Friday, April 13, 2012

Dr. Klumpp reviews countywide judicial results

In peacetime, the old saying goes, generals spend all their time preparing to fight the last war. I suppose we might be accused of doing something similar here, poking amongst the ashes of the March 2012 judicial primary, looking for patterns in the dust that will be helpful in 2014.

But Albert J. Klumpp, PhD, a Research Analyst with the Chicago firm of McDermott Will & Emery LLP, has made a scientific study of results in judicial races, applying statistical principles with academic rigor to produce patterns that can really be used to help predict future outcomes.

Dr. Klumpp recently provided this illustrative table to FWIW to explain his findings. This table looks at countywide contests in 2012, and in the aggregate from 1986 to 2010 (long term) and 2002-2010 (near term).

1986-20102002-20102012
Female Name14.612.012.2
Irish Name11.411.19.6
Black Name7.18.06.8
Top Ballot Line4.46.78.5
Slating10.39.213.0
Recommendations10.814.18.4
Spending (per $10k above avg.)0.20.20.03


Klumpp writes that the numbers in this table are "all percentage-point numbers, as in, percent of the vote." The spending variable is per units of $10,000 above the average spending level within each contest; the person with $100,000 more to spend than the average candidate in his or her race would have 10 times the number shown in the table -- and it would still be a small number.

"Recommendations," Klumpp says, is a composite "that refers to the combined impact of the Chicago Tribune, Chicago Sun-Times, Chicago Bar Association and Chicago Council of Lawyers." If more than one candidate in a race shares one of these advantages -- such as more than one woman or Irish-surnamed candidate in a contest, the statistical advantage is diluted ("divided among them," Klumpp says).

With regard to the 2012-specific numbers, Klumpp worries that the 2012 figures may be based on too small a data set (43 candidates). "But," Klumpp writes, within this limitation, "I think there are three important things in the 2012 numbers. First, the high value for slating -- more than twice as high as in 2010, and one of the highest values in many years. An exceptionally large number of people supported the slated candidates this time." And, Klumpp adds, assuming that all the committeemen remain faithful to the slate, "one of the big perks of being slated is that" the slated candidate ordinarily gets the full advantage of this percentage boost.

But what about those instances -- some of which we've been able to document here in post-election posts -- where individual committeemen 'dumped' slated candidates for someone else? We've been able to document instances where, in a given ward, the inclusion of a non-slated candidate on an ward palm card in lieu of a slated candidate has provided a big boost for that candidate vis a vis that candidate's performance in other wards.

Klumpp acknowledged "there are instances in which a ward or township organization will deviate from a party slate, but I haven't seen evidence that it's widespread enough to justify adding into a big-picture model like mine." Does this mean that the occasional deviation from the slate is factored into the overall slating number? Klumpp would not go that far; he preferred to state "the slating measure is strictly a countywide value."

On the other hand, Klumpp stated, "If you can give me evidence that a significant number of local organizations dumped a slated candidate and pushed someone else in a particular contest, then there would have to be an adjustment to the slating variable for that contest -- provided that we can answer the question about what proportion of the voters presented with conflicting slates would use one versus the other." That would be a big and presumably difficult question to answer because, even in wards where committeeman back one or more unslated candidates, voters do receive party mailings that include the orthodox county judicial slate.

Klumpp also sees the large statistical boost for slated candidates as "further evidence of Mayor Emanuel's popularity, beyond the Theis victory."

"I can't prove that directly without a survey," Klumpp conceded, "but, indirectly, the voting patterns in the Supreme Court primary strongly suggest it. So does that big 13.0 number I got for the other countywide contests." Although Emanuel is not the chairman of the county Democratic Party, or even a committeeman, people "tie the mayor to the party in their minds, and that will affect the value of party slating. That also helps explain the poor slating number from two years ago although," Klumpp added, "the ugly contests at the top of the ballot was probably the main reason."

Moreover, Klumpp said, "to be fair, I have to think that [Cook County Board President and 4th Ward Committeeman] Toni Preckwinkle deserves some credit too. Her personally endorsed candidates didn't do nearly as well as Emanuel's, but I think some voters just feel a little more comfortable with the local party in 2012 from seeing how the two of them have been conducting their business, and at least to that extent, it boosted that slating number."

Also important, Klumpp said, was that "the overall value of recommendations fell by roughly 40 percent. The drop was caused by the Sun-Times' decision not to endorse political candidates and to instead print a giant table containing nearly 1,300 individual bar association ratings. It's exactly what happened between 1996 and 2004 when they didn't endorse judicial candidates and printed giant bar tables -- a 40 percent reduction in the overall value of recommendations. You could see this coming a mile away. This is something that I've found without exception in all of the judicial voting I've looked at, both contests and retention, in whatever jurisdiction."

Klumpp had some bad news for this blog, too: "Voters will not use voluminous or complicated forms of information on judicial candidates." That bodes ill for both the famous Alliance 'grids' and the Organizing the Data posts here. On the other hand, Klumpp stated, "If a trusted source like a newspaper reduces it all down to a simple endorsement, they'll respond. But they simply will not do 'homework' for low-visibility contests that are of little or no direct relevance to them."

Although, Klumpp said, "there are civic-minded, responsible voters out there who diligently study judicial elections, there just aren't very many of them. At least not enough to remove poorly rated retention candidates, for instance." He added, "Unless we start doing a better job to educate young people and new voters about the judicial part of the ballot, it's not going to change."

The third key factor in analyzing this year's judicial primary results, Klumpp said, is "the continuing importance of the first ballot position. The 8.5 figure may be a slight overestimate, but the first-position advantage has been growing over the years, particularly since the county's switch from punch-card voting to the touch-screen and optical-scan ballots." Klumpp said he is not quite sure why this it, but this phenomenon "should be triggering some serious debate over whether or not the county should implement ballot rotation and do away with the first-position lottery. In fact," Klumpp stated, "ballot position was determinative in two of last month's contests. Jesse Reyes would have lost if William Boyd or Ellen Flannigan had been first on the ballot, and Stanley Hill would have defeated Karen O'Malley if Hill had been listed first instead of O'Malley."

As for the seeming minimal impact of campaign spending, Klumpp admitted "0.03 may be a little low." However, he said, "even the historical value of 0.2 is trivial. It's very difficult to get any kind of impact from campaign spending in these lower-court contests." These numbers would seem to undermine the argument of those who fret about the corrosive influence of money in judicial elections. Money may be decisive in high-level judicial races, such as Supreme Court contests, but it seems almost immaterial in races for the Circuit Court.

No comments: