Thursday, April 28, 2016

Inmate Ethnicity Data


I have a researcher seeking historical (and current data) on federal prisons, with particular attention on inmate profiles, precisely...inmate ethnicity.

Annual reports of the OCI, and Commissioner of Penitentiaries (and its previous incarnations) have provided clues that this data is out there.


The Canadian Centre for Justice Statistics (CCJS) does not collect data by ethnicity, with the sole exception that some of our surveys have an Aboriginal identity indicator. For the purposes of corrections data, this is available in the following tables;

The annual Corrections and Conditional Release Statistical Overview from Public Safety Canada provides a national level snapshot. See Figure C9 in the 2015 annual report (p. 49)with the following self-reported race groupings for offenders: 
  • Aboriginal,
  • Asian, 
  • Black, 
  • Caucasian, 
  • Hispanic, 
  • Other/Unknown
Some provincial government departments may also release similar statistics. For example, the Province of BC produces a series of Justice Dashboards, two of which cover corrections: Adult Custody and Community CorrectionsThese dashboards include data on ethnicity in the following categories: 
  • Aboriginal (broken down further into Aboriginal, Native, and Metis), 
  • Asian, 
  • Black, 
  • Caucasian, 
  • East Indian, 
  • Hispanic, 
  • Other, and 
  • Unknown 
The data that feeds into these two dashboards is also available through the DataBC website. I don't know if similar sources are available in other provinces.

There has been quite a lot written about the difficulty in accessing data on ethnicity in the Canadian criminal justice system. Some notable works include:

1. Collecting Data on Aboriginal People in the Criminal Justice System: Methods and Challenges

AuthorRebecca Kong
AuthorKaren Beattie
PlaceOttawa, ON
DateMay 2005
ExtraISBN: 0-662-40132-8
InstitutionStatistics Canada, Canadian Centre for Justice Statistics
AbstractThis objective of this report is to present the status of national data on Aboriginal people who come into contact with the criminal justice system as offenders and victims. The report examines the current and potential collection of an individual's Aboriginal identity through various justice-related surveys at Statistics Canada, the challenges within these surveys to collect these data and provides some insight into the quality of these data. The data and sources are examined within the context of information needs for the justice and social policy sectors, and in relation to the preferred method of measuring Aboriginal Identity at Statistics Canada. Data sources examined include the Incident-based Uniform Crime Reporting Survey, the Homicide Survey, the Integrated Criminal Courts Survey, the Adult Corrections Survey, the Youth Custody and Community Services Survey, the Youth Alternative Measures Survey, the Transition Home Survey, the Victim Services Survey and the General Social Survey on Victimization. Finally, the report briefly describes efforts by other countries to improve justice-related information on their indigenous populations.
Report Number85-564-XIE

2. Race-based criminal justice data in Canada: Suggestions for moving forward 

AuthorAkwasi Owusu-Bempah
DateMarch 2011
InstitutionPublic Safety Canada
Note:Copy obtained under Access to Information request (IA-2013-00078) for previously-released documents held by Statistics Canada (CCJS)​. Originally listed onDecember 2012 Completed ATI Requests list (file # A-2012-00097). 
Please contact me if you want a copy. I haven't attached it here because the 29-page report is a huge 86 MB PDF because ATI documents are released as images.
Excerpt from page 2: 

"Information on individual alleged offenders and victims processed by the criminal justice system, including descriptions of Aboriginal status and racial background, is routinely collected from the records of  criminal justice agencies. However, because this information is collected to meet the disparate operational needs of the agencies involved, the data often lacks the consistency necessary for comparative purposes.

Police services, for example, have the ability to collect Aboriginal and racial data, and many are, in fact, currently recording such data for intelligence purposes and/or when it is relevant to a criminal investigation. The internal data management systems utilized by police services collect detailed information on the racial background of accused persons and the victims involved in crime incidents. While some police forces report Aboriginal information to the CCJS, the majority of forces refuse to do so. The Integrated Criminal Court Survey and the Integrated Correctional Survey are also valuable sources of information on individuals processed through the criminal justice system. Racial information is not systematically collected as part of the Integrated Criminal Court Survey, however, this may be included in th survey where deemed important by stakeholders. The Integrated Correctional Survey currently collects self-reported racial information of individuals entering the correctional system. Furthermore, the potential exists to link police-reported and court data to corrections records, which would allow for a more extensive examination of criminal justice and social policy issues.

The vast majority of data on the racial background of individuals processed through the Canadian criminal justice system is used for internal purposes and is not made publicly available in any systematic way. The  most readily available criminal justice data that includes information on race is the Corrections and Conditional Release Statistical Overview, produced by Public Safety Canada; this document includes a "one day snapshot" of the race of all offenders under federal supervision."

​​3. Whitewashing Criminal Justice in Canada: Preventing Research through Data Suppression


Journal Article
AuthorPaul Millar
AuthorAkwasi Owusu-Bempah
PublicationCanadian Journal of Law & Society / La Revue Canadienne Droit et Société
DateDecember 2011
Accessed2016-04-27, 8:46:50 PM
AbstractRace and racism have long played an important role in Canadian law and continue to do so. However, conducting research on race and criminal justice in Canada is difficult given the lack of readily available data that include information about race. We show that data on the race of victims and accused persons are being suppressed by police organizations in Canada and argue that suppression of race prevents quantitative anti-racism research while not preventing the use of these data by the police for racial profiling. We also argue that when powerful institutions, such as the police, have knowledge that they keep secret or refuse to discover, it serves the interests of those institutions at the expense of the public. Fears that reporting of racial data will result in racial profiling or the stigmatization of racialized communities are not assuaged by the repression of this information. Stigmatization may still occur, and racial profiling can continue to happen, but without public knowledge. Quantitative anti-racist research requires consistent, institutionalized reporting of race data through all aspects of Canadian justice. We outline what data are available, what data are needed, and where consistency is lacking. It is argued that institutional preferences for white-washed data, with race and ethnicity removed, should be subrogated to transparency.

Monday, April 25, 2016

Geographic reference files

Does anyone know if the following geographic reference products exist?
  • Geographic attribute file - 2001
  • Block face data file - 2001 and onwards


We believe that the block face data file was discontinued for 2001 onwards due to concerns about maintaining confidentiality.

There is more information on the block-face definition, and changes made to it in 2001, in the Illustrated Glossary:

I couldn’t find the file in our 2001 Geography folder either, I asked someone and I was told that GeoSuite 2001 offers the same data as the 2001 Geographic Attribute File.

It can be downloaded from the Geography page of our website:

Monday, April 18, 2016

Further PCCF Usage Clarifications


Appendix II (Terms of Use) of the license states that:

"3. You will not use any part of the Data Product to develop or derive any other data product or data service for external distribution or commercial sale."
Appendix A (Approved Data Matching Uses) states that data can be matched for:
“Research Purposes
E.g.: Can be used in analysis to write articles that are published in journals. The data is not shared but the results are published. This also includes thesis for masters or Doctorate where results are required to be public”
1. Can StatCan clarify what it means by “data product or data service.” Is a dataset developed through secondary analysis with the PCCF considered a “data product” in this instance? I ask because “product” is a term often used by StatCan to refer to its many PUMFs, statistical tables, documentation sets, and reports/analyses.

2. Can results produced by research that incorporate the use of the PCCF be shared with an external party when no commercial transaction has occurred? i.e. Is there anything in the license that stops the researcher from sharing research results with an external party (e.g., a publisher, a government department, a municipality, a public health unit, a school board) when the research is non-commercial (cf. Term of Use #3, above)? The examples used in the Approved Matching Uses Appendix (cf. Appendix A clause, above) under “Research” illustrate instances that speak of results that are “published” (e.g., scholarly articles in journals) or theses and dissertations (which quite clearly must be open). However, not all research is published, so the example can confuse stakeholders, librarians and researchers alike. Much research our scholars undertake is not “published” in the traditional sense of the word.

3. In my case today, I am working with a researcher who is undertaking the analysis while writing his dissertation, and results of this analysis would in all likelihood be part of that document. As well as what is mentioned above, we are wondering if there is a difference between him (1) sharing unpublished results with an external party prior to the completion of his dissertation and, (2), an external party reading the results for themselves in the dissertation after he has received his PhD.

Thanks for your time on this. The PCCF products are valuable tools to our stakeholders, and we take the license itself seriously. Clarifications like these help build our understanding of licensee and licensor obligations.


1. We are referring to the PCFF, PCFRF and PCCF+ Statistics Canada data product, all of these data products contain the Postal codes that are owned by Canada Post Corporation. Anytime you create a new data product or dataset through your analysis using PCCF, PCFRF or PCCF+, you are creating a data product and you cannot share it or distribute it if it contains the Postal Codes.

2. Research results can be shared externally as long as they don’t contain a list of more than 1% of the PCCF, PCFRF or PCCF+. You can publish an extract of up to 1% of the PCCF, PCFRF or PCCF+ data products as long as the published data was used to support your published report or analysis.

‘The Licensee may publish an extract of up to 1% of the data product pursuant to Appendix ‘A’. This permission includes the use of the extracted data in support of analyses and in reporting of results and conclusions. The Licensee shall obtain approval from Statistics Canada before publishing extract of the data product in excess of 1%. ‘
3. Same answer as above, Research results can be shared externally as long as they don’t contain a list of more than 1% of the PCCF, PCFRF or PCCF+. You can publish an extract of up to 1% of the PCCF, PCFRF or PCCF+ data products as long as the published data was used to support your published report or analysis.

Thursday, April 14, 2016

Longitudinal Immigration Database


For the Longitudinal Immigration Database, what can we expect in terms of data dissemination? I see some have been released through the Daily and CANSIM in January of this year.
Any other plans? PUMF or RDC?

Could emigration data at the sub-national scale be gleaned from IMDB custom tables (that is, where in particular countries immigrants to Canada are originating from)?


The IMDB data have been released in December 2015 (Canada level) and January 2016 (provincial level) through the Daily and CANSIM. The IMDB micro-data are not released to users and they are not available in RDCs.

Please note that users can get custom tables by sending requests to the IMDB team.

The IMDB contains variables for Country of birth, Country of citizenshipand Country of last permanent residence. (Three distinct concepts.) Unfortunately, it does not contain any variables that would indicate from which specific region of a country they immigrated from. Sub-national emigration data, prior to landing in Canada, is unavailable.

Microdata for Monthly survey of food services and drinking places


I have a faculty member looking for microdata for the Monthly survey of food services and drinking places.

He would like the "type" of restaurant and the "location".


We publish NAICS 722511 (full service) 722512 (limited service) 7223 (The special food services industry group, which includes food service contractors, caterers and mobile food Services) and 7224 (drinking places).

The data is available at the provincial level by North American Industry Classification System (NAICS).

Long-term Care Survey


I have a researcher here who’s interested in the Long-term Care Facilities Survey. I’ve been able to find a couple of tables on CIHI’s site, but she’d like to know if we have access to more than these


While the Residential Care Facilities Survey was in CANSIM until 2010, the Long Term Care Facilities Survey that replaced it is not. Some summarization was made by CIHI. There is no other data published or freely available. Besides the data produced by CIHI, LTCFS data can only be retrieved by custom tabulation.

Monday, April 11, 2016

PCCF licensing questions


1. If we do a survey and collect postal code and use the conversion file to situate a non-identifiable person into a LHIN or Public Health Unit, and then provide the data to a health researcher, are we breaking the spirit of the agreement?

2. Commercial sale (number 4). Our research institute, as part of our university, is a not-for-profit, but we do get paid our costs for research we do outside of the institution. Are we correct in thinking this is not commercial work in the sense used here?

3. I am not sure I understand (number 8) the meaning off "publish an extract of up to 1% of the Data Product." Can you elaborate?

4. Appendix A number 2. The data is not shared but the results are public. If the data is shared but does not have postal code and other geographic units, such as LHIN, FED, together in the file is this

5. At our research institute, should I have everyone who uses the postal code file sign an agreement or should there be one form?


1. The answer is no, as long as the postal codes are not published.

2. Your research institute can use the data product to develop and derive work/services for its own use. It cannot distribute or share these products or services outside regardless of whether or not costs are recovered.

3. You can publish an extract of up to 1% of the file as long as the published data was used to support your published report or analysis.
 8. The Licensee may publish an extract of up to 1% of the data product pursuant to Appendix ‘A’. This permission includes the use of the extracted data in support of analyses and in reporting of results and conclusions. The Licensee shall obtain approval from Statistics Canada before publishing extract of the data product in excess of 1%.
4. The data product and/or extracts are not shared but the results based on the product or where the product was used to tabulate results are public.
Research purposes
E.g.: Can be used in analysis to write articles that are published in journals. The data is not shared but the results are published. This also includes thesis for Masters or Doctorate where results are required to be public.
5. This is a question for the DLI to answer and not Statistics Canada

NAICS at 6 digit level data needed


Is it possible to obtain input data (capital, labour) at the detailed (six digit) level? The researcher has been able to find it only at the 3 digit level. (e.g. CANSIM Table 383-0032).


We do not estimate or publish labour, capital or productivity data at the 6-digit level of NAICS. In most cases, our sample sizes are not sufficiently large enough to produce accurate or publishable estimates at that level of detail.

Access to PCCF products - legitimate use?


1. A researcher is working on a pain clinic program at the hospital and needs to find a way to easily measure distance between postal codes or FSAs and the different pain clinics. There are up to 2000 postal codes for the unique visits. Can you let me know if there is a product the researcher can use?

2. I know that that this is a legitimate use of the PCCF products (since the contact is an Assistant Professor) if they are simply writing a research paper for publication. I also know that the PCCF would be able to support this use (through GIS manipulation) with either “as the crow flies” distances or with other products for route mapping distances.

3. However, what if the research is being done simply to evaluate the program for the hospital’s purposes? Can I then provide the PCCF data for the project under our license? (I think that the answer is no, but want to confirm it).


1. There are no products that can provide this.

2. If you are using the Latitude and longitude on the PCCF file, please make sure you understand that they are the dissemination area, dissemination block, or block-face representative point and not the postal code (please refer to page 17 of the reference guide for more information).

3. No, unless the hospital has access to the PCCF product under a different licence other than the prof’s or the DLI.

Monday, April 4, 2016

Long term bond returns by Province


I am looking for annual average long-term bond yields by province from 1970 - 2008. I am trying to get some indicator of province X's long-term borrowing risk in year T. The only requirement is that the 'type' of bond be consistent across the observations. So, an example of exactly what I'm looking for is here: in the last column. Unfortunately, that is only for the federal government, and Statistics Canada doesn't seem to have anything for the provinces.I was able to pull something out of Bloomberg.


Statistics Canada unfortunately does not have any information on bond returns. I’ve sent a message to the Bank of Canada directly as they are cited as the main source of information. I suspect I won’t get a reply all that quickly, so it might be best if you follow up with them directly as well.

Labour Force Survey


I know that this has been asked before re: the .prn files, and that you need to edit the SPSS syntax file (.sps) to point to the data file location, and then you can run the syntax within SPSS to load the data.

Looking on the DLI mirror site (under doc -- spss) the .sps files stop in 2000.

Are there any current .sps files for LFS files post 2000 in another location?

To access the SPSS files (.sav’s), use the path on the MAD_DLI safe:

/MAD_DLI/Root/other-products/Labour Force Survey - lfs/1976-2015/data

The folders from 2000 onwards will have SPSS files located therein.

The .sps for post 1999 LFS data is on the main DLI site at /MAD_DLI/Root/other-products/Labour Force Survey - lfs/1976-2015/doc/spss/lfs-2000-yyyy.sps

Tuition and Living Accommodation Costs (TLAC) Survey


I have a researcher looking for data from the Tuition and Living Accommodation Costs (TLAC) Survey:

She seeks the average undergrad tuition by province (which I can find here: but I suspect she will have further questions.

Is the actual survey available?


The client just needs to go to the link below(IMDB) that you provided and select by clicking the related products and select the Free CANSIM tables. We have the weighted average tuition fees by Canadian and International by graduate/undergraduate and by provinces