Keyword search
Filter results by
Search HelpKeyword(s)
Subject
Type
Survey or statistical program
Results
All (3)
All (3) ((3 results))
- 1. Collecting crime data in Canada ArchivedArticles and reports: 11-522-X200800011012Geography: CanadaDescription:
Justice surveys represent a unique type of survey undertaken by Statistics Canada. While they all essentially use administrative data, Statistics Canada has had considerable input into the type of data that is collected as well as quality assurance methods guiding the collection of this data. This is true in the areas of policing, courts and corrections. The main crime survey, the Uniform Crime Reporting Survey (UCR), is the focus of this paper and was designed to measure the incidence of crime in Canadian society and its characteristics. The data is collected by the policing community in Canada and transmitted electronically to Statistics Canada. This paper will begin by providing an overview of the survey and its distinctive properties, such as the use of intermediaries (software vendors) that convert data from the police's information systems into the UCR survey format, following nationally defined data requirements. This level of consistency is uncommon for an administrative survey and permits a variety of opportunities for improving the overall data quality and capabilities of the survey. Various methods such as quality indicators and feedback reports are used on a regular basis and frequent two-way communication takes place with the respondents to correct existing data problems and to prevent future ones. We will discuss recent improvements to both the data itself and our collection methods that have enhanced the usability of the survey. Finally, future development of the survey will be discussed including some of the challenges that currently exist as well as those still to come.
Release date: 2009-12-03 - Articles and reports: 11-522-X20010016295Description:
This paper discusses in detail issues dealing with the technical aspects of designing and conducting surveys. It is intended for an audience of survey methodologists.
In order to compensate for unreported, missing, unreasonable, or unusable data, the Uniform Crime Reporting (UCR) Program conducts data estimations and imputations using a variety of statistical methods. This paper illustrates how offence and arrest data are estimated using a variety of different approaches. The paper also points out the strengths and the shortcomings of each approach.
Release date: 2002-09-12 - 3. Probability of victimization over time: Results from the U.S. National Crime Victimization Survey ArchivedSurveys and statistical programs – Documentation: 11-522-X19980015033Description:
Victimizations are not randomly scattered through the population, but tend to be concentrated in relatively few victims. Data from the U.S. National Crime Victimization Survey (NCVS), a multistage rotating panel survey, are employed to estimate the conditional probabilities of being a crime victim at time t given the victimization status in earlier interviews. Models are presented and fit to allow use of partial information from households that move in or out of the housing unit during the study period. The estimated probability of being a crime victim at interview t given the status at interview (t-l) is found to decrease with t. Possible implications for estimating cross-sectional victimization rates are discusssed.
Release date: 1999-10-22
Data (0)
Data (0) (0 results)
No content available at this time.
Analysis (2)
Analysis (2) ((2 results))
- 1. Collecting crime data in Canada ArchivedArticles and reports: 11-522-X200800011012Geography: CanadaDescription:
Justice surveys represent a unique type of survey undertaken by Statistics Canada. While they all essentially use administrative data, Statistics Canada has had considerable input into the type of data that is collected as well as quality assurance methods guiding the collection of this data. This is true in the areas of policing, courts and corrections. The main crime survey, the Uniform Crime Reporting Survey (UCR), is the focus of this paper and was designed to measure the incidence of crime in Canadian society and its characteristics. The data is collected by the policing community in Canada and transmitted electronically to Statistics Canada. This paper will begin by providing an overview of the survey and its distinctive properties, such as the use of intermediaries (software vendors) that convert data from the police's information systems into the UCR survey format, following nationally defined data requirements. This level of consistency is uncommon for an administrative survey and permits a variety of opportunities for improving the overall data quality and capabilities of the survey. Various methods such as quality indicators and feedback reports are used on a regular basis and frequent two-way communication takes place with the respondents to correct existing data problems and to prevent future ones. We will discuss recent improvements to both the data itself and our collection methods that have enhanced the usability of the survey. Finally, future development of the survey will be discussed including some of the challenges that currently exist as well as those still to come.
Release date: 2009-12-03 - Articles and reports: 11-522-X20010016295Description:
This paper discusses in detail issues dealing with the technical aspects of designing and conducting surveys. It is intended for an audience of survey methodologists.
In order to compensate for unreported, missing, unreasonable, or unusable data, the Uniform Crime Reporting (UCR) Program conducts data estimations and imputations using a variety of statistical methods. This paper illustrates how offence and arrest data are estimated using a variety of different approaches. The paper also points out the strengths and the shortcomings of each approach.
Release date: 2002-09-12
Reference (1)
Reference (1) ((1 result))
- 1. Probability of victimization over time: Results from the U.S. National Crime Victimization Survey ArchivedSurveys and statistical programs – Documentation: 11-522-X19980015033Description:
Victimizations are not randomly scattered through the population, but tend to be concentrated in relatively few victims. Data from the U.S. National Crime Victimization Survey (NCVS), a multistage rotating panel survey, are employed to estimate the conditional probabilities of being a crime victim at time t given the victimization status in earlier interviews. Models are presented and fit to allow use of partial information from households that move in or out of the housing unit during the study period. The estimated probability of being a crime victim at interview t given the status at interview (t-l) is found to decrease with t. Possible implications for estimating cross-sectional victimization rates are discusssed.
Release date: 1999-10-22