Dateline: April, 2021, Issue 1
Jurors in criminal cases are instructed to find guilt when proven beyond a reasonable doubt. Jurors also are provided imprecise definitions of what constitutes reasonable doubt: an actual and substantial doubt, an abiding conviction, being firmly convinced, being morally certain, not wavering or vacillating, having no real doubt, causing a careful person to hesitate, and (tautologically) a doubt that is reasonable.
Due to the imprecise nature of the reasonable doubt standard, the certainty jurors require of themselves to find guilt varies across jurors, cases and defendants.
Conklin (2020) studied the certainty individual jurors feel is minimally necessary to find guilt in three hypothetical criminal cases (assault of an elderly woman based on a dare, male gas station clerk who sold alcohol to someone underage, female gas station clerk who sold alcohol to someone underage). Each research participant was given a summary of one of the three hypothetical criminal cases which were designed to have either a sympathetic or unsympathetic defendant. All particpants stated if they thought the defendant was guilty beyond a reasonable doubt, how certain they were as to the defendant's guilt and, importantly, the least amount of certainty required to be 'beyond a reasonable doubt' as to someone's guilt.
When asked to affix a minimum percentage to the reasonable doubt standard, 14.5% of participants reported a number of 50% or less, a certainty level below that specified by law for the preponderance of the evidence standard. Conklin's research is not an anomoly. Stoffelmayr and Diamond (2000) report that mock jurors returned the same percentage of guilty verdicts regardless of whether the standard they were given was beyond a reasonable doubt or preponderance of the evidence.
Conklin rejects the idea that jurors are unable to distinguish among the various standards of proof. Conklin instead suggests that jurors ratchet up and ratchet down the standard of proof they are given in response to extralegal factors such as defendant sympathy, defendant gender, and their own political leanings.
Particpiants in Conklin's research required a lower burden of proof for reasonable doubt when the defendant was unsympathetic (minimum certainty required = 62%) and a higher burden of proof when the defendant was sympathetic (minimum certainty required = 80%). Liberal participants (minimum certainty required = 82%) required a higher burden of proof than did conservative participants (minimum certainty required = 75%). Said differently, the less sympathetic the defendant and the more politically conservative the juror, the lower the minimum certainty the juror required to find a defendant guilty beyond a reasonable doubt.
The male defendant received more favorable treatement than the female defendant. While participants were more certain about the male defendant's guilt, the male defendant was convicted at a lower rate (male defendant convicted 55% of the time vs 60% for the female defendant). The reason for this surprising result is that participants required less certainty to convict the female defendant (minimum certainty required = 74%) than the male defendant (minimum certainty required = 86%).
Even slight changes in the minimum certainty required to find guilt beyond a reasonable doubt produce drastic differences in conviction probabilities. Conklin offers this example: Assume that the majority of cases that go to trial have a defendant who is 70% to 95% likely to be guilty. A 75% burden of proof would result in convictions in 85% of these cases, while an 82% burden of proof would result in convinction in only 38% of these cases.
Conklin concludes that jurors adjust the certainty required for proof beyond a reasonable doubt to reach their desired verdict, and that these adjustments can have significant effects.
Source Conklin, M. (2020). Reasonable doubt ratcheting: How jurors adjust the standard of proof to reach a desired result. North Dakota Law Review, 95(2), pp. 281-290.
Source Stoffelmayr, E. & Diamond, S.S.. (2000). The conflict between precision and flexibility in explaining 'beyond a reasonable doubt', Psychology Public Policy and Law, 6(3), pp. 769-787.