09 January 2007

Decline. [Updates]

Anglican blogger Karen B. came across the just-posted membership statistics for the Episcopal Church and did a quick compare and contrast excericise. It ain't a pretty picture, especially when the non-domestic dioceses are broken out:

Click on image to enlarge.

Update: Considerable email discussion of these statistics, most of which is hard for me to follow because I'm pretty innumerate (got a "D" in statistics, as a matter of fact); however, there is some skewing in these numbers due to (1) non-domestic dioceses added between 2002 & 2005; and (2) calendar anomalies in 2005 - 53 Sundays, and Christmas falling on a Sunday.

More Update: In addition to Karen B.'s comments here, TitusOneNine has a summary of and links to discussion and number crunching. The only diocese to grow relative to general population growth? South Carolina.



Blogger Karen B. said...

Thanks for posting this Patrick+ One of these days I'll get my own (non-prayer) blog and post this stuff myself rather than bugging all my blogging friends! LOL!

In any case, let me clarify the e-mails about the non-domestic dioceses. As soon as I'd sent out that table, a friend e-mailed and asked (essentially) "but couldn't the growth in the non-domestic dioceses be due to the fact that some overseas dioceses are included in 2005 that weren't counted in 2002?"

Great question. I checked. It is true that the data for all 12 non-domestic dioceses is not reported every year of 2002 - 2005. Data for 3 dioceses are missing in one or two years (2002 - 2003). To avoid the possible skewing of the data that the presence or absence of these 3 dioceses in the count might cause, I excluded them totally and did a new table which counted only the 9 non-domestic dioceses for which full data is available:

Virgin Is
Dominican Repub
Ecuador Litoral
Puerto Rico

Diocese excluded are:
Micronesia, Taiwan, Ecuador Central

Excluding those 3 dioceses very slightly diminishes the growth seen in the non-domestic dioceses, but the pattern still holds in spades.

09 January, 2007 18:23  
Blogger Karen B. said...

OK, question # 2. The "Christmas Effect"

Anyone who is really interested in this perhaps should read the discussions on T19 or Stand Firm


But here's the gist. Some of us have been wondering about patterns we've seen in the ASA data for the past 10-12 years, wondering if there is a "Christmas effect" whenever Christmas falls on a Sunday or Monday.

We think that the ASA decline may actually be worse in 2005 than reported because Christmas on Sunday is padding the numbers. (Christmas on Monday, as in 2006, may also result in padded numbers.)

The Parochial Report worksheet form specifically instructs that Sat. evg. data should be included in Sunday ASA when Christmas falls on a Sunday. Here’s the link:

(see page 6 and especially page 9)

Do I count Christmas?
No (unless Christmas Day falls on a Sunday; in that case, count your Christmas Eve services as well as any services on Christmas Day).

PATRICK: would you be able to share your parochial data with us?

What's your normal ASA for a normal (not Christmas or Easter) Sunday in 2005 and 2006?

What was your Christmas Eve Attendance in 2005 and 2006?

What data did you include for Christmas on your 2005 report (Saturday evening included?)

What will you (or have you) included for 2006? (Will you count Advent 4 and Christmas Eve services? Or just Sunday a.m. Advent 4?)

You can imagine that if Christmas Eve gets counted as a Sunday some years and not in others it makes for a problem with the data comparison!

09 January, 2007 18:39  
Blogger PSA+ said...

I'll get the numbers, but I am confident that our numbers were skewed last year due to the "Christmas effect" - in fact I believe I calculated it both ways.

Stand Firm has posted Karen's updated tables:
updated domestic v. foreign table

non-domestic data

09 January, 2007 18:43  
Blogger PSA+ said...

By "last year" I mean, of course, 2005. I haven't recovered yet from New Year's Eve on St. Lucia.

09 January, 2007 18:44  
Blogger Karen B. said...

By the way, I perhaps should have noted that ECUSA research director, Dr. Kirk Hadaway estimated in an e-mail to me that the "Christmas Effect" for 2005 could possible have added as much as 18,000 to the ASA figures, meaning the "real" ASA drop might be more along the order of 26,000 rather than the reported 8,000 decline. That drop of 26,000 is much more consistent with the drop in membership we see of 42,500. So, we're not talking about 10's or hundreds, but a possible discrepancy of THOUSANDS due to the possible Christmas effect from counting Christmas Eve data in some years and not others.

Also, rereading my comments from last night, I cringed at my use of the word "padding." I don't mean to imply that there is any intentional distortion of the data, merely that Christmas Eve attendance augments and inflates the ASA average some years.

10 January, 2007 09:42  
Blogger PSA+ said...

Thanks, Karen - and I understood what you meant by "padding." I haven't had the chance to dig up our numbers, but I do recall when calculating the 2005 statistics wondering if it was quite "fair" to include the Christmas services, but the directions seemed quite clear. Again, I am not by any means a numbers/statistics person, but I would be shocked (shocked, I say) were there not a huge "Christmas effect." Glancing at the 2005 service register prior to preparing Christmas bulletins this year reminded me that on Christmas Day 2005 we had about 70 people (we had only one service that morning - normally 2 on a Sunday), about 2/3's of an average Sunday. In 2006 (Christmas on Monday) we had 30. In 2005, we had fairly massive (for us)Christmas Eve attendance (included is ASA), this year somewhat fewer total but spread fairly evenly among the 4 Dec. 24 services (2 morning IV Advent service, 2 Christmas Eve services).

10 January, 2007 10:00  

Post a Comment

Links to this post:

Create a Link

<< Home