Reading the 2022 Corrections and Conditional Release Statistical Overview (CCRSO)
A data literacy companion

While the format of the HTML and PDF versions of the CCRSO Data Literacy Companion may differ in some areas in order to meet different access needs, the content is the same.

Public Safety Canada is committed to creating barrier-free documents. To receive this document in an alternate format, please contact the Research Division.


This resource is designed to provide some context and clarity around select key data literacy topics and terminology relevant to the 2022 Corrections and Conditional Release Statistical Overview (CCRSO). It is by no means exhaustive in its coverage, but is intended as a guide to facilitate an accurate and more critical interpretation of some of the data presented in the CCRSO.

Data Literacy Concepts

Victimization versus Incident versus Arrest versus Charge versus Conviction


Refers to the self-assessment by a person under Canadian legal jurisdiction that they have experienced an incident that should be considered to be a crime. Not all criminal victimizations are reported to police. In the CCRSO, information on self-assessed victimization is captured in the General Social Survey (GSS).


Refers to the total number of incidents reported to police, or discovered by police, which police classify as being potential crimes. Not all criminal incidents will result in an arrest because of reasons such as a lack of evidence or due to the incident not being a crime. As such, the total number of incidents reported to police will be greater than the number of arrests, charges, or convictions. In the CCRSO, information on police-reported crime is captured in the Uniform Crime Reporting (UCR) Survey.


Occurs when police detain an individual based on evidence of potential involvement in criminal activity. Arrest data show the number of individuals who are apprehended by police for various types of crime during a given time period. Not all arrests lead to charges.


Occurs when there is a formal accusation that an individual has committed a crime, and is laid by the police or counsel, depending on jurisdiction. Only a subset of those individuals arrested for a crime will formally be charged, and only a subset of charges lead to convictions.


Occurs when the court finds a person who has been charged with a crime guilty of a criminal offence based on evidence provided in court or through a guilty plea.


Not all victimizations are reported, not all reported criminal incidents are crimes or can be solved, not all arrests lead to charges, and not all charges result in convictions! Understanding these differences and examining data related to each of these concepts can lead to a better understanding of the resource needs at each level of criminal justice processing.

Types of Conditional ReleaseFootnote 1 from Federal Custody – Day Parole versus Full Parole versus Statutory ReleaseFootnote 2

Day Parole

A form of conditional release, day parole allows an offender to participate in daytime community-based activities in preparation for full parole or statutory release. The individual must generally return to a community-based residential facility, halfway house or other locationFootnote 3 each night, or at another specified interval. In addition to the standard conditions of day parole, other special conditions may also be imposed by the Parole Board of Canada.

Generally, an individual is eligible for day parole 6 months before the full parole eligibility date (PED) or 6 months into their sentence – whichever is later. Eligibility for offenders serving life sentences is 3 years before their PED.

In 2021-22, 3,302 federal offenders were granted day parole (75.8% of eligible applicants who had reached day parole eligibility date) and 1,072 federal offenders were granted full parole (30.4% of those who had reached full parole eligibility date).

Full Parole

A form of conditional release, full parole allows an offender to serve a portion of their sentence under community supervision, generally staying at a private residence, with certain conditions imposed. Full parole typically follows the successful completion of day parole and facilitates the process of community reintegration.

Generally, an individual is eligible for full parole after completing one-third of their sentence or after serving 7 years (whichever comes first). In life sentences, the court sets a parole eligibility date at the time of sentencing. Parole eligibility for first-degree murder is automatically 25 years and may range between 10-25 years for second-degree murder.

Over the past 10 years (2012-13 to 2021-22), successful completion of both day parole and full parole has been approximately 90% (i.e., completion without a return to custody for breach of conditions or a new offence).

Over the past 5 years (2017-18 to 2021-22), the percentages of violent recidivism for federal offenders on day parole and full parole are particularly low, averaging 0.2% and 0.6%, respectively (during the community supervision period).

Statutory Release

Unlike parole, which is granted discretionarily based on an assessment of the offender's level of risk, statutory release (SR) is mandated by law after an offender has served two-thirds of their sentence, except for those serving life or indeterminate sentences or those who have been deemed to have met the criteria to be detained past their statutory release date.Footnote 4 Offenders on SR are subject to standard conditions of release, and the Parole Board of Canada may impose special conditions on the SR.

Offenders on SR may be returned to custody if they violate a condition of their release (as with day or full parole), or if their level of risk increases.

In 2021-22, 68.4% of all releases from federal institutions were statutory releases (total of 4,383 statutory releases).

The percentage of revocation for a violent offence among those statutory release has averaged 1.4% over the past 5 years (2017-18 to 2021-22).


The goal of conditional release is to facilitate community reintegration in a gradual, structured, and supportive manner and to optimize an individual's chances of remaining crime-free once their sentence is complete.

Frequency (count or absolute number) versus Rate


Tells us about the absolute number of crimes that take place in a given period of time. This count helps researchers and policy-makers understand the capacity of the systems (e.g., police, courts, correctional facilities) needed to respond to the crimes committed in a specific population. Although understanding frequencies is important, these can sometimes be misleading in the context of making comparisons.

In the CCRSO, figures or tables denoted as “Number of…” depict frequency data (i.e., absolute numbers).


Tells us about the intensity of a crime problem. Rates are created by calculating a ratio of the crime count against the size of a specific population (typically per 100,000 population) and as such, rates can sometimes be more helpful for comparison purposes, particularly as related to risk. Indeed, rates describe the risk that a member of that population group might be victimized by a crime or commit a crime.

In the CCRSO, figures or tables denoted as “Rate per…” depict population adjusted rate-based data.

Frequency (number) of crimes in Canada in 2010 vs. 2022

In 2010, the number of crimes in Canada was 2,379,130. In 2022, the number of crimes in Canada was 2,411,535. This represents 32,405 more incidents of crime in 2022, compared to 2010.Footnote 5

Rate (risk) of crimes in Canada in 2010 vs. 2022

In 2010, the rate of crimes in Canada was 6,996 per 100,000 population. In 2022, the rate of crimes in Canada was 6,195 per 100,000 population. Although there were 32,405 more incidents of crime in 2022 compared to 2010, Canadians were roughly 11.5% less likely to experience crime in 2022 compared to 2010.Footnote 6


Number and rate data answer different policy questions and so inform policy and program responses to crime in different ways.

Percentage Change – Interpretation is Dependent on Event Frequency

Percentage Change

Measures the relative difference between two values in terms of percentage. It is a statistical method commonly used to quantify the extent of change between one value and another. It can be applied to any trend or event measured over time. The formula for percentage change is:

Percentage Change =  Value at Time 2 - Value at Time 1 Value at Time 1  x 100

Here is the formula applied to figure A3 of the CCRSO:

Figure A3. Number of police-reported violent victimizations from 2017 to 2021

Number of police-reported non-sexual assaults in 2017 (225,350) vs. 2021 (266,020)

Percentage Change =  (266,020 – 225,350) 225,350  =  40,670 225,350  x 100 = 18.0%


Number of police-reported sexual assaults in 2017 (23,905) vs. 2021 (33,091)

Percentage Change =  (33,091 – 23,905) 23,905  =  9,186 23,905  x 100 = 38.4%

The interpretation of percentage change depends on context, and specifically, on the frequency of a particular event. Although the percentage increase of police-reported non-sexual assaults is 20.4% less than the percentage increase of police-reported sexual assaults over the 5 years, the increase of the absolute number of non-sexual assaults is 31,484 greater than sexual assaults:

Because percentage change is heavily dependent on the absolute frequency of the respective event type (e.g., non-sexual assault vs. sexual assault), a comparison that is restricted to percentage increase may artificially bias interpretation of the actual magnitude of change – particularly in cases where there is a large discrepancy between the frequency of the respective events.


Be cautious when comparing differences in percentage change! With high-frequency incidents, even a substantial increase in absolute numbers may only lead to a slight percentage increase; conversely, a lower-frequency incident can lead to a more significant percentage increase. In other words, a given increase of X in the number of events will lead to a higher percentage increase in a low-frequency event than in a high-frequency event.

Trend Lines and Time Anchor Points

A trend line is a visual representation of a dataset's underlying pattern or trend over time.

A time anchor is a specific point that acts as a reference or starting point for a trend analysis. It can help assess changes or patterns in a dataset relative to that anchor point. The choice of the anchor point depends on the goals of the analysis and the characteristics of the data.

The time anchor points that are chosen to be included in a trend line affects how the data is understood. Take for example, the following crime trend data interpretations, which use different time anchor points.Footnote 7

Between 2000 and 2021, the violent crime rate decreased (1,494 vs. 1,323 = 11.4% decrease).

Between 2014 and 2021, the violent crime rate increased (1,044 vs. 1,323 = 26.7% increase).

Between 2008 and 2021, the violent crime rate remained relatively stable (1,334 vs. 1,323 = 0.8% decrease – not a statistically significant change).


All three statements above are accurate. However, they present very different conclusions about violent crime trends. Remember to take note of the time anchor points for the data when interpreting a trend.


We welcome your feedback on this product, which is a reflection of data contributions and insights of our valued partners including Correctional Service of Canada, Parole Board of Canada, the Office of the Correctional Investigator, and the Canadian Centre for Justice and Community Safety Statistics (Statistics Canada). Please reach out to the Research Division, Crime Prevention Branch, Public Safety Canada with any questions or comments that you might have, including suggestions for new CCRSO data literacy products!

Date modified: