Predicting Liability Decisions: The Importance of Juror Characteristics and Perceived Case Strength

Introduction

The DOAR Research Center compiled data from a collection of focus groups and mock trials for civil cases that we conducted between January 2020 and December 2025. This includes data on over 2,000 focus group participants and mock jurors (N = 2,125) across 64 research exercises. The goal was to identify broad patterns in juror characteristics in civil cases—that is, what characteristics are generally associated with favoring one side over the other.

We used predictive statistical models to identify which juror characteristics most strongly predicted two key outcomes: jurors’ initial leanings^[1], and their final liability decisions. Our results painted a clear picture of the types of jurors who consistently favor the defense in civil cases. We also examined how jurors’ initial leanings relate to their final decisions about liability, and the extent to which perceived case strength^[2] predicts their final decisions. We review our findings for each of these below.

Our Sample

The selected research events (i.e., focus groups and mock trials) took place across the country, with the largest samples drawn from New York (35.8%), California (27.2%), Massachusetts (10.9%), Delaware (4.7%) and Connecticut (4.0%). For each event, participants were recruited to proportionally mirror the demographic characteristics of the jurisdiction. This resulted in an exceptionally diverse group of jury-eligible participants in our combined dataset. Our sample was evenly split between men (49.8%) and women (48.5%), with 46.6% under 45 years old and 54.4% 45 years or older^[3].

Key Findings

Juror Characteristics

Across the board, two traits consistently predicted jurors’ decisions: gender and socioeconomic status^[4] (SES). Even after controlling for other demographic factors,^[5]  gender was a reliable predictor of case leanings (p = .015)^[6]. Men were consistently more likely to be defense-leaning than women (Men: M = 3.55; Women: M = 3.38)^[7].

Socioeconomic status (SES) was also a significant predictor of juror leanings (p < .001).^[8] High SES jurors (i.e., wealthier and more educated jurors) were more likely to lean toward the defense. Controlling for other demographics, jurors with the lowest SES (generally high school education level, making less than $50K annually) were the most plaintiff-leaning (M = 3.38), followed by those with average SES (generally bachelor’s level education, making $50-99K annually) (M = 3.46) and those with the highest SES (generally graduate degrees with household incomes of $150K or more) (M = 3.55), reflecting a pattern of increasing support for the defense as wealth and education rise.

When we examined SES and gender together, a meaningful interaction emerged. The overall effect of SES on case leanings was driven primarily by men (p = .009)^[9]. As men became wealthier and more educated, they tended to lean more toward the defense, but women’s views stayed stable regardless of their SES. In other words, differences between men and women are small among lower SES jurors but widen substantially among higher-SES jurors. Gender may therefore be a particularly important factor important to consider in higher-SES jury pools.

Notably, these findings hold true even when we account for variability between individual research events. This means that gender and SES predict jurors’ initial leanings even in different types of civil cases. We also discovered that gender and SES predict jurors’ final, post-discussion or post-deliberations decisions about liability. This is unsurprising given that jurors’ initial leanings typically predict their final decisions about liability, as described below.

Jurors’ Initial Leanings as Predictors of Final Decisions About Liability

We examined whether jurors’ initial leanings predict their post-discussion (i.e., final) decisions about liability.^[10] As we would expect, jurors’ leanings strongly predicted their final liability decisions (p < .001)^[11]. The more defense-leaning the jurors were before group deliberations, the less likely they were to ultimately find liability. Conversely, the more plaintiff-leaning the jurors initially were, the more likely they were to find liability.

As shown in the figure below, each one-point increase in leaning towards the defense relative to the group average reduced the odds of a liability vote by nearly half (OR ≈ 0.48).  Stated differently, each one-point decrease in leaning towards the defense relative to the group average (and therefore perceiving the plaintiff’s case as stronger) more than doubled the odds of finding liability (OR ≈ 2.07).

Among jurors who initially leaned towards the defense, 81.9% found no liability, compared to only 18.1% of plaintiff-leaning jurors who found no liability. In contrast, 73.1% of jurors who initially leaned towards the plaintiff found full liability, compared to only 26.9% of defense-leaning jurors who found full liability. This indicates that jurors’ individual opinions are highly predictive of how they will vote in a group (i.e., during deliberations). At the same time, these results also demonstrate that jurors’ opinions can be swayed by group discussions, as evidenced by the fact that approximately one in four defense-leaning jurors found full liability after discussions (i.e., 76.3% defense-leaning found no liability compared to 23.7% who found full liability), and approximately one in five plaintiff-leaning jurors found no liability after discussions (i.e., 79.3% of plaintiff-leaning jurors found full liability compared to 20.7% found no liability).

Perceived Case Strength

Finally, we examined how much the overall strength of the defense’s case predicts jurors’ final decisions about liability. Case strength was measured by averaging jurors’ leanings across a particular research event. We found that case strength was a powerful predictor of liability (B = −2.08, p < .001).^[12]  Each one-point increase in perceived defense strength reduced the odds of finding liability by nearly 90%. In other words, while jurors’ case leanings were shaped by their demographics, their final decisions were highly sensitive to the overall strength of the evidence. No matter who the jurors are, one of the most important factors in determining liability is the overall strength of the case being put forward.

Moreover, perceptions of case strength—at both the individual juror level and at the group level—accounted for about 40% of the variation in liability judgements. This indicates that perceived case strength predicts liability judgments both across jurors and across different types of civil cases.

Importantly, once we accounted for perceived case strength, demographic differences no longer predicted liability (p = .087). However, demographics continued to predict jurors’ leanings, which in turn strongly predicted liability. Taken together, this suggests a mediation effect: demographic characteristics affect case outcomes indirectly by shaping how jurors interpret and evaluate the strength of the evidence.

Key Takeaways

Our results indicate that in civil cases—whether it’s intellectual property, employment, antitrust, contracts, securities, and so on—the people who are going to be the best defense advocates on the jury are often going to be the most like individual defendants or representatives (or witnesses) of the corporate defendants: men who are high SES.

The buck does not stop at jury selection, either. We found that the most important factor in predicting how you will fare is case strength. Yes, SES and gender matter—but only to a certain extent. Building a strong case—utilizing research to test and refine your case themes and see how they chalk up against the burden of proof and jury instructions—are critical for increasing your chances of success at trial.