Police Administration Survey

Detailed information for 2002

The objective of this survey is to collect baseline information on police personnel and expenditures to enable detection of historical trends as well as permit comparisons at the provincial/territorial and municipal levels.

Data release - December 20, 2002

• Questionnaire(s) and reporting guide(s)
• Description
• Data sources and methodology
• Data accuracy

Description

The objective of this survey is to collect baseline information on police personnel and expenditures to enable detection of historical trends as well as permit comparisons at the provincial/territorial and municipal levels.

This survey collects data from police services across Canada under the authority of the Statistics Act, Revised Statutes of Canada, 1985, Chapter S19. All municipal, provincial and federal police services in Canada are surveyed.

Respondents provide the number of police officers employed by the service (full-time equivalents), categorized by rank (senior officers, non-commissioned officers, and constables) and by sex. They also provide the number of civilian employees, categorized by job type and by sex. Other questions collect data on the number and type of police vehicles and operating expenditures broken down into salaries/wages, benefits, and other operating expenditures.

Data from this survey provide information on total expenditures on policing and the number of officers in each province and in Canada as a whole, as well as the number of officers per hundred thousand population and per capita cost for each municipal police service.

The data are intended for police services, for officials with responsibility for police budgets, for policy-makers with policing-related responsibilities, and for members of the general public.

Statistical activity

The survey is currently administered as part of the National Justice Statistics Initiative (NJSI). Since 1981, the federal, provincial and territorial Deputy Ministers responsible for the administration of justice in Canada, with the Chief Statistician, have been working together in an enterprise known as the NJSI. The mandate of the NJSI is to provide information to the justice community as well as the public on criminal and civil justice in Canada. Although this responsibility is shared among federal, provincial and territorial departments, the lead responsibility for the development of Canada's statistical system remains with Statistics Canada.

Reference period: Personnel: June 15 of the survey year. Expenditure: the previous calendar year for most municipal services and the fiscal year for provincial and federal services as well as some municipal services.

Subjects

• Crime and justice
• Crimes and offences
• Justice system spending

Data sources and methodology

Target population

All Canadian municipal, provincial and federal police services that were active on June 15th of the survey year.

Instrument design

For municipal and provincial police services, a questionnaire is used. The questionnaire is a paper-and-pencil respondent-completed instrument. The questions either ask for a single number (e.g. the number of male constables or the dollar amount spent on salaries), or provide closed response check-boxes. The RCMP submit data electronically.

The last major revision to the questionnaire was done in 1996. Changes were made to both the personnel and the expenditure section of the questionnaire. A separate category for "school crossing guards" was added and adjustments were made to civilian personnel back to 1986 to reflect this change. Expenses associated with the purchasing and leasing of vehicles were to be included only if they were part of the operational budget of the police service. The "benefits" category was expanded to include those paid by other government sources. The removal of expenses due to the capital purchases of vehicles, and the addition of expenses due to the inclusion of benefits paid by other government sources were examined in detail. It was estimated that these revisions accounted for an overall change of less than 1% to reported expenses.

Sampling

This survey is a census with a cross-sectional design.

Data are collected for all units of the target population, therefore, no sampling is done.

Data sources

Responding to this survey is voluntary.

Data are collected directly from survey respondents.

A survey form has been developed pertaining specifically to police administration statistics. The survey is sent to all municipal and provincial police forces across Canada. After completion, the survey forms are returned to the Centre where they are clerically edited and data captured. The Royal Canadian Mounted Police (RCMP) submit data to the Centre in electronic format which is then merged with the data captured from the other respondents.

View the Questionnaire(s) and reporting guide(s) .

Error detection

Respondents have no difficulty in completing the survey due to the rudimentary level of information requested. Even though few respondent-generated errors are anticipated, a comprehensive set of edits has been devised which check various logical and year-to-year calculations. Data are captured and the first set of edits checks for legibility, verifies totals and compares the submitted data to the previous year (to check for any significant changes). In cases where the data provided is either not legible or exceeds predetermined tolerances, contact with the respondent is initiated to clarify survey responses. Edits are in place to identify incorrect totals and invalid respondent codes. In case a variation is deemed acceptable after contacting the respondent, a comment is added to those records explaining the reasons. Once all the data has been captured, tolerance and outlier edit programs are run against the data to identify any data falling out of the prescribed tolerance range. Such cases are resolved after verifying the respondent status for the current year. Any jurisdictional changes are taken into account when resolving data issues.

Historical trend comparisons are made to identify any other anomalies.

Imputation

Each year some police services do not respond to the survey. If data for a non-responding police service are available for the previous year, estimates are made based on that data.

Estimation

This methodology type does not apply to this statistical program.

Quality evaluation

Data quality is evaluated by doing historical trend analyses of key variables. Historical analyses of rates are done based on personnel and expenditure data (police per 100,000 population, per capita costs, etc.). Police-civilian ratios and male-female ratios are also checked over time to verify historical consistency. Overall per capita costs are derived and analyzed historically.

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects which could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.

A cover letter requesting their participation and an authorization form confirming their signed permission to publicly release the data is sent to each police service.

Revisions and seasonal adjustment

Each year, some respondents send in revisions to data submitted in previous years. These revisions may result from internal audits done by the respondents, or they may result from data quality queries initiated by the survey manager. If there is inconsistency in the year-over-year data comparisons, the respondent may review the information sent by them in previous years and decide to revise them. In some cases, revisions are made to correct prior misinterpretation of definitions.

Data accuracy

The survey collects census data as compiled by police services. Formal data quality indicators are not part of the survey methodology, beyond annual review for consistency and logical relationships between responses. While considerable effort is made to ensure high standards throughout all stages of collection and processing, the resulting estimates are inevitably subject to a certain degree of non-sampling error. Examples of non-sampling error are coverage error, data response error, non-response error and processing error.

Coverage error can result from incorrect addresses of police services or absence of contact information for new police services.

Data response error may be due to questionnaire design, the characteristics of a question, inability or unwillingness of the respondent to provide correct information, misinterpretation of the questions or definitional problems. These errors are controlled through careful questionnaire design and the use of simple concepts and consistency checks. However, since the respondents and/or their staff have been changing quickly due to amalgamations and regular turnover, some lack the background to ensure consistency. Therefore, the survey analyst is conscious of the need to monitor reporting and to discuss any anomalies with the police services in question. Detection of data errors is done at a very early stage. As soon as a survey response is received, the responses are compared to the previous year's data and if necessary, a trend analysis is done. Jurisdictional changes are kept in mind when doing these checks. If a response is over a tolerance range, the respondent is called and explanations and/or changes are made as a result. In some cases, part-time employees are not converted to full-time equivalents and this is resolved after discussing with the respondent.

Non-response error is related to respondents that may refuse to answer, are unable to respond or are too late in reporting. In these cases, data are imputed. The extent of any imputation error decreases with increases in the response rate and attempts are therefore made to obtain as high a response rate as possible while minimizing the response burden. The survey manager liaises with the respondents to maintain the high response rate of approximately 98% for the whole survey.

Processing error may occur at various stages of processing such as in data entry. Measures have been taken to minimize these errors. Processing errors are minimal due to consistency edits built into the processing system. Edits are in place to verify totals and invalid respondent codes. A respondent level tolerance edit process has been built into the data capture system to identify anomalies.