Sunday, November 7, 2010

A (Mostly) Thoughtful California Electorate

Regular readers of this blog may assume that I think the recent election demonstrates thoughtfulness on the part of California voters because mostly liberal candidates won statewide offices. That is not the point I wish to make here.

Voters demonstrated sophistication and thoughtfulness, first, because they failed to allow the onslaught of money to purchase their brains. As we all know, Meg Whitman vastly outspent her victorious opponent, Jerry Brown, yet came up short in the polling booth. Proposition 23, which would have "suspended" air pollution control regulations (probably forever, due to the ridiculous criterion established for lifting the suspension), was supported in large part by out-of-state oil companies. The voters saw through it and defeated it by a hefty margin -- 61.2% "No" to 38.8% "Yes."

Dramatic evidence of careful voting is also exhibited by a comparison of the results of Propositions 20 and 27. The former was designed to extend the mission of the pre-existing Redistricting Commission -- which was created by an earlier proposition to put the post-census redistricting process for state officials in the hands of an impartial citizens' commission rather than the gerrymander-prone legislature -- to federal Congressional seats as well. The latter would have scrapped the Redistricting Commission altogether. Thus, Propositions 20 and 27 were, for all practical purposes exact opposites. It is difficult to imagine a rationale for voting for or against BOTH of them; the more logical approach would be to vote for one and not the other (in either direction).

In fact, most Californians did precisely that -- voted for one but not the other. Statewide, Proposition 20 passed by a vote of 61.4% to 38.6%; Prop 27 failed by a vote of 59.6% to 40.4%.

Clearly, some voters did vote for or against both of these propositions; otherwise, the passage rate for one would have been precisely equal to the failure rate of the other. By subtracting the "No" vote percentage of Prop 27 from the "Yes" vote percentage of Prop 20 (and taking the absolute value, i.e. using a statistical method that ignores the direction of the deviation, which in this case is irrelevant), one can calculate a "discrepancy" score that might be considered a measure of the degree to which voters acted in an intellectually consistent manner. Statewide, that discrepancy score is 61.4 minus 59.6 -- or 1.8. (See last week's post for a related issue on intellectual consistency.)

In mostly liberal Los Angeles County (where Brown trounced Whitman by 62.7% to 32.5%), the discrepancy score was 1.1. Reasonably liberal Santa Barbara County (Brown got more votes than Whitman but not by a huge margin) had a discrepancy score of 1.32. Extremely liberal San Francisco had a score of 2.78.

Conservative counties Kern, Orange, and Riverside, where major statewide contests went Republican by substantial margins, had discrepancy scores of 3.22, 5.35, and 6.09, respectively.

The pattern (did I just see some of you liberals smiling?) did not hold across the board. Fiorina clobbered Boxer in conservative Kings County, with a discrepancy score of 2.28. Fresno County, where both Whitman and Fiorina came out on top, had a moderately low discrepancy score of 2.07. In conservative and tiny Butte County, the discrepancy score was a lowly 0.85.

In liberal Humboldt County, where Brown got 56% of the votes for governor to Whitman's 36%, the discrepancy score was a whopping 10.51. What in the world do you suppose those voters were smoking?

1 comment:

  1. I don't think it's entirely logically inconsistent to have opposed both props 20 and 27, though I am in favour of prop 20. The position that the commission should only govern state elections is not logically contradictory, and would imply voting no on both. It's not a very sensible position, but it does not meet the strict standard of being logically inconsistent. It is fair to say that if you support either ballot initiative, it is logically inconsistent to support the other. But that does not imply you cannot support neither position. Because of that, you would need pairwise breakdowns to determine voter's logistical consistency. Given the roughly 2/3 to 1/3 split, one would expect the group of double no voters to create a statistical artifact pushing support for 20 down by twice as much as support for 27. Assuming (and that's a big assumption) that voters were otherwise either logical or reverted to the mean positions, a 1.8% gap would imply a 2.7% double-no vote.