The effect of verdict system on juror decisions: a quantitative meta-analysis

Figures & data

Table 1. Mock trials for mock jurors.

Mock trial	Design	Presentation	Crime	2-Verdict		3-Verdict
				Convictions	Acquittals	Convictions	Acquittals
Smithson et al., 2007	B/W	Transcript	Homicide	9	43	5	47
Smithson et al., 2007	B/W	Transcript	Death by negligence	26	26	33	19
Smithson et al., 2007	B	Audio clip	Homicide	6	18	2	22
Hope et al., 2008	B	Transcript	Sexual assault^a	27	17	31	29
Hope et al., 2008	B	Transcript	Physical assault	24	45	16	57
Ormston et al., 2019	B	Mock trial video	Rape	95	119	75	140
Ormston et al., 2019	B	Mock trial video	Physical assault	69	147	43	175
Curley et al., 2021	B/W	Vignette	Homicide	27	37	14	50
Curley et al., 2021	B/W	Vignette	Homicide	15	49	12	52
Curley et al., 2022	B	Mock trial video	Homicide	46	32	28	51

Note: Mock jurors were assigned to the English/Anglo-American two-verdict and the Scottish three-verdict system, where each mock juror was assigned to only one verdict system in each mock trial but made decisions in two different mock trials. B = between-subjects design, B/W = mixed between/within-subjects design; convictions = guilty; acquittals = not guilty, not proven. The odds ratios of each mock trial are not shown but correspond to the estimates presented in the forest plot of Figure 2.

^a This sexual assault trial may be better classified as a rape trial due to the description of the vignettes in Myers et al. (2003).

Table 2. Number and percentage of convictions and acquittals for the Scottish three-verdict and Anglo-American two-verdict system.

Juror decision verdict system	Conviction		Acquittal		Total
	N	%	N	%	N	%
3-Verdict	259	15	642	36	901	51
2-Verdict	344	19	533	30	877	49
Total	603	34	1175	66	1778	100

Note: Convictions = guilty; acquittals = not guilty, not proven. Juror verdicts are pooled across mock trials.

Figure 1. Funnel plot of odds ratios (log scale) centred on the odds ratio estimate 0.59 (vertical line) with standard errors (SE) of estimates on the y-axis. The number above each black data point refers to the corresponding mock trial.

Figure 2. Forest plot of estimated odds ratios for 10 studies and mock trials based on a logistic model with random effects (Viechtbauer, 2010). CI = confidence interval. An odds ratio below 1.0 indicates a lower number of convictions for the 3-verdict system than for the 2-verdict system (see text for details).

Smithson, M., Deady, S., Gracik, L. (2007). Guilty, not guilty, or …? Multiple options in jury verdict choices. Journal of Behavioral Decision Making, 20(5), 481–498. https://doi.org/10.1002/bdm.572

 Web of Science ®Google Scholar

Hope, L., Greene, E., Memon, A., Gavisk, M., & Houston, K. (2008). A third verdict option: Exploring the impact of the not proven verdict on mock juror decision making. Law and Human Behavior, 32(3), 241–252. https://doi.org/10.1007/s10979-007-9106-8

 PubMed Web of Science ®Google Scholar

Ormston, R., Chalmers, J., Leverick, F., Munro, V. E., & Murray, L. (2019). Scottish jury research: Findings from a large-scale mock jury study. Scottish Government.

 Google Scholar

Curley, L., Munro, J., Frumkin, L., & Turner, J. (2021). Informing reform: The views of legal professionals on the unique aspects of Scottish Law. Medicine, Science, and the Law, 61(4), 256–265. https://doi.org/10.1177/0025802421992913

 PubMed Web of Science ®Google Scholar

Curley, L. J., Munro, J., Turner, J., Frumkin, L. A., Jackson, E., & Lages, M. (2022). Proven and not proven: A potential alternative to the current Scottish verdict system. Behavioral Sciences & the Law, 40(3), 452–466. https://doi.org/10.1002/bsl.2568

 PubMed Web of Science ®Google Scholar

Myers, B., Rosol, A., & Boelter, E. (2003). Polygraph evidence and juror judgments: The effects of corroborating evidence. Journal of Applied Social Psychology, 33(5), 948–962. https://doi.org/10.1111/j.1559-1816.2003.tb01933.x

 Web of Science ®Google Scholar

Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor package. Journal of Statistical Software, 36(3), 1–48. https://www.jstatsoft.org/v036/i03. https://doi.org/10.18637/jss.v036.i03

 Web of Science ®Google Scholar