Beyond Reasonable Doubt

Factors Impacting Jurors’ Interpretation of Criminal Standards of Proof

Beyond reasonable doubt (BRD) is the standard of proof used to convict defendants charged with crimes in the English criminal justice system. If the decision maker perceives that the probability the defendant committed the crime as charged (based on the evidence) is equal or greater than their interpretation of BRD, than he/she will decide to convict. Otherwise, the decision maker will acquit the defendant. It is generally agreed that BRD should be interpreted as a .91 probability. The main goal of the present research was to better understand how members of the general public (potential jurors) quantitatively interpret BRD under different judicial instructions, case factors, attitudes, and demographic backgrounds. We used Wallsten and Budescu's (1995) theory of linguistic probabilities and their membership function (MF) method to capture peoples' quantitative interpretations of BRD. Recently, Dhami (2008) showed that this method was a reliable and valid measure of BRD. In a series of studies, we examined the impact of specific court-level factors, case-level factors, and juror-level factors on interpretations of reasonable doubt. The studies involved individual mock jurors at the predeliberation stage of a trial summarised in written form. This methodology is typical of jury research.

1. Judicial level factors

Legal academics and judges have expressed that the undefined version of BRD ("the defendant is presumed innocent unless the prosecution has proved guilt beyond a reasonable doubt") is difficult for jurors to understand. As a result several jurisdictions in the Anglo-American legal system have proposed other wordings with a view to aiding jurors’ understanding. In England and Wales for instance the Legal Studies Board advocates the wording, "The defendant is presumed innocent unless the prosecution has proved guilt beyond a reasonable doubt. Proof beyond reasonable doubt is proof that makes you sure of the defendant’s guilt". A very commonly used instruction in US jurisdictions is “Proof beyond reasonable doubt is proof that makes you firmly convinced of the defendant’s guilt.” We compared mock jurors’ interpretation of BRD under these 3 different versions of the judicial instruction: the standard undefined instruction, “sure” and “firmly convinced”. Mock jurors gave their interpretation of each standard of proof and rated how helpful they found the instruction.

The undefined BRD, “sure” and “firmly convinced” led overall to similar interpretations, very close to what is intended by the law (.92). However, for the undefined version we found significant differences in interpretation and understanding across demographic groups. This was not the case for the “sure” and “firmly convinced” instructions. Thus these two wordings of BRD are effective: They decrease demographic group differences in interpretation and in understanding while at the same time leading to the same evidentiary threshold.

2. Case-level factors

A further set of studies investigated the influence of case characteristics on jurors' interpretations of beyond a reasonable doubt. In the first of these participants read a case summary including information in which the perceived likelihood that the defendant is guilty was manipulated to be either high or low by making the defendant’s description comply with a “typical offender” stereotype or not, and where the perceived penalty was manipulated to be high or low, thus creating 4 different conditions. All participants gave their interpretation of reasonable doubt as elicited by the MF method. Two further studies followed-up on the impact of case factors. In these, participants read a case in which we varied the severity of the charges or the severity of personal consequences of imprisonment and elicited the interpretation of reasonable doubt. When the defendant did not fit the “offender stereotype”, mock jurors required a higher evidentiary threshold for BRD than when the defendant fit the stereotype. The severity of a possible sentence if convicted did not have an effect on the interpretations of BRD, nor did the severity of charges brought against the defendant or possible personal consequences a prison sentence would bring (as could be made known to the jury though pre-trial publicity). Thus this set of studies showed that while case characteristics exist that can lead to unintended variation in the interpretation of BRD, overall it is quite robust and not affected by severity of charges, severity of possible sentence, or personal consequences of a sentence for the defendant.

3. Juror-level factors

Finally jurors' attitudes were examined to see if persons with, for example, a prosecution bias adopt a lower interpretation of reasonable doubt, i.e. require a lower evidential threshold to find a defendant guilty. Attitudes were measured using the Juror Bias Scale (Kassin and Wrightman, 1984), Revised Legal Attitudes Questionnaire (RLAQ-23; Kravis, Cutler and Brock, 1993), Pre-trial Juror Bias Questionnaire (PJAQ; Lecci and Myers, 2008). While attitudes were important in predicting the case verdict, they did not predict how stringently the BRD instruction was interpreted.

Demographic differences were observed for jurors’ interpretation of BRD, as well as in their reported understanding of and confidence in applying the BRD standard. The differences occurred for the undefined version of the BRD, but not for the wording “sure” or “firmly convinced”. For the undefined version greater difficulties in understanding BRD and low confidence in applying it correctly were reported by people aged 30 or younger, women and persons without university education.

Overall this project showed that juror interpretations of BRD are affected by the wording of instructions, by demographic differences between jurors, and at least to a certain extent by case characteristics. However, they are not as much affected by case characteristics as could be feared, and are not influenced by juror attitudes found to be related to verdicts.

Dissemination conference

Project findings were presented at a successful 1-day dissemination conference on 6th March 2009 at the Faculty of Law, University of Cambridge. Eight international speakers discussed their empirical research on legal evidence and proof with an audience of practitioners and academics from a variety of backgrounds (police, lawyers, magistrates, academic lawyers, criminologists, sociologists, psychologists).



Dr. Katrin Müller-Johnson


Dr. Mandeep K. Dhami

Research Associate

Dr. Samantha Lundrigan

This 18-month project was funded by the Arts and Humanities Research Council (AHRC). It took place at the Institute of Criminology, University of Cambridge, from October 2007 to March 2009.