculations at the National Congregations Study website.  56% of congregations report having less than 100 "regular attendees including children."  But just 38% of congregations report having less than 100 in "Total attendance for ALL services last weekend."  Apparently, a number of pastors estimated that of the people attending last Sunday, only a smaller percentage are "regular participants" saying something like, "Yes, we had over 125 people in attendance last Sunday but I would only consider 75 to be 'regular participants.'"  It seems to me all of the other sociological research on congregations deals with attendance because this notion of "regular participants" is too subjective.  The pastor who perhaps attracts a large attendance but conscientiously reports a lower number of "regular participants" looks to be ministering to a smaller number of people than he or she really is.  (I of course may be missing something here in my statistical analysis but I think I am right about this). 

The question of what is more important "membership" (described here in NCS as "regular participants") vs. "attendance" is not new.  For example, the United Methodist Church gives "Average Worship Attendance" while The Presbyterian Church (USA) emphasizes membership statistics (though you can also find attendance statistics).  Sometimes, in the PC(USA) the membership exceeds the number who attend each week.  Other times, it is the opposite.  For example, "Your congregation's reported total membership, 548, was larger than the 2007 PC(USA) average, 205. . . . Your congregation's reported worship attendance, 456, was larger than the 2007 PC(USA) average, 114."  Like I said, I think the people who actually show up is the more important number. ( . . . and my Presbyterian friends mutter about Andy's anabaptist ecclesial instincts . . . but I digress). 

Eight Warnings for Church Leaders about Using Sociologist Data by Andy Rowell

All of this information in these reports should be used by the church leader judiciously. 

For my course for Mark Chaves last fall, I wrote my term paper on how pastors should use sociological data.  (Someday I'd like to publish it--any ideas where?) 

I will post below the outline for church leaders and consumers of statistics to keep in mind. 

Warning 1: Theological convictions should determine what gets measured.  Consider measuring both quantity and quality. 

Warning 2: Statistics are descriptive not prescriptive.

Warning 3: Correlation does not mean causation.

Warning 4: It is very difficult to determine the most important causative factor—the right hypothesis—and without it, there will be failed expectations. 

Warning 5: There are always exceptions. 

Warning 6: Good social science is very difficult and all of it needs significant peer review.

Warning 7:  Statistics should also be gathered from outsiders.

Warning 8: Businesses and other organizations are not necessarily more effective organizations than churches. 

See also
Evangelicals Behaving Badly with Statistics
Mistakes were made.
Christian Smith | posted 1/01/2007

See my categories Megachurches and Sociology for more on these topics.
Or see my topic Ecclesiology for more theological reflection on the church in which I always try to stay cognizant of the sociological data.

Update June 15, 2009.
Three stats I have been thinking about a lot so I tweeted about them.  http://twitter.com/AndyRowell

1. Chaves, "51% [of congregations], with 59% of participants, do not allow women to be full-fledged senior clergy." p. 16.  No wonder women in ministry is such a hot issue.  1/2 of congregations are egalitarian and half are complementarian/traditional!
2. Chaves, "Only 9% of congregations [in the U.S.] describe themselves as theologically liberal." p. 13.
3. Thumma, 65% of attenders of megachurches cite "senior pastor" as the most important factor that keeps them at the church. p.18.

Comments

[Josh Rhone]

Andy, excellent post which brings to light some of the issues regarding what separates good research methods from those which are mediocre. Additionally, the 8 warnings regarding how one employs/uses sociological data is excellent!