Public Release: 

Crime, British Muslims and their relationships with the police

University of Cambridge

Muslim communities may not be as victimised by violent crime, or as dissatisfied with the police as is widely suggested and believed, according to new research by a Cambridge academic.

An examination of statistics taken from the Crime Survey of England and Wales between 2006 and 2010 reveals a surprising counter-narrative to commonly-held perceptions of British Muslim communities and their relationships to crime victimization and the criminal justice system.

Analysis of crime data generated by nearly 5,000 Muslims reveals few differences between Muslims and non-Muslims in relation to a range of violent personal crime including assaults, wounding and threats - the types of crime that scholarly literature, media reports and anecdotal evidence all suggest have disproportionately affected Muslim communities.

Likewise, statistical analysis reveals few statistically significant differences between Muslim, Hindu, Jewish and Sikh respondents in respect of many personal crime types included within the Crime Survey.

Research carried out by Julian Hargreaves from Cambridge's Centre of Islamic Studies, and published today by the British Journal of Criminology, also shows surprisingly high levels of positive attitudes towards the police. Among Muslims surveyed by the Crime Survey of England and Wales, 61.2pc of respondents rated their local police as either good or excellent compared to 53pc of non-Muslims.

Despite widespread condemnation of police counter-terrorism measures such as stop and search, data from the Crime Survey reveal that a sizeable majority of Muslim respondents not only reported positive attitudes towards a range of subjects such as police fairness, reliability and relevance to the community but are in many cases more likely to do so than non-Muslim respondents.

"Such findings, derived from a large nationally-representative dataset, clearly challenge the dominant narrative within the criminological literature which describes all or most Muslim people as being at greater risk of violent crime, as well as having a deep-rooted sense of police dissatisfaction," said Hargreaves.

"Much of the discourse in this area is highly politicised and rhetorical in character, rarely rooted in statistical evidence and seldom substantiated by empirical findings. The findings suggest a growing need to move beyond misleading and potentially damaging generalisations which seek to cast British Muslim communities only as the victims of violent crime and police discrimination."

Research undertaken by Hargreaves also examined the relationship between socio-economic disadvantage and household crime such as burglary and car theft. Here, the analysis revealed a strong relationship between household crime victimization and the socio-economic disadvantage suffered by many British Muslim communities - topics rarely, if ever, discussed by criminologists and journalists.

The research incorporated data from a 'Multiple Deprivation Index', a ten-tier classification system designed to map socio-economic deprivation in the UK.

While there were no discernible differences between Muslim and Christian households in each tier (or decile) there were marked differences in the numbers of Muslim and Christian households contained within areas of disadvantage. Two thirds of Muslim households (66.2pc) were distributed among deciles containing the most deprivation and the highest levels of reported household crime.

While Hargreaves' work provides a fascinating counter-balance to traditional victimization narratives concerning Muslim communities, he is at pains to point out the need to use large-scale and anecdotal evidence together in order to complete as broad a view as possible of Muslim life in 21st century Britain.

"Research in this field is dominated by descriptions of British Muslim communities which appear to eschew the use of large, nationally-representative data samples, placing emphasis instead on interviews and focus groups and on describing the nature of anti-Muslim hostility rather than quantifying the scope and extent of any problems.

"Descriptions of the lived experiences of Muslim victims of crime have undoubtedly provided an invaluable contribution but it is clear that the Crime Survey offers crime statistics at odds with commonly-propagated views - and a greater range of attitudes towards the police than might be assumed from an uncritical reading of the literature.

"Current debates around British Muslim communities, crime and the police - and especially those around the scale of current problems - unfortunately rest on rather inconclusive statistical evidence. Much more research is needed to complete the criminological picture in this area, including large-scale data concerning damage to property such mosques and small businesses. There is also a need for research to focus more often on 'everyday' hate crime, such as verbal abuse from strangers, which often goes unreported to both the police and the Crime Survey."


Members of the press, please email Stuart Roberts, Deputy Head of Research Communications: or call 01223 764982 for further details and a copy of the paper.

Notes for Editors:

Statistics taken from the Crime Survey of England and Wales between 2006 and 2010 and used in the paper: Half a story? Missing Perspectives in the Criminological Accounts of British Muslim Communities, Crime and the Criminal Justice System

Data summary


No statistically significant differences found between Muslim and non-Muslim respondents in respect of:

  • All personal crime (6.3% and 6.2% respectively)
  • All violent crime (3.3% and 3.4% respectively)
  • All assault crime (2.3% and 2.8% respectively)
  • Common assault (1.8% and 2.1% respectively)
  • Theft (1.6% and 1.2% respectively)
  • Robbery (0.8% and 0.5% respectively)
  • Wounding (0.6% and 0.8% respectively)
  • Threats (2.5% and 2.3% respectively)
  • In all above cases the differences were due to sample error (p>0.05) not real differences between the groups

Statistically significant differences found between Muslim and non-Muslim respondents in respect of:

  • Mugging (1.1% and 2.1% respectively)
  • But no such differences found between Muslim, Hindu, Jewish and Sikh (all groups shared a broadly similar likelihood of reporting being mugged to the Crime Survey - among minority religion groups Muslims are not disproportionately victimised).


Statistically significant differences found between Muslim and non-Muslim respondents in respect of:

  • All household crime victimisation (20.8% and 17.4% respectively)
  • Burglary victimisation (4.0% and 2.4% respectively)
  • Car crime victimisation (12.1% and 9.7% respectively)

The Multiple Deprivation Index (MDI)

  • The MDI collates information concerning using factors such as income, employment, housing and education.
  • Ten tiers (or deciles) map socio-economic disadvantage across England and Wales.
  • 40% of household crime reported by households identified as being in the lowest three tiers (i.e. the most deprived 30% of England and Wales).
  • Together these three deciles were found to have 40% more household crime than if such crime were to be distributed evenly across all ten.
  • Two thirds of Muslim households (66%) identified by survey data as being in these lowest three deciles.
  • The increased risks of household crime faced by Muslim communities appeared to be related to socio-economic disadvantage.


Muslim respondents more likely than non-Muslim respondents to report positive attitudes ('excellent' or 'good') towards the statements:

  • How good a job are the police in this area doing? (61.2% and 53% respectively)
  • This difference was found to be statistically significant

Muslim respondents more likely than non-Muslim respondents to report positive attitudes ('strongly agree' or 'tend to agree') towards the following statements:

  • The police in this area can be relied on to be there when you need (63.4% and 47.7% respectively)
  • The police in this area treat everyone fairly regardless of who they are (68.8% and 63.7% respectively)
  • The police in this area can be relied to deal with minor crime (56.2% and 44% respectively)
  • The police in this area understand the issues that affect this community (65% and 63.7% respectively)
  • The police in this area are dealing with the things that matter to this community (59.7% and 52% respectively)
  • Taking everything into account I have confidence in the police in this area (71.3% and 66.1% respectively)
  • All differences were found to be statistically significant


1. In the findings reported here an observation that is said to be 'statistically significant' is one where the probability of observing such a difference was less than 5% (assuming that there are no such differences between the two groups). A statistically significant difference may be thought of as a real difference in characteristic between the specified groups (rather than one appearing at random, or one due to sampling error).

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.