Polls tell you what they tell you - and that's all they tell you

In my first public service job, one of my lesser responsibilities was to analyse the data from the periodic tracking polls we did about how people viewed the education system and the government's management of the system. As a result, I actually understand a little bit about polls, margins of error, the formulation of meaningful questions and cetera.



One of the things I learned was that polls tell you what they tell you - and that is all they tell you. Yesterday's poll about voting intentions doesn't tell you how people are going to vote next week. It merely tells you how a random sample of people questioned yesterday said they intended to vote. Within limits, it may help you to extrapolate possible outcomes. But at the end of the day, Dewey still did not defeat Truman.

Sam Clemons (aka Mark Twain) is attributed with the saying, "There are three types of lies: Lies; Damned lies; and Statistics." Personally, I don't think it's so much that statistics lie as that people assume statistics show things they simply don't.

*

Back in that public service job, we had assessment data which showed that the more time children spent with the math teacher, the lower their marks. Really. No, really.

*

So, did the statistics show that spending time with math teachers lowered students' math marks?

*

Maybe. I suppose.

*

But isn't it more likely the other way around? That students who are having trouble in math are more likely to spend extra time with the math teacher?

*

Aaaaah. That makes more sense.

*

In any event, over the last week or so, we've been inundated with all sorts of statistics about attendance in the Church of England and about attendance in the so-called Anglican Church in North America (ie, the schismatics). Thinking Anglicans comments on the former here and on the latter here. Canon Mark Harris comments on the ACNA numbers here and here.

*

It is curious that Christianity Today describes the CofE numbers as "stable," while the Telegraph focusses on a 12% decline in Christmas attendance. Same news release, mind you.

*

Similarly, there are any number of schismatic spin doctors happy to claim that the ACNA numbers document a major split in North American Anglicanism and a host of defections from both the Episcopal Church in the US and from the Anglican Church of Canada. In fact, the numbers show a small denomination with attendance about 5-7% of the combined attendance of the real Anglican churches in Canada and the US. Since about 16% of the ACNA claimed attendance is from congregations of the Reformed Episcopal Church, which split from mainstream Anglicanism in the 1800s, it is more than a bit bizarre (perhaps, may I say, a tad dishonest) to imply these are recent defectors from the Episcopal Church.

*

So what do all these statistics tell us?

*

I don't know, really. They tell us that an Anglican-style sect in formation has an average Sunday attendance of about 83,000. And that's about all they tell us.

*

Notwithstanding my previous post, membership numbers and Sunday attendance numbers don't necessarily, on their own, tell us much about the vitality of congregational life. A parish which challenges its members to grow spiritually may grow, or it may lose members who are uncomfortable with the challenge. We need other indicators (and I don't know what they are) to tell us that.

*

In any event, this whole thing made my head start to hurt.

*

So instead, let me reflect on something important.

*

I'm going to become a grandfather in a few months. I'm still adjusting to the idea. Surely I'm too young.

*

To Hades with all the meaningless bumph about unconfirmable attendance claims. Instead, let me refer you all to my daughter's blog, where she's reflecting on important matters like her "early morning porcelain prayers."