Copyrighted material.
Posted 7/03 by permission of publisher.
Journal of Police and Criminal Psychology, 18 (1),
9-14.
Prediction of First Ballot Mock Jury Votes by the
Analytic Juror Rater
C. Gary Pettigrew
Dawson Psychological Associates
And
Lewis Unglesby
Unglesby, Koch, & Reynolds
Two scales of Abbott’s (1987) Analytic Juror Rater (AJR) were used with 24 mock jurors to predict first ballot mock jury votes. Each participant observed one of two mock trial proceedings involving an actual second degree murder case. In a moot courtroom, they heard arguments from attorneys and witnessed examination of the defendant and actors portraying witnesses. The Cosmopolitan Lifestyle Scale of the AJR successfully predicted first ballot votes of participants (p < .02), while the Non-Authoritarian Scale showed a non-significant trend in the hypothesized direction. It was concluded that, in cases where evidence is not strong, the AJR may lend modest assistance to the attorney using peremptory challenge to eliminate venire members who may be biased against a defendant.
Abbott (1987) described a systematic approach to ranking potential jurors based upon a secondary analysis of a nationwide pooled sample of 15,579 respondents from the General Social Survey for 1972-1983, sponsored and conducted by the National Science Foundation, Russell Sage Foundation, the Roper Public Opinion Research Center, and the National Opinion Research Center. His resulting tool, the Analytic Juror Rater (AJR), attempts to predict four types of values for jury candidates on the basis of certain demographic characteristics. The values are authoritarianism, economic orthodoxy (conservatism), cosmopolitanism, and racial tolerance (whites only). The basic predictors are age, sex, race, occupation, marital status, and spouse’s occupation.
The current study explores whether AJR classification of 24 mock jurors into three preference categories is related to first ballot (pre-deliberation) votes in two mock trials. Using data from 225 criminal cases, Kalven and Zeisel (1966) demonstrated that final jury verdicts are usually identical to the majority vote on the first ballot. When all jurors voted not guilty on first ballot, the final verdict was always not guilty. If 7 to 11 jurors voted not guilty on first ballot, 91% of trial verdicts failed to convict. Conversely, a not guilty verdict was rare if that vote was not obtained on the first ballot from at least half of the jurors. In a replication study, Sandys and Dillehay (1995) found from interviews with 142 felony jurors that the correspondence rate between first ballots and final verdicts was similar to that found by Kalven and Zeisel.
The second degree murder case presented to the mock jurors in this study is thought to be one in which the evidence was ambiguous. It involved a couple who argued while traveling in the husband’s vehicle after a night of drinking with friends. The woman apparently jumped out of the vehicle. Her husband said he then stopped, changed gears, and backed up to retrieve her. In doing so, the physical evidence supported the conclusion that his vehicle struck a traffic sign, which then came down and fatally hit his wife in the head. This involvement of a third object, the trajectory of which could at best have been only imprecisely predicted, would appear to be an awkward and uncertain means of intentionally murdering someone, and suggests accidental death. On the other hand, there is no dispute that the victim’s husband was emotionally aroused and at the wheel when the incident took place. Because of the evidentiary ambiguity, it was hypothesized that attitudes predicted by selected AJR scales would have a significant impact upon first ballot voting.
Participants
The 24 participants were hired through an employment agency. The agency only accepted applicants who qualified for jury duty, and an effort was made to achieve racial (black/white) and gender balance.
On the first mock jury, there were 6 males and 6 females, among whom 5 were black and 7 white. Age ranged from 34 to 57, with a mean of 43.8 (SD = 7.29). Years of education ranged from 2 to 15.5, with a mean of 12.4 (SD = 3.46). All 12 were employed. Two earned less than $15,000 per year, 4 earned $15,000 – $35,000, 5 earned $36,000 – 60,000, and 1 earned more than $60,000. Three described themselves as politically conservative, 5 as moderate, and 4 as liberal. Four were Democrats, 4 were Republicans, and 4 belonged to neither party. Five were married (never divorced) and 2 were single (never married). There were 5 divorced participants, two of whom were re-married. Eleven had children.
Serving on the second mock jury were 7 males and 5 females, among whom 6 were black and 6 white. Age ranged from 18 to 72, with a mean of 37.8 (SD = 15.08). Years of education ranged from 12 to 16, with a mean of 13.3 (SD = 1.57). Eight were employed. Four earned less than $15,000 per year, 3 earned $15,000 – $35,000, 3 earned $36,000 – 60,000, and 2 earned more than $60,000. Six described themselves as politically conservative, 4 as moderate, and 2 as liberal. Seven were Democrats, 2 were Republicans, and 3 belonged to neither party. Six were married (never divorced) and 5 were single (never married). One had divorced and re-married. Nine had children.
Instruments
The AJR requires that one (or two) of the following value scales be chosen as theoretically applicable to the attorney’s case:
The Cosmopolitan Lifestyle scale was also selected, since the present case involves an episode of late-night socialization in a cocktail lounge and apparent alcohol intoxication. This scale applies (in part) to behavior regarded in a cosmopolitan subculture as legitimate, although it may violate legal or traditional religious norms. The six General Social Survey items from which Abbott derived this scale concern the following aspects of lifestyle: legalization of marijuana, premarital sexual relations, pornography, drinking, the right to an abortion, and the right to commit suicide. Abbott sees the central theme as allowance of maximum freedom on the part of the individual in the conduct of his personal daily life.
Procedure
Jurors were given a pre-trial questionnaire that included basic information necessary to describe the sample and apply the AJR.
The Non-Authoritarian and Cosmopolitan Lifestyle Scales were calculated and separately analyzed, resulting in a list of jurors for each scale in rank order of preference. The first third of the list (8 jurors) was assigned to Category A (good), the next third to Category B (acceptable), and the final third to Category C (undesirable).
Two mock trials were conducted, with 12 jurors in each. In order to maximize the realism of mock trial proceedings, participants in each trial were empanelled in a moot courtroom, the interior setting of which was typical of actual courtrooms. Closed circuit television cameras were well hidden among the furnishings. Jurors were told by the first author that the simulated trial in which they would participate was an effort to obtain their reaction to a real case scheduled for trial in the near future. They were informed that the actual defendant and his attorney would be taking part, an attorney would be assigned to play the role of opposing counsel, and other persons would act in the roles of witnesses and judge. A summary of background facts was then read to them, followed by attorney opening statements of both sides. Key witnesses took the stand and were questioned, followed by closing statements. The jury was then isolated at a conference table, and asked to take a first ballot before beginning deliberations. Balloting was secret from other jurors, given in writing to the first author and identified by name. For purposes of preparation for actual trial, jury discussion was observed by consultants and defense team on closed circuit television, then terminated after 15-20 minutes. Observers were satisfied that mock jurors took their roles seriously, a conclusion supported by the fact that one jury requested additional and uncompensated time for deliberation.
Results
Table 1 presents outcome data for the Cosmopolitan Lifestyle Scale. Use of this scale with tripartite juror classification was significantly related to first ballot vote in the desired direction (p < .02). Both A (good) and B (acceptable) jurors predominantly voted not guilty, while C jurors tended to vote guilty.
In our sample, a non-significant Pearson correlation of .252 (p = .234) was obtained between Cosmopolitan and Non-Authoritarian scales. Using the General Social Survey sample, Abbott (1987) calculated a correlation of .01 for these scales.
Discussion
Results of this study provide modest support for the hypothesis that, where evidence is weak, at least the Cosmopolitan Lifestyle Scale of the AJR may have some utility in reducing bias related to prejudicial values if used at voir dire in some criminal cases. The method has the advantage of low cost when compared with profiles based on local community survey and/or mock trial data analysis. A major AJR disadvantage may be its lack of detailed case specificity. That is, it predicts broad values rather than specific reactions to evidence presentation, arguments, appearances, and persons. AJR utility is confined to jury selection, and is therefore not useful in improving the presentation of evidence for optimum impact, increasing witness credibility, or enhancing the efficacy of attorney technique.
The superior contribution of the Cosmopolitan Lifestyle Scale in the present sample seems consistent with conclusions drawn from Hastie, Penrod, & Pennington’s (1983)large-scale study of 828 mock jurors, who were presented with a case involving barroom violence. The juror’s world knowledge concerning the domain of events relevant to the case was the most important individual ingredient in decision making. This knowledge included understanding of the social customs governing barroom quarrels, the effects of alcohol on physical coordination, and the acuity of visual perception under various viewing conditions. In the present study, this experiential knowledge also appears relevant to the judgment of events, and may more likely be represented among those who score higher on the Cosmopolitan Lifestyle Scale than by those who score lower on the AJR Authoritarian Scale. The latter purportedly measures predisposition to impose retributive punishment and coercion in controlling illegal behavior (see Centers, Shomers, & Rodriques, 1970, Mitchell & Byrne, 1973, Snortum & Ashear, 1972). Abbott (1987) distinguishes this concept from that of the authoritarian personality. As Abbott defines it, a less authoritarian tendency may theoretically relate more to lenient sentencing than to non-conviction, although one source component of the AJR Authoritarianism Scale, attitude toward death penalty, has been repeatedly correlated with tendency to convict (Cowan, Thompson, & Ellsworth, 1984, Ellsworth et al., 1984, Filkins, J. W., Smith, C. M., & Tindale, R. S., 1998, Jorow, 1971, Mertz, Miller, & Balance, 1966). However, there is some evidence that research which investigates the impact of jury death qualification upon verdict may actually be measuring the effects of the death qualification process itself rather than attitude. Haney (1984) found that exposure to such proceedings significantly increased the juror’s likelihood of voting guilty.
In actuality, there is a great deal of uncertainty about the relationships among variables explored in jury research. When decisions are made upon the basis of mock jury data, it is critical to the analysis that the best parts of each side’s evidence be presented, in order to minimize the risk of confounding the result. The extent to which, and the circumstances under which, values impact juror votes is difficult to measure, as is the accuracy of prediction of such values from a demographic instrument such as the AJR. As the current study illustrates, the method requires an exercise of judgment in selecting the most relevant scale(s). Where two scales are selected, subjective weighting is necessary if a mathematical combination is desired, and subjective category boundaries are required if a cross-tabulation scheme is attempted. In summary, the AJR may be a useful point of departure in advising the attorney how to proceed at voir dire, but it cannot entirely replace the role of experience and intuition in developing a strategy for selecting whom to challenge among the venire members.
References
Abbott, W. F. (1987). Analytic Juror Rater. Philadelphia: American Law Institute.
Centers, R., Shomers, R., & Rodrigues, A. (1970). A field experiment in interpersonal persuasion using authoritative influence. Journal of Personality, 38, 392-403.
Cowan, C. L., Thompson, W. C., & Ellsworth, P. C. (1984). The effects of death qualification on juror’s predisposition to convict and on the quality of deliberation. Law & Human Behavior, 8 (1-2), 53-79.
Ellsworth, P. C., Bukaty, R. M., & Cowan, C. L. (1984). The death qualified jury and the defense of insanity. Law & Human Behavior, 8 (1-2), 81-93.
Filkins, J. W., Smith, C. M., & Tindale, R. S. (1998). An evaluation of the biasing effects of death qualification: A meta-analytic/computer simulation approach. In T. S. Tindale & L. Heath (Eds.), Theory and Research on Small Groups: Social Psychological Applications to Social Issues, Vol. 4. New York: Plenum Press, pp 153-175.
Haney, C. (1984). On the selection of capital juries: The biasing effects of the death-qualification process. Law & Human Behavior, 8 (1-2), 121 – 132.
Jurow, G. (1971). New data on the effect of a death qualified jury on the guilt determination process. Harvard Law Review, 84, 567-611.
Kalven, H. & Zelsel, H. (1966). The American Jury. Boston: Little, Brown.
Mauet, T. A. (1996). Trial Techniques (4th Ed.). New York: Aspen Law Business.
Mertz, R., Miller, G., & Balance, L. (1966). Open- and closed- mindedness and cognitive conflict. Journalism Quarterly, 43, 429-433.
Mitchell, H. E., & Byrne, D. (1973). The defendant’s dilemma: Effects of juror’s attitudes and authoritarianism on judicial decisions. Journal of Personality and Social Psychology, 25, 123-129.
Saks, M. (1976). Social scientists can’t rig juries. Psychology Today, 9, 48-50, 55-57.
Sandys, M. & Dillehay, R. C. (1995). First-ballot votes, predeliberation dispositions, and final verdicts in jury trials. Law and Human Behavior, 19 (2), 175-195.
Snortum, R. J. & Ashear, V. H. (1972). Prejudice, punitiveness, and personality. Journal of Personality Assessment, 36, 292-296.
Thompson, W., Cowan, C., Ellsworth, P., & Harrington, J. (1984). Death penalty attitudes and conviction proneness; the translation of attitudes into verdicts. Law and Human Behavior, 8, 95-113.
Correspondence may be addressed to:
C. Gary Pettigrew, Ph.D.
Dawson Psychological Associates
7510 Highland Road
Baton Rouge, LA 70808
