4524.0 - In Focus: Crime and Justice Statistics, December 2011  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 01/12/2011   
   Page tools: Print Print Page Print all pages in this productPrint All  


Statistics presented in this article are based on data collected by the ABS national Crime Victimisation Survey conducted from July 2009 to June 2010, as part of the Multipurpose Household Survey (MPHS). The MPHS is conducted each financial year throughout Australia from July to June as a supplement to the ABS’ monthly Labour Force Survey (LFS), and is designed to provide statistics for a number of self-contained topics. For more information about the scope, coverage, and data collection methodology of the MPHS please refer to the explanatory notes in Crime Victimisation, Australia, 2009-10 (cat. no. 4530.0) (Endnote 4).

The Crime Victimisation Survey included a social disorder module for the first time in 2009-10, which comprised a series of questions relating to respondents’ perceptions and opinions of social disorder issues in their local area. The definition of ‘local area’ was left to the interpretation of the respondent, and may include a whole town or suburb, or just the streets surrounding the respondent’s home. However, if the respondent was unsure the interviewer defined it as ‘the area around where you live.’ Questions about social disorder were asked of all respondents aged 18 years and over. Multiple social disorder issues could be reported, as footnoted at the bottom of each relevant graph (see Appendix for full list of questions asked and social disorder issues).

As a sample survey, the Crime Victimisation Survey is subject to both sampling and non-sampling error. Sampling error is the difference between the published estimates, derived from a sample of persons, and the value that would have been produced if the total population (as defined for the scope of the survey) had been included in the survey. In addition, information collected in the survey is essentially ‘as reported’ by respondents, which makes errors in reporting by respondents a potential source of non-sampling error. For example, non-sampling error can occur as a result of recall bias – the ability of the respondent to accurately and completely recall memories of past experiences and events. Crime Victimisation Survey respondents were asked to recall their experiences and perceptions of events that occurred in the 12 months prior to interview (For more information refer to Explanatory Notes 19-27 of Crime Victimisation, Australia, 2009-10 (cat. no. 4530.0) (Endnote 4).

A statistical significance test for a comparison between estimates was performed to determine whether it was likely that there was a difference between corresponding population characteristics. The standard error of the difference between two corresponding estimates was used to calculate a test statistic. If the absolute value of this test statistic was greater than 1.96 then there was evidence, with a 95% level of confidence, of a statistically significant difference in the two estimates with respect to that characteristic. Any proportions or values described as significant in the text are significant at this level. For more information on the statistical concepts and methodology underpinning the Crime Victimisation Survey please refer to the explanatory and technical notes in Crime Victimisation, Australia, 2009-10 (cat. no. 4530.0) (Endnote 4).

Previous PageNext Page