Examining How Prevalent Can Rape Be Defined Criminology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Prevalence can be defined as the proportion of people who have been the victim of a particular offence or group of offences in a fixed period of time (Myhill & Allen, 2002). The true prevalence of crime is important to attempt to establish as it can assist government and other agencies to target specific crimes, develop policies and procedures focusing on crime prevention and control, and serve as a benchmark against which to track trends (Pakes & Pakes, 2009). Equally, understanding the changes in types of crime committed provides valuable information for research purposes, focusing academics on relevant issues.

It is generally accepted across all methods of measuring crime that the number of crimes recorded has increased almost every year since records began. Between 1981 and 1987, crimes recorded by police in England & Wales increased by 41%. The British Crime Survey however, found an increase of only 30% (Mayhew, Elliott & Dowds, 1989) with the difference being mostly accounted for by the increase in people reporting crimes to the police (Blackburn, 2002).

However, what actually counts as a crime can vary depending on the way in which such data is collected. The various means that are used to count crime are described in this essay and their strengths and weaknesses evaluated to examine the extent to which they can be considered to indicate the prevalence of the specific crime of rape. Studies produce varying figures about how common rape is, depending on factors such as the definitions used, whether or not the assaults were reported to the police and what methods are used to obtain the information.

The majority of the essay will focus on the methods used in England and Wales to record the prevalence of rape. There are now numerous different sources of crime data available to examine to understand the prevalence of rape, including the British Crime Survey (BCS), St Mary's database and official government data based on offences recorded by the police. However, as other countries utilise other data collection methods these will also be briefly considered. Consideration of other countries prevalence rates may enable identification of variables that perhaps impact upon higher or lower rates for particular offences.

Definition of rape

Prior to discussion of the various methodologies used to count the prevalence of rape it is first necessary to consider how such a crime is defined. In England and Wales the definition of rape was substantially changed by the Sexual Offences Act 2003, which came into force on 1 May 2004.

Offences committed before 1 May 2004 are prosecuted under the Sexual Offences Act 1956 where the statutory definition of rape was any act of non-consensual intercourse by a man with a person. Intercourse can be vaginal or anal. Offences committed on or after 1 May 2004 are prosecuted under the Sexual Offences Act 2003 which extends the definition of rape to include the penetration by a penis of the vagina, anus or mouth of another person.

It is also important to note that definitions of rape vary both within and across countries. Indeed, in the UK different methods of measuring prevalence use differing definitions of rape making comparability across sources difficult. The definition may impact upon the responses obtained. International research highlights this issue with several community-based studies using the phrase of 'violence against women, rather than 'crime' which has see greater disclosure of sexual offences; perhaps as some victims do not consider rape within some contexts to be a crime (e.g. within a marriage). REFERENCE!! Also, the use of the word 'rape' in questions has been reported to have the effect of decreasing the reporting of forced sex/sex without consent (Schwartz, 1997). The redesign of questions, for example, in the US National Crime Victimisation Study in 1992 resulted in findings four times higher than previous versions (Greenfield, 1997).

Additionally, surveys often do not offer definitions of rape and sexual assault to their respondents (School of Psychology, Module 1: Psychology, the legal system and criminology, 2009) so may not be measuring the same thing as official statistics that have a very proscribed definition. Therefore, any reported differences between the statistics need to be interpreted with some caution (Module 1, 2009).

Official statistics

The most easily accessed statistics for the prevalence of most offences are crimes that are recorded by the police. Historically, the main 'official' compilation of crime figures in the UK was 'Criminal Statistics', published since 1876, which contained the data derived from various court and police records.

Examination of the statistics reveals a progressive increase in the number of rapes reported to the police for over 20 years (HMCPS & HMIC, 2007). Female rape in 2007 / 08 was reported to be 12,630 cases, with 1,150 recorded cases of male rape (Thorpe, Rob & Higgins, 2007). However, 2007 / 2008 figures showed a decrease in both female rape (n=11631) and male rape (n=1008) on the previous year. In the last published figures for 2008 /2009, 12, 165 cases of female rape and 968 cases of male rape were recorded (Home Office, 2009). This was an increase on the previous year of 5% for female rape, and a decrease of 4 % for male rape. Male rape has been gradually decreasing.

On initial examination it would seem that the figures provided above are a good indicator of the prevalence of rape in England and Wales. However there are a number of flaws in the methodology used to produce these statistics that require careful consideration and must be taken into account when using these figures as indicators of prevalence.

Police recorded statistics are based on figures supplied by the police to the Home Office and include crimes which are reported to and recorded by the police. Police recording practice is governed by the National Crime Recording Standard (NCRS). which was introduced in all police forces in April 2002 to ensure greater consistency of crime recording between forces. The standards also emphasised that there should be a victim focus where crimes would be recorded once reported unless there was credible evidence to the contrary. It may have been anticipated that reports of rape would increase following the NCRS since recording and classification of all crimes was now expected to be based on the perception of the victim that an offence occurred. A long recognised flaw in establishing the prevalence of rape is that often there is little evidence so cases are not pursued and complaints are not recorded. With the change of rules in 2002 this flaw in theory should have been addressed and therefore we should expect to see an impact upon the numbers of recorded rape. Simmons and Dodd (2003) suggest that changing this counting rule may have increased recorded offences against the person by 23 per cent in police statistics. However, in reference to sexual offences they reported that the difference would not be so significant and, on in support of this the figures do not alter radically with the introduction of the new rules (Crime in England & Wales 2008/09).

It has also been recently reported in the media that many rapes which have been reported to the police are not included in official crime records. A Freedom of Information request in 2009 by the BBC showed some UK police forces were failing to record more than 40% of rape cases reported to them.

Wide regional variations existed with some forces having high number of cases removed from records indicating that, despite the NCRS, different forces continued to apply different rules for recording rape. For example, in Northumbria, there were 382 reports of rape yet only 210 of these cases were recorded in Home Office figures. In contrast, forces in Humberside, Gloucestershire, and Northamptonshire recorded at least 90% of reported cases for investigation.

In Scotland it appeared that much less stringent criteria were applied to recording of crime with the Northern Constabulary recording every case of reported rape. (BBC, Monday, 21 September 2009) This potentially could lead those examining the data to think that Scotland had particularly high rates of rape offences, when in fact it may be that rapes are more often recorded as a crime than they are by forces in England & Wales.

A further methodological flaw is that if rape is committed in the context of a sexual homicide then the offence of murder will be the one that is recorded as it is only the most serious offence that is counted (Module One, 2009). For something to be recorded as an offence would therefore depend on a number of different variables. As Bottomley & Pease (1986) stated, the officially recorded crime figures reflect the filtering of events through people with their own agendas. The police forces main concern when recording and investigating offences is not the generation of statistics on prevalence (Maguire, 2002) and therefore the accuracy of the recording process is likely to be subject to individual biases.

Crime Surveys

The British Crime Survey (BCS) was first carried out in 1982 and this demonstrated that only a minority of crimes committed were officially recorded as such (Hough & Mayhew, 1983). It is now an annual survey and the sample size has increased over the years to around 40.000 respondents per year (Maguire, 2002).

The BCS surveys a representative sample of people aged 16 and over from households in England and Wales. The survey is carried out by interviewing people in their own homes and asking them about their experiences of crime over the last twelve months. Considerably more crime is recorded in this way than that which is revealed in official statistics (Blackburn, 2002). BCS estimates for 2008/09 are based on face-to-face interviews with 46,286 respondents. The BCS has a high response rate (generally around 76%) and the survey is weighted to adjust for possible non-response bias and to ensure the sample reflects the profile of the general population as afar as is possible.

It could be considered to be a more reliable measure of the extent of crime than the police record statistics as it is not affected by whether the public report crime or by changes to the way in which the police record crime. However, the BCS figures cannot be directly compared with Home Office statistics as they both approach the recording of crimes in different ways. For example, recorded crime includes crime experienced by under 16's and the BCS does not. However, since January 2009 children aged 10 to 15 have been included in the survey and these results will be available next year. Also, sampling errors occur as generally only one member of the household is interviewed which may then miss significant high risk groups and estimates of rape are believed to be particularly prone to error (Block & Block, 1984; Mayhew, Elliot & Dowds, 1989).

The rationale for carrying out such a survey is that those people who for various reasons have not reported crimes to the police may report in different contexts and as such will provide a fuller picture of crime prevalence in the UK. Hence many commentators report that BCS data is more valid in giving a picture of crime occurrence in the UK. (REFERENCE). However, the survey method requires respondents to understand the questions put to them and as mentioned previously, there are a variety of definitions of rape and many may not think such an offence can be committed within a relationship.

The accuracy of the BCS may be considered to have improved since 1992 when the Postcode Address File was used to select respondents. Previous to this the electoral register was the sampling source and as such a proportion of the UK population was automatically excluded including many high risk groups such as the young people and ethnic minorities (Mayhew, Maung & Mirrlees-Black, 1993). It also does not include people in psychiatric institutions or the homeless which are both groups that are vulnerable to being victims of violent crime (Pakes & Pakes, 2009).

It is incorrect to assume that rape is more likely to reported in the British Crime Survey than in official statistics. For example in the 1982 and 1984 BCS reports only one report of rape was included, both recorded as attempts. Maguire (2002) notes that up to and including 2002 sexual offences were still reported very infrequently to BCS data collectors making any estimate of prevalence impossible. It was posited that the low rates of reporting may be associated with the need to verbally respond to an interviewer. As such the BCS have now included a computerised self-completion module, which was designed to provide an accurate estimate of the extent and nature of sexual victimisation (Myhill and Allen, 2002) and subsequently the reporting of rape offences has increased.

The 2008/2009 survey revealed a 0.2% prevalence rate for rape which is the same rate as reported for the previous year. There was a slightly larger sample size in 2007/9 (46,773 compared to 46,286 in 2008/9 although this is not statistically significant). The BCS revealed that women are most likely to be raped by men they know, and that a considerable proportion had experienced repeat incidents by the same perpetrator.

An important research study based on the findings of the BCS (Myhill & Allen, 2002) found that only 60% of rape victims classified what had happened to them as rape. Additionally, only approximately 1 in 5 rapes were reported to the police - demonstrating that the police figures substantially under-estimate the prevalence of rape. There was also reluctance amongst victims to report perpetrators who they had a relationship with as there seemed to be an assumption that rape was not possible within a relationship (Myhill & Allen, 2002) or would not be treated as ac rime if reported. Perhaps it is these assumptions that need to be challenged in order that more rapes are reported and we have a clearer picture of the prevalence of rape in England & Wales. - reword

Supporting the above findings, in the 2003/04 BCS, less than half (43%) of women who had experienced an assault that met the legal definition of rape defined it as such, and this was even lower where the perpetrator was a current or ex-partner (31%). However, where there was also a physical injury, the proportion of respondents defining their assualt as 'rape' increased markedly to 62 per cent (Walby and Allen, 2004).

It was recognised in recent years that BCS and Home Office figures contained markedly different results in their reporting of crime prevalence. In recognition of this, the Home Office now combines the two sources of data in an attempt to give a more accurate picture of the level of crime in England and Wales.

Further sources of information

Additional sources of information on the prevalence of particular crimes are local surveys. These are often beneficial in examining a specific type of crime as they tend to have a more detailed focus (Maguire, 2002) One study designed solely to provide information on the extent of unreported rape was carried out by Painter (1991) and involved 1,007 women in 11 cities. This study found that one in four women had experienced rape/attempted rape in their lifetime, with current and ex-partners the most common perpetrators, and the vast majority (91%) not disclosing the crime at the time. The survey included a significant proportion of women who declared themselves as victims within their marital relationships (14%). Again though, the results from this study must be interpreted with caution. The sample size is small and it is unlikely that it is a representative sample of the population as it covered 11 specific cities. Also, the study was focused on a specific question and does not include male or child victims of rape so only provides a basic understanding of prevalence of rape within a narrow definition. The study does however highlight that many women are victims of rape within their relationships which may well impede the likelihood of the offences being reported to the police or disclosed to an interviewer who is present in the marital home.

A further source of information local to the author of this essay is St. Mary's Sexual Assault Referral Centre which is a collaborative venture between Central Manchester and Manchester Children's University Hospitals NHS Trust, Greater Manchester Police, and Greater Manchester Police Authority. St Mary's holds a database on recent rapes in the UK comprising almost 8,000 cases. Analysis of the St Mary's database (covering the years 1987-2002) showed that there were increasing numbers of young victims of rape who were aged under 20 (Kelly, Lovett & Regan, 2005). As the BCS does not question those under the age of 16 it is likely therefore that they are missing a significant proportion of victims. St Mary's database also supports the BCS finding that assaults are most likely to be committed by known men (Kelly, Lovett & Regan, 2005) which may also impact upon how likely an offence is to be reported to the police or BCS interviewers.

Examination of international examples of rape prevalence studies which, through focusing specifically on violence against women, (rather than using the word 'crime'), seem to have enabled greater disclosure of such offences revealing figures similar to those found by Painter (1991)

One of the most frequently cited research studies is that conducted by Statistics Canada in 1992, which telephone interviewed a random sample of 12,300 women. The BCS findings on known men being the most common perpetrators and repeat assaults were supported, but a far higher prevalence rate with over one in three reporting a sexual assault than any of the UK data gathering methodologies. This is comparable to Painter's findings. The Canadian study also found a very low reporting rate to the police of 6% for rape compared to 25% for domestic violence.

In America, the Uniform Crime Reports (UCR) and the National Crime Victimization Survey (NCVS) are the two primary sources of official rape statistics. In 2005, the UCR figures showed that 93,934 completed or attempted rapes were reported, or 31.7 reports for every 100,000 adult women (Federal Bureau of Investigation, 2006). Also in 2005, the NCVS reported that 64,080 completed rapes and an additional 51,500 attempted rapes occurred or 0.5 rapes/attempted rapes per 1,000 adult females (BJS, 2006).

Using data from the National Crime Survey (NCS) from 1973-1991 and the National Crime Victimization Survey (NCVS) from 1992-2000, Baumer (2004) found that overall reporting rates continued to increase during the 1990s, and that differences in rates of reporting between stranger and non-stranger rapes diminished. Some of these changes were attributed to campaigns that focused attention on rapes by known offenders. It is important to note also that in 1992 the NCVS changed their wording and asked women if they had experienced unwanted sexual activity. This resulted in an increase in reporting, again highlighting the importance of the wording used in rape prevalence research. It is not possible however to make a direct comparison between prevalence rates in different countries as practices of recording and definitions of rape can vary significantly (MacDonald, 2002).

However, as in the UK, a majority of rape victims still do not report their attacks to police. The Bureau of Justice Statistics (BJS) reports that only 36% of rapes between 1992 and 2000 were reported to the police (Rennison, 2002). Even lower figures are indicated in the Netherlands where studies indicate that only 6% of the victims of sexual crimes reported to the police (Wittebrood & Junger, 2002.) Further study is needed to understand what it is that specifically impacts on reporting behaviour and to then identify whether there are any practices that would facilitate higher reporting rates.

It has been suggested that the police now respond more sensitively to victims of rape (Crime in England & Wales, 2009) and this may have an impact on the willingness of victims to report a crime if they share this perception. This will then potentially demonstrate a corresponding increase in the number of crimes recorded. However, there are potentially still incidents where victims of rape are blamed in part for being intoxicated or taking drugs for example. If these factors are present in the offence this may also impact upon the willingness of a victim to making a report fearing negative judgement from authorities (Skogan, 1994).

As rape is not an offence that is 'targeted' as part of police crime initiatives (although preventative campaigns may appear) it may be that to increase the reliability of prevalence reports we need to focus on providing information to victims about the crime of rape (including the information that rape can occur in the context of a relationship) the process of reporting a crime and an emphasis on the increasingly sympathetic response form the police force and other voluntary agencies. This approach had a demonstrable effect in the US on reported rape figures.


Block & Block (1984) note that there is no real number of crimes which exists independently of the various processes by which data is collected. Al methods of data collection contain some sources of error (Blackburn, 2002).

Across different methods rape remains one of the most under-reported crimes and is recognised as being significantly under represented in official crime statistics (Walby and Allen, 2004).. Even in confidential victim surveys it may be the case that victims of rape are unwilling to record their experience as there is an element of shame for many.

Also, it is possible that those carrying out the interview conduct them in the presence of others where disclosure of theft and burglary may occur but more 'personal crimes' may not be so easily revealed (pg. 42 module handbook). According to the most recent statistics however the majority of respondents to the BCS self-completion module completed the nodule on their own. However, those who completed in the presence of others were mainly young adults a group who are increasingly being recognised as victims of rape (Lovett etc)

There are no ideal ways of measuring the prevalence of rape but this does not imply that we should not continue to record its occurrence in a variety of ways. It is particularly important in a widely underreported crime such as rape that society continues to acknowledge its prevalence and the effects that such crimes have on victims. Just because there is no perfect measure does not mean we should stop recording it. Further attempts are being developed to access unrecorded crimes by the examination, for example of Accident and Emergency department statistics. (Home Office, 2000). This has been demonstrated to be a worthwhile source of nformation about the prevalence of violent assaults (Silvarajasingam, Shephard, Matthews, Jones, 2003). However, victims of rape may not always be willing to disclose to A & E staff the circumstances which they received the injuries particualryl if the rape was committed by a known offender.

As Maguire (2002) notes it is essential to approach the question of crime within a critical framework as there is much variety between methods of data collection and recording as well as differences in definitions of particular types of crime.

As with all prevalence research, inconsistent findings are attributed to methodological differences with respect to: the sample; the number and content of questions asked; the format (questionnaire, telephone or face-to-face interview); and the definition of rape/sexual assault used by the researchers (see Schwartz, 1997, for a more detailed discussion).

Home Office data on reported rape cases in England and Wales show a continuing increase in reporting of rape to the police over the past two decades/ However the number of convictions has remained relatively static leading to what is known as the 'justice gap'. Figures released in 2009 revealed that only 6.5% of reported rapes resulted in a conviction. (Guardian, 2009) Such figures, which have been widely reported in the media, may contribute to a reluctant or report rape as there is the perception that it is unlikely to result in a conviction.

Increasingly the validity of official crime statistics has been questioned and other sources have been looked at to understand the prevalence of certain types of crime (Blackburn, 2002).