Dr. Klumpp: What the data does - and does not - reveal about the 2024 Cook County judicial primary

FWIW is once again pleased to present a Guest Post by Albert J. Klumpp, a PhD in public policy analysis with a national reputation for expertise on judicial races, the author of several scholarly works analyzing judicial elections, and a generous and frequent contributor to FWIW for many years.

by Albert J. Klumpp

As the number of candidates seeking elective judicial vacancies in Cook County continues to dwindle—more on that later—studying the elective process itself becomes more and more difficult. Data analysis, by whichever qualitative or quantitative method, does actually require data. Go figure.

Nevertheless, I’ve finally had a chance to compile the results of our March 19 primary, and have taken the usual run at analyzing those results to provide whatever insight can be gained from them.

Based on the latest available results, roughly 678,000 ballots were cast in the primary, for a total turnout of 21.8%. As illustrated, the number is low:

…but not surprisingly low, given the lack of any substantive high-profile contests in either party for president or governor or senator.

The Democratic voters who did turn out did not neglect the judicial contests on the ballot. As the figure below illustrates, judicial votes cast as a percentage of total voters was the second-highest ever, just behind 2022.

As far as how those voters voted, that question is doubly difficult this year. Not only is the data set small—just seven two-candidate contests on the countywide ballot—but for the most part it was the same candidates who were slated by the party, favored by bar associations, and endorsed by progressive voices like Girl I Guess. And as well, nearly all of those same candidates secured the first ballot position in their contests. All of this overlap means that a statistical analysis will struggle to sort out how much of the vote for those candidates can be attributed to each source. (The jargon term for it is “multicollinearity.”) So unfortunately it simply isn’t possible to provide precise estimates of effect sizes in this primary, even within reasonable margins of error.

What the analysis does show more generally, though, is that the judicial vote was surprisingly different from that of what seemingly was a very similar primary two years ago (low turnout and nothing of significance at the top of the ballot). In this primary the big winner was party slating. Two years ago it was worth roughly 15 percent of the vote, a figure reasonably within historical averages. This year its value was well into the twenties, certainly 25 percentage points and possibly slightly higher.

This is a good example of how the value of a party endorsement does not uniformly rise and fall as turnout falls and rises. Given that the current Chicago mayor is popular with at least one wing of his party, versus two years ago when the mayor had lost support partywide, the shift in slating value is explainable and parallels similar situations in the past.

The Girl I Guess choices mirrored the party’s choices, so without the benefit of exit polling it isn’t possible to measure its influence separate from the party’s. Two years ago I measured its influence at roughly six percent of the vote, among a small electorate. There is no reason to think that the figure this time is different.

The big winner wo years ago was bar ratings, which I measured at a whopping 29 percent of the vote. The figure was difficult to estimate this year but is far lower and almost certainly in the single digits, likely no more than nine percent of the vote. This would be much more in line with historical averages than the 2022 result is. For the moment there is no clear explanation for the difference. Similarly, gender was an important variable two years ago at 17 percent of the vote, but it appears to have been less than half as influential this year. This is more understandable; the absence of women at the top of a ballot typically leads to a smaller advantage for female judicial candidates.

Finally, the tentative estimate for the first ballot position was roughly six percentage points, but its margin of error was too large to be confident of the figure.

Subcircuit contests always provide a useful comparison for conclusions drawn about countywide voting. The observations here do look to be supported by the subcircuit contests in this primary, although just as with the countywide ballot there is a shortage of useful data in the subcircuits (only nine contests and 22 candidates).

Speaking of subcircuits, one major factor that always influences subcircuit contests is campaign spending. A proper look at the numbers on spending will have to wait until the next quarterly campaign finance documents are submitted by candidates in July. But it does appear that, true to form, spending was a major factor in subcircuit contests. An imbalance in campaign funds looks to have helped decide at least four of the nine contests, and possibly five. I will return to this with full data in a future post.

To conclude, and very much not to bury the lead, here is the single most important chart for the 2024 primary:

The overall number of candidates per vacancy this year was 1.53, the lowest in the modern history of judicial contests in Cook County. And as the chart shows, this was not an anomaly but rather the continuation of a long-term trend. The trend was discussed here in FWIW recently and certainly will be discussed again, because obviously it has consequences far more important than my ability or inability to provide vote percentages. It also should raise questions about the wisdom of creating new elective judgeships at a time when filling elective judicial vacancies is becoming a significant problem.