Lay Episcopalians for the Anglican Communion (LEAC) has released the findings from the blind, anonymous survey mailed out to members of the ECUSA House of Bishops about a month ago.  As expected, the response rate was low: only 80 of 298 surveys were returned.  Moreover, the results cannot be considered indicative of the views of all ECUSA bishops.

LEAC tries to put a positive spin on the results, but it really can't be done.  Two press releases have been issued with content that, strangely, differs in crucial aspects.  One is posted at LEAC's website, while the other was released through PR Newswire.

The headline and opening sentence of the PR Newswire release are seriously misleading.

Episcopal Bishops, If Voting Secretly, Would Oppose Church's Stance on Homosexual Agenda Items Adopted in 2003, a Lay Poll Reports

WASHINGTON, March 31 /PRNewswire/ — Episcopal bishops in the United States, if voting secretly now, would turn away from two critical votes in 2003 which championed a pro-homosexual agenda, according to a survey by Lay Episcopalians for the Anglican Communion (LEAC).

In fact, no statements about the voting preferences of Episcopal bishops can be made on the basis of the survey.  Given that respondents effectively selected themselves into the survey sample (i.e., only those who took the initiative to respond are included), not everyone in the sample frame had an equal chance of being included in the results, and therefore the LEAC survey results are not based on a random sample.  In cases involving such self-selected samples, the respondents cannot be assumed to represent the larger population to which they belong; the results refer only to those who actually responded.  In this case, the findings are not generalisable to the entire ECUSA HoB.  (See this blog post for more discussion of self-selected samples.)

The headline and opening text of the press release at LEAC's own site are much more circumspect in this regard.

Bishops responding to LEAC survey reject key homosexual agenda items

WASHINGTON, DC (3/29/2006): LEAC's blind, confidential survey of the U.S. House of Bishops found that 56.25% of respondents now would disapprove of the 2003 General Convention resolution which led to consecration of Bishop V. Gene Robinson, and 57.5% would oppose provisions for church blessing of same-sex domestic partnerships, another of the convention's historic resolutions.

From a statistical perspective, there is a world of difference between referring to, on the one hand, "bishops responding" and "respondents" (per the release posted at LEAC's site) and, on the other, "Episcopal bishops" in general (per the PR Newswire release).

The release posted at LEAC's site also says:

A satisfactory sample of 27% of the nation's active and retired bishops participated. Response was likely suppressed during the data collection period by a special, unfavorable letter to bishops from Presiding Bishop Frank Griswold. The derogatory letter said the sponsor of the research was anonymous, although the covering letter explaining the study and questionnaire was on LEAC's letterhead, with name and address. Griswold's erroneous words and unfavorable tone apparently caused reduced responses.

The survey is emphatically not "satisfactory" from a statistical perspective.  Not because it's small—public opinion pollsters commonly make statements about the voting preferences of US citizens based on samples of 1500 or less—but because, as just discussed, it is self-selected, not randomly selected, and therefore not necessarily representative of all ECUSA bishops.

As for the effect of Bp Griswold's letter criticising the survey, LEAC's contention is debatable.  In a mail-out survey, it is not unusual for responses to be returned quickly in relatively large numbers and then to taper off quickly after the initial rush.  Those who wish to respond typically do so shortly after receiving the survey form, especially if the survey is short, as in this case.  (That's not to say that LEAC is wrong, merely that the pattern of returns observed in this instance is not an uncommon experience in mail-out surveys.)

In my previous post on this survey, I suggested that LEAC would have been better off hiring a professional polling firm.  Admittedly, that would have been an expensive proposition, but the results would have been more useful.  So, I was puzzled to see this in the PR Newswire release: "The study was aided by pro-bono opinion research professionals".   Did those professionals point out to LEAC the problems with an anonymous, mail-out survey?  If not, I think they were remiss.

Even on its own terms, I don't think the LEAC survey provides much encouragement for traditionalist Episcopalians.  Eighty surveys were returned, of which 56.25% disapprove of the GC2003 decision to confirm consecration of Bp Gene Robinson, and 57.5% percent oppose providing for church blessing of same-sex domestic partnerships.  In numbers, 45 and 46 respondents, respectively, oppose the GC2003 decisions.  Those numbers, while not insignificant, would appear insufficient to change the direction of the ECUSA HoB.

Finally, I'm not up on ECUSA church governance in general, or how decisions are made at General Convention in particular, but based on this post at Blog of Daniel, changing the opinion of the House of Bishops is only part of the challenge LEAC has taken on.

LEAC links via Thinking Anglicans, where you can find links to more information about LEAC.
Blog of Daniel link via Drell's Descants.

Previous related post: Survey of ECUSA House of Bishops