(Disponible en français : L’Enquête sociale générale, un instrument pour élucider le « chiffre noir » de la criminalité – mise à jour)
For many years, statistics provided by police were considered the most reliable way to measure crime rates, because they were generally believed to accurately reflect all crime. However, in the 1960s and 1970s, research showed that a large portion of crime – crime that went unreported – escaped the attention of the police.
To address this shortcoming, Statistics Canada has been compiling data on victimization since 1988 through the General Social Survey (GSS). Information is collected from Canadians by phone to determine, among other things, if the respondents believe they have been victims of crime in the previous 12 months and if they reported it to the police. This survey complements a Statistics Canada survey based on police-reported data established in 1962.
The advantages and disadvantages of the GSS
The GSS measures both reported and unreported crime. It also collects important data on why people choose to report a crime to the police, what their victimization experience was, what their risk factors are (age, lifestyle, etc.), how they perceive the criminal justice system and how much they fear crime. It is also worth noting that the reported incidents identified in the GSS are not necessarily included in police-reported data, as the police include in their statistics only those crimes that they have substantiated.
GSS data are not infallible: they are subject to sampling error and based on the assumptions that the respondents correctly identified when the incident in question occurred, recognized the incident as a crime, agreed to talk about it, and remembered and described the incident accurately.
In general, victim surveys offer a more accurate picture of less-serious crimes involving a victim, while police statistics are considered accurate for more serious crimes, for which the reporting rates are fairly high.
Statistics Canada recognizes that the victimization survey and police-reported data are complementary when assessing crime and its impact on society. The organization continues to undertake the survey on a regular basis, once every five years. Other countries such as the United States and England and Wales conduct such surveys annually.
Police-reported data and the GSS
Given that the survey based on police-reported data shares neither the objectives, methodology nor population of the GSS (see Table 1), the two surveys will not necessarily produce similar data.
Table 1 – Statistics Canada Crime Surveys
Survey based on police-reported data
Survey on victimization (GSS)
|Frequency||Annually||Every 5 years|
|Survey frame||All offences in the Criminal Code and other federal and provincial legislation.||Random sampling of the population over the age of 15 with access to a landline (note: as of 2014, cellphones are included) and not residing in an institution (e.g., a hospital or prison).|
|Exclusion||When several crimes occur during a single incident, only the most severe crime is recorded, unless they are violent crimes involving a victim. In these cases, one crime per victim is recorded. This means that less-serious crimes are underestimated.||Crimes committed against a company or an institution and crimes between consenting parties (e.g., drug deals, the purchase of sexual services) are not recorded. Crimes involving victims under the age of 15, without access to a landline (before 2014), or who reside in an institution are not recorded.|
|Crimes reported||Approximately 200 crimes||Eight types of crimes (see Table 2)|
Crime according to the victimization survey
According to the 2014 GSS, victimization rates fell between 2004 and 2014 for almost all crimes measured, with the exception of sexual assault, which remained stable (see Table 2). Unlike in previous surveys, women recorded a higher rate of violent victimization than men (85 incidents per 1,000 women compared to 67 incidents per 1,000 men). Moreover, Aboriginal women were victims of sexual assault more often (115 incidents per 1,000 women) than non-Aboriginal women (35 incidents per 1,000 women).
The proportion of crime reported to the police fell just slightly during this period, dropping from 34% to 31%. In 2014, only one sexual assault in 20 was brought to the attention of the police, while 93% of cases of child victimization were never reported.
The GSS showed that the rates of reporting to the police varied according to such factors as the type of crime, the victim’s relationship with the offender and the value of damaged or stolen goods. One of the most common explanations for not reporting incidents to the police was that the incident was not important enough (78%). A number of violent crime victims felt that it was a personal matter (63%) and did not want to involve the offender in the justice system (27%). For incidents of household victimization, many victims believed that the police would not consider the incident important enough (66%), that they could not do anything about it (65%) or that there was a lack of evidence (61%).
Table 2 – Victimization incidents reported by Canadians, by type of offence
Number in thousands
Number in thousands
|Total violent victimization||2751||106||2245||76|
|Total household victimization||3206||248||2029||143|
|Breaking and entering||505||39||441||31|
|Theft of motor vehicles / motor vehicle parts||571||44||261||18|
|Theft of household property||1136||88||766||54|
|Theft of personal property||2408||93||2154||73|
¹ Rates of violent victimization and theft of personal property are calculated per 1,000 population aged 15 years and older. Rates of household victimization are calculated per 1,000 households. Source: General Social Survey on Victimization (4504).
Despite significant differences in methodology between the GSS on victimization and the survey based on police-reported data, both show similar trends over the 10-year period under review, specifically a decline of both violent crime (down by 28% and 26% respectively) and property crime (down by 42% and 40%) from 2004 to 2014.
Coleman, Clive, and Jenny Moynihan. Understanding Crime Data: Haunted by the Dark Figure, Open University Press, 1996.
John van Kesteren, Jan van Dijk & Pat Mayhew. “The International Crime Victims Surveys: A retrospective”. International Review of Victimology, 20(1): 49-69, January 2014
Jan van Dijk. Closing the doors: Highlights of the International Crime Victims Survey 1987-2012. Tilburg, 2012
Jan van Dijk, et al. Final report on the study on crime victimisation. Tillburg University. August 2010
van Dijk, Jan, et al. Criminal Victimisation in International Perspective, United Nations Office on Drugs and Crime, 2007.
Author: Lyne Casavant, Library of Parliament
Categories: Law, Justice and Rights