The silence of the stats

Rangel points out a positively morbid WSJ column by Larry Eastland in which he attempts to make a point about abortions. He examines what fellow WSJ editor James Taranto politely calls the "Roe Effect":

The number of abortions accumulate in size and political impact as the years roll along. Like an avalanche that picks up speed, mass, and power as it thunders down a mountain, the number of Missing Voters from abortion changes the landscape of politics. The absence of the missing voters may not be noticed, but that doesn't mean its political impact disappears.

Eastland goes on to cite statistics that liberals are more likely to abort than conservatives. Assuming kids will vote like their parents, he reasons, America has lost more Democrats to abortion than Republicans in the decades since Roe v. Wade. This demographic shift was enough to give George W. Bush the margin of victory he needed in 2000.

It's fascinating, disturbing concept, which I think Eastland has pursued it. He's collected a lot of data, to his credit, but not nearly enough to clinch the case. Eastland acknowledges he's making an assumption about children voting like their parents. He doesn't, however, acknowledge another key assumption he's made: that abortion means fewer children. I'm not sure that's true.

I can think of a few examples of colleagues and friends who had a child very early, and as a consequence, haven't had other kids. On the flip side, those who wait until they're financially secure, older, married, etc, might be more inclined to raise a large family.

Is the effect I noted above enough to reverse the missing voter trend? It doesn't even have to be: Another Eastland assumption is that these missing voters would have voted like their parents. But how many adults switch parties as they age and start families? I remember that adage, "if you vote Republican at 18 you've got no heart, and if you vote Democrat after 40 you've got no brain."

In other words, growing responsibility, home ownership, and I think, starting a family -- all tend to make people more conservative. And I suspect conservatives have larger families, but which came first: the political leanings or the SUV full of kids?

As for Democrats and abortion, are they damned if they do, damned if they don't? And is the answer the same among Volvo-driving New Englanders as it is among poor Southern Baptists?

Eastland has stumbled onto a dissertation-sized topic. He needs to examine the assumptions that babies born in a world without Roe wouldn't cost their parents children down the line, wouldn't change their parent's political leanings, and would, in fact, vote like their parents. Further, I think each assumption's truthfulness varies by geography, religion, and socioeconomic class... even family size. It's not an easy question, and certainly Eastland's facile first analysis fails to convince me.

The one point Eastland is trying to prove is: liberal policies are costing them elections. Of course, this is true, but probably not in the convoluted, ghastly way he suggests.

And Rangel's got more -- about some liberal policies that might be saving pregnancies.

UPDATE: A thorough analysis is up at the Church of Critical Thinking. The comments include stats on infant mortality and voter rates for twentysomethings, further minimizing the "missing voter" effect.