News-Record.com

allen , the link doesn't work

He's out watching Mitch get fired. Watch for the link correction later tonight or tomorrow.

I presume Allen meant to link to this instead.

To me, rankings like this are fun puzzles (which of course means that I need to get out more). If you follow the link, you'll see how Democracy North Carolina got its ranking. To say the least, it's pretty arbitrary.

What DNC does is take 10 indicators of voting and registration, add up the ranks, and then rank the aggregate rankings. Guilford's aggregate rank is 487, which is the 34th lowest, hence the #34 rank.

But the ranks aren't simply added. DNC assigns double weights to two indicators (i.e. they add them in twice): the rate of turn-out of registered voters, and the rate of turn-out of voting-age people. No doubt those are important indicators. But why do they matter more than, say, the increase in turnout? I would have thought that'd be pretty important in gauging "voter activism." (But alas, Allen, Guilford ranks only 81st in that category.)

And why is the percentage of voting-age people who are registered get any weight at all? Who cares if they're registered but don't vote? That's not voter activism. If anything, it's voter inactivism!

Also, DNC ranks both the total voter turnout rate and the turnout rates of Democrats, Republicans, and Unaffiliated voters. Statewide, that covers all voters (apparently registered Libertarians are lumped in with Unaffiliated). Why would the turnout rate among voters of an individual party matter when you're already ranking total turnout?

Similarly for the break-out of white and black voter turnout. They're interesting measures, but it's hard to see what they add to the ranking.

A final sense in which various indicators cover the same statistical territory: it turns out that the increase in turnout (one of the indicators) is highly correlated with the percentage of registered voters who are black. My rudimentary estimate suggests that each 1% increase in the black share of registered voters moves a county up one place in the ranking for increasing overall turnout. (This is rudimentary because there are other factors at play as well.) Mind you, this isn't surprising in a year in which Barack Obama energized black voters especially. But it does make one wonder why both indicators are used as though they provided independent information.

All in all, this smacks of something put together by folks who don't generally work with numbers.

Still having fun with the numbers...

Guilford's best individual ranking is in the one category that shouldn't be ranked: percentage of voting-age people who are registered (not voting, just registered). We're 7th in that ranking, and if you take it out altogether, it drops us into 42nd place.

Aside from that, it's a curious measure. According to DNC, 98.9% of Guilford's eligible population was registered. Really? 98.9% only gets us into 7th place? Sure, because #6 in that category is Edgecombe, with 100.8%. #1 is Watauga, with 113.0%. You gotta give Watauga credit for that. It's not easy to register 113 people for every 100 people of voting age!

Presumably the population figure in the denominator came from a different year or different source than the voting-registration figure in the numerator. But still.

Very interesting points, Andy.
I would have expected greater turnout in Guilford, however.
Wonder how Democracy North Carolina's Bob Hall would respond.