Juror attitudes and biases in sexual assault cases

Foreword | Sexual assault has among the highest rates of acquittal and lowest rates of proven guilt compared with other offences. Given that more than 70 percent of sexual assault incidents are not reported to police and only about one in 10 reported incidents results in a guilty finding, increasing conviction rates for sexual assault is a key issue for the criminal justice system. This paper presents findings from two recent studies conducted by the Australian Institute of Criminology. These show that juror judgements in rape trials are influenced more by the attitudes, beliefs and biases about rape which jurors bring with them into the courtroom than by the objective facts presented, and that stereotypical beliefs about rape and victims of it still exist within the community. As jurors are members of the community and are randomly drawn in order to be representative of it, the two studies together indicate that successful prosecutions of sexual assault will remain low until we acknowledge that jurors interpret what they see in light of their own beliefs, experience and expectations. We need to know what these belief structures are and how they directly impact upon judgements in rape trials, if conviction rates are to improve.

Toni Makkai

Sexual assault is one of the hardest offences to prosecute. Primarily this is because these offences are usually committed in private, there is often little or no corroborating evidence and it is usually one person's word against another (ODPP & AFP 2005). The onus is upon the Crown to prove beyond reasonable doubt that a defendant is guilty; the defendant is not required to prove innocence. This means that prosecutors have the extremely difficult task of convincing a judge or jury that the offence took place, that the complainant did not consent, and that the defendant knew that the complainant was not consenting or was reckless as to such consent. When it comes down to the word of one person against another, with no witnesses and no other supporting evidence, making a case beyond reasonable doubt is a hard task indeed.

Faced with such difficulties and limited resources, prosecutors often need to make hard decisions about which cases to prosecute and which to let go. While the public interest is a factor, prosecutorial decision-making about whether to prosecute a sexual assault case also usually involves a judgement about the probability of success, based on expectations of how judges and juries are likely to view the complainant and her story. Since defendants can, and often do, choose not to give evidence in court, the credibility of the complainant is crucial to whether she is likely to be believed, the ability of the prosecution to convince a jury beyond reasonable doubt and, hence, the probability of a guilty verdict. Through experience, prosecutors develop knowledge about what jurors are likely to look for in a sexual assault complainant, the attributes that might make a good or bad impression on jurors and whether the circumstances surrounding the incident (e.g. prior relationship, alcohol consumption) are likely to reduce the plausibility of the complainant's story in the eyes of jurors. A study of prosecutorial decision-making by Lievore (2004) found that prosecutors were more likely to prosecute sexual assault cases when:

  • the victim was injured
  • the victim physically or verbally expressed non-consent
  • the assault was severe (involved some level of threat or force, weapon use)
  • there was additional evidence linking the defendant to the assault
  • the defendant used force
  • the defendant was a stranger.

In a study of New South Wales sex offences, Fitzgerald (2006) found that criminal proceedings were more likely to be instigated where the incident involved aggravation and was reported to police earlier rather than later.

Essentially, the above are indicators which either support the evidence given by the complainant (injury, severity, weapon use) or which reflect community preconceptions about 'real' sexual assault cases (rape is committed by strangers, a victim would always clearly articulate that she was not consenting to sex and would physically fight against it). Most sexual assault cases, however, do not have supporting evidence such as injury, weapon use or witnesses (Lievore 2003). Juror beliefs and attitudes about what a sexual assault case looks like and how a victim of sexual assault would behave therefore become critical to understanding why complainants may or may not be believed by jurors, and whether a particular sexual assault case is likely to achieve a guilty verdict.

This paper draws on two studies, conducted recently by the Australian Institute of Criminology, to illustrate that:

  • pre-existing juror attitudes about sexual assault not only influence their judgements about the credibility of the complainant and guilt of the accused, but also influence judgements more than the facts of the case presented and the manner in which the testimony is given
  • myths and stereotypes about rape and sexual violence are common within the general community. Since jurors are members of the general public and are randomly drawn in order to represent the views of the community, attitudes they bring with them into the courtroom will, to a large degree, reflect the attitudes and beliefs of the wider community. Gauging the extent of community myths and stereotypes will provide some indication of the types of beliefs likely to be prevalent among jurors.

Sexual assault can be prosecuted in lower courts, magistrates courts and higher courts (including Supreme courts) across Australia. The court in which a matter is prosecuted can vary with jurisdiction, seriousness, plea entered and/or appeal. In general, less serious charges are dealt with by lower courts. A large number of sexual assault matters are dealt with in lower courts. More serious charges proceed first to a committal hearing at a lower court to determine if there is a case to answer. If there is, they proceed to a higher court where they are finalised. The committal hearing filters out cases which should not proceed to the higher courts. Consistent with this, a higher percentage of sexual assault charges heard in the NSW local courts are dismissed without a hearing compared with the NSW higher courts (Fitzgerald 2006). However, just under one quarter of sexual offences are dismissed without a hearing in the NSW higher courts, again revealing the difficulties prosecutors face when trying to mount a case.

The Australian Bureau of Statistics (ABS) publishes data on finalised matters in the higher and magistrates courts across Australia. Of all sexual assault matters finalised in both these courts each year, approximately one quarter result in defendants being acquitted (ABS 2007a; 2006; 2005). This is similar to the proportions acquitted in the NSW study (Fitzgerald 2006).

While about 60 percent of all finalised sexual assault cases in the higher courts, as reported by the ABS, involve the defendant pleading guilty, this overstates the proportion of all defendants who plead guilty, because it does not take into account those whose cases have been dismissed. Fitzgerald (2006) found that, after taking into account those whose cases were dismissed or disposed of (27%), only 35 percent of all sexual assault defendants in NSW higher courts in 2004 pleaded guilty. In the NSW lower courts only 24 percent pleaded guilty.

If the defendants who pleaded guilty in the ABS higher courts data are removed from the equation, this means that 58 percent of those who pleaded not guilty in the higher courts where a decision was finalised were acquitted in 2005-06, 57 percent in 2004-05 and 61 percent in 2003-04 (Figure 1). Only four in 10 defendants were found guilty when the case went to trial and a verdict was reached. For cases going to trial in NSW higher courts, 67 percent of sexual assault defendants were acquitted in 2004 (Fitzgerald 2006), a slightly higher proportion than that recorded nationally by the ABS.

When the fact that many cases that go to court are those judged to have a higher probability of success is considered, it becomes clear that prosecuting sexual assault successfully is a difficult task.

Figure 1 also shows that the percentage of finalised defendants pleading not guilty who were acquitted in the higher courts was highest for sexual assault cases. Except for homicide (in which about half of finalised defendants pleaded guilty) the offence of sexual assault generally has the smallest proportions of adjudicated defendants pleading guilty (ABS 2007a; 2006; 2005). When those whose cases are dismissed without a hearing are included, this difference is accentuated (Fitzgerald 2006). Compared with other types of offence, it seems that defendants charged with sexual assault are less likely to plead guilty and more likely to take their chances in front of a judge or jury. Given the high rates of acquittal shown in Figure 1, the odds would appear to be in their favour.

Figure 1 : Acquittals in the higher courts for finalised defendants pleading not guilty (percent)

Acquittals in the higher courts for finalised defendants pleading not guilty

Source: ABS 2007a; 2006; 2005

Although consent is at the heart of most rape trials (did the complainant consent to sex or not?), there are inherent difficulties in establishing whether it occurred. Legal constructions of consent vary across jurisdictions and there is wide debate about the appropriateness of trying to define it. Even where consent is interpreted to mean 'free and voluntary', this may be confusing for juries where consent can also be given reluctantly or after persuasion (CLRD 2007). While judges do give some direction to juries about what constitutes consent, this is minimal, varies across jurisdictions and has been criticised as inadequate (NSW Criminal Justice Sexual Offences Taskforce 2005). Further, knowing what consent is, as defined by the judge, is a far cry from being able to determine in a criminal trial whether it occurred. In NSW, for example, there is no statutory definition of consent. The NSW Bench book provides the following directions for a judge to give to a jury in a sexual assault trial:

Consent involves conscious and voluntary permission by the complainant to engage in sexual intercourse. It can be given verbally, or expressed by actions. Similarly, absence of consent does not have to be in words; it may be communicated in other ways. However the law specifically provides that a person who does not offer actual physical resistance to sexual intercourse is not, by reason only of that fact, to be regarded as consenting to the sexual intercourse.

A key difficulty here is that while the absence of consent does not necessarily have to be communicated in words or actions (a person being sexually assaulted may freeze and say and do nothing but this does not equal consent), jurors may query a claim of non-consent if a complainant did not verbalise it or physically resist. Jurors may also question how a defendant could reasonably be expected to know that the complainant was not consenting if she gave no overt sign that she was not consenting. Legal developments in several jurisdictions have moved towards a requirement that the defendant demonstrate the action they took to determine whether the complainant was consenting - this is also currently under consideration in New South Wales (CLRD 2007). While legal developments are important, the divide between legal definitions and how juries interpret and make decisions about consent is large. The legal guidelines on consent also do not help jurors when its communication, or lack thereof, is either contested in court or ambiguous.

Alcohol consumption by both parties adds an extra dimension of ambiguity - too much renders the complainant legally incapable of giving voluntary consent, but how much is too much? Again, and for good reason, there is a legal reluctance to be definitive about this. While alcohol or drugs may render a person incapable of giving voluntary consent, it is also possible that someone may be affected by alcohol or drugs but still be capable of giving voluntary consent. It comes down to a subjective judgement about fact and degree, contributing to the ambiguous nature of such cases. A UK study (Finch & Munro 2005) found that participants in a simulated trial and in focus groups tended to attribute responsibility for the ensuing sexual intercourse to the complainant when she had consumed alcohol, despite understanding that alcohol can impact on the ability to offer meaningful consent. When both the complainant and defendant were equally intoxicated there was agreement that the defendant should not be held criminally liable for the intercourse that followed. These findings illustrate that the role of alcohol in sexual assault cases, as perceived by jurors, is not straightforward and may work against the complainant, regardless of her capacity to give meaningful consent.

In cases where consent is disputed and there is no supporting evidence, determining whether consent occurred cannot be established through fact. Rather, jurors must resort to making a probability judgement about consent through their assessment of the testimony given, the surrounding elements of the case, the context in which the incident occurred, the plausibility of each story, and their own impressions, knowledge and experience of people and human affairs. As noted by Finch and Munro (2005) this leaves a wide discretion in the hands of the jury and raises the likelihood that differently composed juries would reach different decisions in cases where the facts were identical.

Juror beliefs and attitudes

Given that the issue of consent is not straightforward in contested sexual assault trials, jurors look to other cues to help them judge whether it occurred. In cases where only the complainant gives evidence, she and the story she is telling are all that jurors have available to them. Since there is no objective way to test the truth of the complainant's testimony, her perceived credibility becomes crucial to the judgements that jurors make. However, the credibility of a complainant is not objectively determined. Whether a juror perceives a complainant as credible is not simply related to the consistency of her story or the manner in which the testimony is presented. This was shown clearly in a recent Australian experimental mock sexual assault trial study in which 210 members of the general public participated as jurors (Taylor & Joudo 2005). Across 18 trials and despite the fact that only the manner in which the complainant presented her testimony varied (in court, CCTV or via pre-recorded video) - everything else was held constant - jurors had different opinions about her credibility and the plausibility of her story. Even within the same jury, opinions about her credibility differed, which could not be attributed simply to how she presented her testimony or what she said since all jury members watched exactly the same trial.

Figure 2: Influences on juror beliefs about guilt in sexual assault trials

Influences on juror beliefs about guilt in sexual assault trials

Source: Taylor & Joudo 2005

Rather, individual juror differences in terms of demographics and the beliefs, attitudes and expectations that the jurors (members of the public) brought with them into the courtroom were what primarily influenced their judgements about the credibility of the complainant's testimony and guilt of the accused (credibility and guilt were highly correlated). On average, and consistent with previous research, males were significantly less likely than females to perceive the complainant as credible. Higher credibility was also associated with more positive and less stereotypical attitudes toward rape victims in general. Influences on jurors' personal beliefs about the guilt of the defendant are shown in Figure 2, based on the variables measured in the study. Less favourable attitudes towards rape victims in general were significantly associated with:

  • being male
  • lower income
  • political conservatism

and were linked to believing that the defendant was probably not guilty.

Stronger personal beliefs in guilt were significantly associated with:

  • higher levels of education
  • personal knowledge of sexual assault victims
  • positive attitudes toward rape victims in general
  • higher perceptions of complainant credibility
  • low empathy with the defendant.

Although not measured quantitatively, one of the key qualitative insights obtained during discussions with jurors following the study related to the high degree to which many jurors believed many of the myths which surround rape in general, and the influence of these beliefs on their perceptions of the complainant's credibility. During these discussions, it became clear that many of these issues had been discussed during the jury deliberations as part of the process of weighing up the evidence. Many jurors had strong, often stereotypical, expectations about how a 'real' victim would behave before, during and after a sexual assault, and these expectations affected their perceptions of the complainant and how they interpreted her testimony.

Examples which arose regularly and worked against the complainant included:

  • the complainant flirted and danced with the defendant (some degree of encouragement)
  • she did not scream or shout for help (why not?)
  • there was no evidence of injury and no medical evidence to support her claim (surely there would be evidence of injury or DNA)
  • the complainant went back to the party afterwards; she did not leave immediately (she would leave)
  • the complainant composed herself and pretended nothing had happened (why would she pretend nothing had happened?)
  • she continued to work with the defendant for two weeks after the incident (a rape victim could not continue to work with the person who had raped her)
  • the complainant did not report the incident to police for two weeks (why did she delay in reporting the rape?)

Some jurors indicated that they had advanced some of these arguments as a rationale for not believing the complainant and hence choosing a not guilty verdict. Others stated that they did not hold these views about rape and did not believe they were relevant in reaching a verdict. A lot of argument within the jury room centred on the truth and relevance of some of these issues. However, for many, at least some of these issues raised sufficient doubt about the truth of the complainant's testimony to prevent a guilty verdict. Although 16 of the 18 juries were unable to reach a unanimous verdict within the allotted time period (two juries reached a unanimous not guilty verdict), three quarters of all jurors in the study favoured a not guilty verdict.

Unfortunately for prosecutors, this type of scenario in which sexual assault occurs is common. Rape is not always committed by strangers, victims do not always scream for help, obvious physical injury is uncommon and the majority of victims do not report the incident to police at all. The reality for many sexual assault victims is that, as long as misinformation about rape and stereotypical beliefs about how a victim would behave exist within the community, the likelihood of convincing a jury that a sexual assault did occur in the absence of supporting evidence will remain low.

The finding that many of the jurors in the study not only brought their own beliefs and misconceptions about rape into the courtroom, but also interpreted the complainant's testimony in light of them, highlights the need to identify the nature and prevalence of beliefs about rape and sexual assault held more widely within the community. This is because jurors are members of the public - identifying the types of attitudes and beliefs held within the community should shed some light on those that prosecutors and complainants need to counter in the courtroom.

Table 1 : Community beliefs about rape (percent)
  Males Females
Agree Disagree Neither/unsure Agree Disagree Neither/unsure
** significant to p
a n=487 males and 516 females
b n=488 males and 508 females
Source: Taylor & Mouzos 2006
Women are more likely to be raped by someone they know than by a stranger a 72 15 13 80 10 10
Women rarely make false claims of being raped a 65 23 12 68 22 10
Women often say 'no' when they mean 'yes' a 15 75 10 15 79 8
Women who are raped often ask for it b 7 91 3 4 95 1
A woman cannot be raped by someone she is in a sexual relationship with b 6 91 3 4 94 1
Rape results from men not being able to control their need for sex b 44** 52 4 32 63 5
Sexual assault can be excused if the victim is heavily affected by alcohol b 5 95 0 3 97 0
Sexual assault can be excused if the offender is heavily affected by alcohol b 4 96 0 2 97 1

A recent survey of a random sample of the general Victorian population, conducted on behalf of VicHealth (Taylor & Mouzos 2006), revealed that myths and stereotypes exist within the community to varying degrees. Table 1 shows that:

  • one in 10 respondents believed that women are more likely to be raped by strangers and another one in 10 couldn't say
  • about one quarter disagreed that false claims of rape are rare and one in 10 couldn't say (if a juror starts with the assumption that women often lie about rape, this will influence the way s/he interprets testimony)
  • fifteen percent agreed that women often say 'no' when they mean 'yes' and one in 10 couldn't say (testimony that the complainant said 'no' is unlikely to convince jurors with this belief that she did not consent)
  • seven percent of males and four percent of females agreed that women who are raped often ask for it
  • forty-four percent of males and 32 percent of females believed that rape results from men not being able to control their need for sex (responsibility for rape is therefore removed from men because it is not within their control).

These findings do not augur well for rape victims trying to convince a jury that they did not consent, or for the prospects of improving the low rates of conviction.


Increasing conviction rates for sexual assault is a key issue for the criminal justice system. It is important for ensuring that justice is done and essential if victims of sexual assault are to place their trust and confidence in police, the courts and the prosecution process. With high acquittal rates and the re-victimisation and trauma associated with having to give their testimony in great detail and sometimes several times over, there is currently very little incentive for victims to pursue their case in court.

It is time to acknowledge that jurors do not (because they cannot) make objective judgements about consent and guilt based on the facts presented to them in court. Jurors actively interpret what they see and hear, based on their own knowledge, experiences, attitudes, biases and expectations. Identifying these beliefs and expectations and understanding how they impact on judgements relating to credibility and guilt, is essential to understanding some of the barriers to successful prosecution and how prosecution rates might be improved in the future. It is particularly crucial in light of the fact that most sexual assaults occur in private, with no corroborating evidence, and involve one person's word against the other. Research is needed to directly investigate the link between pre-existing juror attitudes and judgements about credibility and guilt in the courtroom. Only when the nature of this link has been established will it be possible to know where to target education within the community, how to counter misinformation in the courtroom, how to better inform jurors about the nature of sexual assault, and how to increase the rate of successful prosecutions.


This paper refers to complainants and victims as female and defendants as male. While both men and women can be victims and perpetrators of sexual assault, 83 percent of all sexual assault victims who came to the attention of Australian police in 2006 were female (ABS 2007b). The 2002 Crime and Safety Survey found that 93 percent of female victims and 47 percent of male victims were assaulted by a male (ABS 2003).


  • ABS 2007a. Criminal courts Australia 2005-06. ABS cat. no. 4513.0. Canberra: ABS
  • ABS 2007b. Recorded crime, victims, Australia, 2006. ABS cat. no. 4510.0. Canberra: ABS
  • ABS 2006. Criminal courts Australia 2004-05. ABS cat. no. 4513.0. Canberra: ABS
  • ABS 2005. Criminal courts Australia 2003-04. ABS cat. no. 4513.0. Canberra: ABS
  • ABS 2003. Crime and safety Australia, 2002. ABS cat. no. 4509.0. Canberra: ABS
  • Finch E & Munro VE 2005. Juror stereotypes and blame attribution in rape cases involving intoxicants. British journal of criminology 45: 25-38
  • Fitzgerald J 2006. The attrition of sexual offences from the New South Wales criminal justice system. Crime and justice bulletin no. 92
  • Lievore D 2004. Prosecutorial decisions in adult sexual assault cases: an Australian study. Canberra: Office of the Status of Women
  • Lievore D 2003. Non-reporting and hidden recording of sexual assault: an international literature review. Canberra: Office of the Status of Women
  • NSW Criminal Justice Sexual Offences Taskforce 2005. Responding to sexual assault: the way forward. Sydney: NSW Attorney General's Department
  • NSW Criminal Law Review Division (CLRD) 2007. The law of consent and sexual assault: discussion paper. Sydney: NSW Attorney General's Department
  • Office of the Director of Public Prosecutions & Australian Federal Police (ODPP & AFP) 2005. Responding to sexual assault: the challenge of change. Canberra: Office of the Director of Public Prosecutions
  • Taylor N & Joudo J 2005. The impact of pre-recorded video and closed circuit television testimony by adult sexual assault complainants on jury decision-making: an experimental study. Research and public policy series no. 68. Canberra: Australian Institute of Criminology http://www.aic.gov.au/publications/rpp/68/
  • Taylor N & Mouzos J 2006. Community attitudes to violence against women survey 2006: full technical report. Canberra: Australian Institute of Criminology http://www.aic.gov.au/publications/reports/2006-11-violenceAgainstWomen....

About the Authors

Dr Natalie Taylor is a senior research analyst and Manager of the Justice and Crime Analysis Program at the AIC.