At the National Center for State Courts seminar on Jury Management and in the book, Jury System Management by G. Thomas Munsterman, most issues were addressed relevant to large court systems having thousands of trials and hundreds of thousands of jurors. The suggested procedures seemed to have little relevance to less populated states like North Dakota with fewer than 325 jury trials in the entire state each year and fewer than 620,000 people. This report, Analysis of Juror Utilization in North Dakota, attempts to apply some of the ideas in the field of jury management to the state court system in North Dakota, specifically to correct the problem of calling in too many prospective jurors for jury duty.

In the jury trial system it is a necessary practice to bring in more prospective jurors for a jury panel than are actually needed for the jury plus alternate jurors. The parties involved have a right to challenge prospective jurors to keep persons of suspected bias from serving on the jury. There is no limit on how many challenges "for cause" the trial judge may allow. This leaves open ended the question of how many prospective jurors to call to the courthouse. How many is too many?

During the three years 1997-1999 the North Dakota state courts called in 23,509 people for jury selection for 969 felony, misdemeanor and civil jury trials. Only 17,181 of these prospective jurors were either sworn in for jury duty or were questioned and denied jury duty. The remaining 6,328 citizens (27% of those attending jury selection) spent 1 to 4 hours in the courtroom and were never called to the jury box. These "unreached" prospective jurors cost North Dakota $158,200 in juror fees ($25 each) plus mileage reimbursement during 1997 - 1999; about 10% of the state's jury budget.

The goal of this analysis matches the statement in the 1992 commentary to American Bar Association (ABA) Standard 13, paragraph (b): "Courts should reduce the number of prospective jurors that comprise a jury panel to the minimum number likely to be required to yield a jury in a given type of case." (Underlined by author). As stated in Jury System Management, "Although panel sizes in the past have been set by judges, by local custom, by rule of court, or by statute, they seldom have been calculated quantitatively on the basis of past experience." "...experience has shown that habitual panel sizes used in many courts have been extended far beyond the limits of prudence, with the result that the number of jurors 'not reached' are inflated" (pages 101, 102). Until this report, no quantitative or numerical analysis of panel sizes had been done in North Dakota.

The three specific goals of this research were (1) to highlight the problem of inflated jury panel sizes, (2) to provide a tool by which the judiciary can determine more appropriate panel sizes and (3) to encourage judicial self restraint in requests for jury panels. Improvement in juror utilization will lead to fewer trips to the courthouse for North Dakota citizens and a better juror experience for those who are called in for jury duty.

The database used in this report includes all jury trials in North Dakota from 1997 through 1999 in which jury selection was completed. Trials which settled before jury selection but after the prospective jurors had reported to the courthouse, were not included in this analysis for two reasons. (1) Jury panel size statistics are intended to be measures of how well (or badly) the trial judge and the jury clerk are doing in minimizing panel size. Since they have little control over last minute settlements or defendants who fail to show up for trial, such trials should not impact the measures of their work. (2) Keeping the information from these last minute cancellations in the database also has an adverse affect on the numerical analysis needed to predict the appropriate number of persons a judge should request for the jury panel. "Settlements on the courthouse steps" are a source of juror frustration and inflated juror costs but they should be recorded and analyzed separately.

Summoning yield (the number of persons actually reporting for jury duty versus the number of persons initially summoned) was not considered in this analysis although it does contribute to the problem of juror over-call. Because no database was available to analyze summoning yield, the panel size used in this analysis is the "number of persons reporting for jury selection" as reported by the clerk of court after each trial.

The data from 969 jury trials was analyzed in three groupings: by county, by judge and by trial type. Four trial types were considered: (1) criminal trials with 12 person juries, (2) civil trials with 9 person juries, (3) civil trial with 6 person juries and (4) criminal trials with 6 person juries. To highlight "best practices" to be emulated (and "worst practices" where remedial efforts should be applied), each group and trial type was compared on sixteen values (defined in Table 1). Also, for each of the four trial types, the four high volume counties in North Dakota were compared to each other and to state statistics on these same sixteen values.

In order to provide a tool by which the judiciary can determine more appropriate panel sizes, each trial type was further analyzed based on the numbers of jurors "reached". Table 2 is the frequency analysis for the 9 person civil trials. The first entry shows that at 8 trials, out of the 165 total, only 9 jurors were "reached" (ie, none of the 9 persons initially placed in the jury box were successfully challenged). In one trial (last entry in Table 2) 32 jurors had to be questioned before nine were sworn onto the jury. The "Cumm. Percent" column shows that 30 or fewer prospective jurors were "reached" in 97.6% of these civil trials. Only 4 out of the 165 trials needed more than 30 prospective jurors. So the "risk" of a panel size of 30 being too small and delaying the trial is only 2.4%.

A similar frequency analysis was done for each of the four trial types. Using this analysis a judge can request a panel size based on the amount of risk s/he is willing to assume.

The austere guidance on jury panel size in the official North Dakota Jury Selection Plan is inadequate. If it had been followed during 1997 - 1999, 18% of jury trials might have been delayed for lack of prospective jurors.

Fourteen of the 53 North Dakota counties hold 90% of all jury trials. Neither the population size of the county nor the number of trials held was a factor in how well the judges and clerks in these counties determined jury panel size. Jury trials in counties with smaller populations did not necessarily need larger jury panels to cover increased challenges. Comparison of the four largest counties showed where "best practices" might be found and where more effort could be applied.

The number of trials a judge presided over was not a factor in how well that judge determined the panel size s/he needed. This may be due to the judge relying on the "standard panel we've always used" in the county.

Comparative analysis of the trial types showed that the six person misdemeanor trials had the worst percentage of "unreached" jurors (unreached jurors / jurors reporting). However, many more people coming to the courthouse for jury duty were "not reached" in the 12 person trials. Any efforts to reduce juror over-call should be concentrated on reducing the "standard" panel size for the 12 person trials.

The frequency analysis as shown in Table 2 can be used by the trial judge to select a reasonable panel size. The following panel sizes were computed using the frequency distribution charts and applying an arbitrary 3% risk of having an inadequate jury panel:

Trial Type	Jury Size	PanelSize	% of trials Covered
Felony & Misdemeanor	12	35	96.8%
Civil	9	30	97.6%
Civil	6	22	96.5%
Misdemeanor	6	21	97.0%

Most of North Dakota's state judges assume more than this 3% risk; many using panel sizes smaller than those shown above.

The following recommendations further the goals stated in the 1992 commentary to ABA Standard 13, paragraph (a): "The goals of effective jury management are to increase the overall efficiency of jury system operations, to reduce costs, and to improve the attitude of the citizenry toward jury service and the court system."

1. Use this study to educate state judges on the use of smaller jury panels. Use the historical database to determine an acceptable level of risk and a jury panel size to match that risk for most jury trials.
   
   
2. Update the official Jury Selection Plan on panel size and encourage the use of standby jurors. Determine an acceptable degree of risk (statewide or by county) and set recommended panel sizes accordingly.
   
   
3. Consideration the use of multiple voir dire technique (selecting several juries from one jury panel) in all four larger counties.
   
   
4. Develop better statewide information on qualification and summoning yield. The judge and jury clerk need to know how many to summon, given that a certain number may be excused or simply fail to show on the day of jury selection.
   
   
5. Include the length of trial in the database to determine what effect this may have on needed panel size. Length of the jury selection process would also be useful.
   
   
6. Track of last minute settlements for separate analysis, not as a part of jury utilization.
   
   
7. Summarized jury utilization by county to show "best practices" across the state and to give the counties a way to track their efforts. Do data entry at the county level.
   
   
8. As panel sizes are reduced a method should be developed to highlight trials which were delayed because the panel was too small and trials where standby jurors were called in. Tracking these items will provide needed system feedback to determine if the percent of risk is set too high.
   
   
9. Make the state's effort at minimizing jury panels a part of the North Dakota Courts Annual Report. Show the public that the judiciary understands the real costs incurred in jury duty. The court has no control over the $25 fee but can change the process to minimize the number of people affected.
   
   
10. Continue looking at the by-county and by-trial statistics after implementing changes to gage the effectiveness of those changes.

To obtain a copy of this research paper, please contact:

Knowledge Information Services
National Center for State Courts
300 Newport Avenue 
Williamsburg, VA 23185
Phone: (800) 616-6164