Jun 21, 2011

What it means if you say you say grace before meals

As a quote to sum up the "God Gap" in American politics, David E. Campbell's is perfect:

"If I know whether you say grace before meals every day, I can probably predict how you vote," said Notre Dame political scientist David E. Campbell.

It's perfect, which is probably why it has cropped up everywhere since Campbell said it earlier this month. It's fast on its way to becoming the quote on this topic, the one always pulled out to bolster the point. All we need now is for it to get misattributed to someone more famous than a political scientist and get repeated in a thousand churches and a million Bible studies.

You can already tell -- this quote is going to be spoken of as simply, obviously and blindingly true.

So it's too bad it's slightly off, slightly but importantly wrong.

What the prognosticator needs is not to know whether or not you say grace regularly, but whether or not you say you say grace every day.

The "God Gap" in American politics may or may not actually be about religious practice. Thing is, we don't really know. What we know is how people self-report their practices, how they self-identify, and how that correlates to voting patterns.

As far as I know, there are no major studies where people have been followed around and, say, every time they pray before they eat a check was made on a clip board. Thus the question has to be, how do they know -- how does David E. Campbell know or the reporters or I know -- who prays before what meals? And, truth is, we don't.

No one's counted bowed heads.

Instead, they've called people up, as polling is usually done these days, and asked them, "do you say grace before meals every day, weekly, monthly, on special occasions, or never?"

I don't have any quibble with the method, per se, but we ought to be clear about what it is we can actually measure like this. What we know is what people say they say, not what they do in their lives.

This turns out to be important, because we know that people are not always honest in their self reports. One might well ask, why would they lie?, and that's a good question, but we know that they do, whatever the explanation for that turns out to be.

There's very clear evidence, for example, that Americans consistently over-report their church attendance. Any even basic comparison of how many people say they're in church on a Sunday with actual attendance records shows that pollsters are consistently given false information. No wide-spread study has been done to see how many people are lying about being in church Sunday morning. It's possible, though, that as many people lie about being in church as are actually in church.

That raises the possibility of the same scenario the other way around: perhaps some percentage of people deny being in church on Sunday, but really do find a back pew somewhere.

The point being that what we know is what people want us to know. That is, their identity. Their public presentation of self.

What we know -- and it's important to point out -- is that measuring self-reports is not the same as measuring what's being self-reported.

The question that has to be asked here is: what is being measured. That might seem silly, since, obviously, these are misrepresentations/over-representations/lies. But that doesn't mean they're simply meaningless. People do say what they say and, presumably, have some reason for the answers they give. The false statements mean something -- there has to be some discernible reason why people answer one way or the other even though that reason isn't a basis in facts.

Put it another way: we know that the statement "I was in church last Sunday" doesn't mean the speaker was in church last Sunday, and that "I say grace before meals everyday" doesn't mean that the speaker says grace daily before meals. But, what is true of everyone who says those things?

Here's one thing we know about all these people for sure: whether or not they go to church and pray before meals, they represent themselves as doing so to strangers.

I.e., it's part of their public image. It's how they present themselves to the world.

It's what they say when strangers call and ask, and, in that, is how they think of themselves, maybe, but certainly how they present themselves. We know this simply from the fact that this is how they answered. This is the one thing we can be certain is true, given the data we have. The data we have is about the images of themselves that people construct for the public or at least for those they don't know who call on the phone and ask questions about prayer life.

It's these images that we can correlate with voting patterns.

So, but, maybe that's incidental, and it's still the case that there's this "God Gap." Perhaps we can just modify the quote so oft-repeated now, and say, "If I know whether you say you say grace ..." etc., noting that the correlation isn't between religious practice and voting tendencies (necessarily), but between two parts of people's public presentations of their identities.

It seems to me though that the whole point of their being a gap is for it to be understood as something more substantial than self-perception. The whole work of the quote was to establish that this gap was real, or had some basis in how we live and how the people on the opposite side of the ongoing political squabble live, which then reinforces this idea that they're different from us on the other side.
But if the point is only that those to perceive themselves to be and present themselves as being religious also perceive and present themselves as voting Republican -- then haven't we just repeated the thing that was supposed to be explained?

For if it's only a matter of perception -- or that's what it is we know or can know with this data, and if the quote just repeats that -- than rather than documenting the gap, this works to establish it. It makes the exact perception it purports to document. Every time it's repeated, it affirms and cements and constructs this perception of the correlation, which makes it more likely that, in the future, those with predilections one way will report and even mis-report themselves as also doing the other, associated things, and as not doing the things that are understood as contradictory to the self-identity they want to present.

We might think here of the way online pornography viewing is self-reported, vs. the actual statistics. There's reason to not believe Little League coaches in Mississippi when they say they don't watch porn. On the other side, certain kinds of college age guys are likely to exaggerate their porn watching. It's not about the act itself; it's the image. And if a social scientists were going to say something important about this, it wouldn't just be repeating unquestioned the image that's out there.

If the point of this quote and the idea of the "God Gap" was just that Americans perceive the "God Gap" and picture themselves as opposite sides of it, or present themselves in congruence with the image one would expect them to have even in cases when it's not true, then, really, it doesn't seem like it was helpful. What we wanted to know was about something more than just, yeah, people do think this.

What we have here is the self-perception of a cultural identity fed to a pollster who feeds the perception of a cultural profile that includes certain practices and voting patterns to the political scientists and pundits who feed it back to the people who have that perception already (largely), so it affirms the divide, all without ever substantiating anything more than "this is how we think of ourselves."

This is what this quote is going to do -- "If I know whether you say grace before meals every day, I can probably predict how you vote" -- around and around and around.