(Tom Arthur / CC 2.0)

How Do Measures of Quality Influence Voters’ Choices for Judicial Candidates?

May 15, 2017

Right now, it’s incredibly difficult to cast an informed vote on the entire ballot. While most voters enter the voting booth prepared to vote for president or senator, they’re usually unprepared for everything else. As a result, voters guess, leave blanks, or stay home altogether.

As founders of BallotReady, our primary goal has been to increase vote quality. However, this is somewhat difficult to measure, as assessments of candidate quality are subjective. As a result, we have paid special attention to judicial candidates: the ratings they receive from bar associations constitute the best proxy for identifying candidates of high and low quality.

Elected judges play an important and often unacknowledged role in setting street-level criminal justice policy. Yet, for the most part, they are re-elected regardless of their courtroom behavior. In Cook County, Illinois, for example, no judge has failed to be retained since 1990, including a judge who pled legal insanity to assaulting someone in her courtroom.

In this past fall’s 2016 General Election, 5.5 million people cast a vote for President of the United States in Illinois, but only 13%—a little over 760,000 voters—cast at least one vote in a circuit court judge retention election. At the state level, we examined how measures of candidate quality affected Illinois election outcomes. The Illinois State Bar Association (ISBA) publishes a list of all judicial retention candidates by county, along with “yes/no” recommendations on retention, a clear signal of candidate quality. Analyzing data from the Illinois State Board of Elections from the 2012, 2014 and 2016 elections, we find that the average “yes” vote for ISBA-recommended judges was 76.8%, about 4.3 percentage points higher than that for judges not recommended by the ISBA. This small difference is statistically significant, but non-recommended judges still receive an average of 72.5% of yes votes, well above the 60% threshold needed to be retained. So, while 13% of voters felt informed enough to cast judicial retention votes, overall we found a limited effect of candidate quality on voting outcomes from all Illinois voters.

From our own data, we sought to measure whether judicial ratings influenced BallotReady users’ behavior. BallotReady allows voters to compare judicial candidates based on biography, endorsements, and multiple bar association ratings, and then to “save” their choices to create a sample ballot to bring into the voting booth.

In our model, we considered the effects of judicial ratings, ballot order, and the overall number of BallotReady views a candidate received on the overall total number of “saves” on the site - users indicating they would vote to retain a specific judge.

Our regression model, using results from 117 judicial candidates, shows that the total number of saves was associated with all three selected factors: judicial ratings, ballot order, and the overall number of views a candidate received on the BallotReady site. The models indicate that a percentage increase in judicial ratings of a candidate (where 100% indicates all bar associations rated a candidate highly qualified) resulted in 4.5 unit increase in the total number of saves of the candidate. By comparison, a candidate changing their ballot position by one place (for instance from 2nd to 1st on the ballot) only led to additional 0.93 total number of saves, and a standard unit increase in total views is related with 0.27 unit increase in total number of saves.

We believe our findings demonstrate that BallotReady users are sensitive to indicators of judicial quality and are using BallotReady to make more informed choices. However, in analyzing our data, we encountered an interesting dynamic—the vast majority of judges are recommended for retention by bar associations. In our data set, 93 candidates (80% of the total candidates) received only positive recommendations, and 105 candidates (90% of total candidates in our data set) received positive recommendations from 90% of bar associations.

This could indicate that most judges are, in fact, highly qualified and deserve to be retained. It could also reflect the realities of the politics of bar association recommendations. Knowing that most judges will be re-elected and will continue working with their lawyers, bar associations may reserve bad ratings for judges guilty of truly egregious actions. So even conscientious voters seeking out objective measures of quality may be unable to find the information they need to truly differentiate.

These insights lead to a number of takeaways for both the voter and the public administrator seeking to increase vote quality, but they may uncover more questions than answers. It is reassuring that many individuals in our country are seeking to vote informed for lesser-known races, and objective measures do seem to impact their decisions. Such objective measures, however, are few and far between, and those that do exist may themselves be unduly impacted by politics.

As BallotReady continues to cover more elections, we will continue to pay close attention to the complex relationships between bar association ratings, user behavior, and election outcomes. In the meantime, all Philadelphia residents have the opportunity to cast an informed vote this month—for judicial candidates as well as other municipal offices. BallotReady has partnered with Philadelphia’s Committee of Seventy to create a voter guide to the May 16 Philadelphia municipal election. Check it out at committeeof70.civicengine.com and don’t forget to vote and vote informed.


BallotReady is a nonpartisan voter guide to every race and referendum on the ballot that was created because elected officials matter—at all levels—and it should be easier to vote informed.  Voters who come to BallotReady enter their address to view the entire ballot. From there, they can compare candidates based on biography, endorsements, issue stances, and news articles, and then save their choices to use in the voting booth. Last fall, BallotReady was live in 12 states, covering 15,000 candidates, and saw over 1 million visits.

Tags: 

Categories: 

Contact Information

Fels Institute of Government
University of Pennsylvania
3814 Walnut Street
Philadelphia, PA 19104

Phone: (215) 898-2600
Fax: (215) 746-2829

felsinstitute@sas.upenn.edu