Methodological overview
The following information gives an overview of the key methodological elements of the New Zealand Crime and Safety Survey (NZCASS) in 2014.
More detail is given in the NZCASS Technical manual [PDF, 5.3 MB]
Overview
Nationwide, face-to-face random probability survey, with one respondent selected per household using multistage stratified cluster sampling methods.
Target population
Total usually resident, non-institutionalised, civilian population of New Zealand aged 15 years and over.
Sampled areas
North Island, South Island and Waiheke Island.
Dwellings included
Permanent, private dwellings.
Composition of sample
Two samples are drawn as part of the NZCASS: a general or 'main sample' and a Māori booster sample that aims to increase sample sizes for Māori.
Sample sizes and response rates
| Sample | Interviews achieved (2014) | Response rates (2014) |
|---|---|---|
| Main | 5,235 | 80% |
| Māori booster | 1,708 | 84.8% |
| Total | 6,943 | 81.0% |
Period when interviews took place
Interviews were conducted between 10 February 2014 and 6 July 2014.
Interview length
On average the questionnaire took 40 minutes and 45 seconds to administer in 2014.
Recall period
For most critical questions, participants were asked to think of incidents that took place between 1 January 2013 and the date of the interview. Some other recall periods may apply at different times throughout the questionnaire.
Crimes / offences
In the NZCASS, questions are asked about different things (incidents) that might have happened to the survey participant or their household. These incidents are then coded by legal experts to determine whether or not the incident was a crime, and what type of offence (or offences) took place.
The NZCASS does NOT ask survey participants directly about crimes that happened to them, such as ‘Have you been burgled?’
This is because people don’t always know what are legally considered crimes, and what aren’t. Once collected, the incidents are coded by legal experts (Offence coding) to determine whether or not the incident was a crime, and what type of crime.
Comparing the NZCASS surveys
Results from the NZCASS 2014 can be compared with results from the NZCASS 2009 and 2006. The methodology and application is largely consistent. Some changes were made to:
- Improve response rates
- Improve project efficiency and transparency
- Bring questions and analysis in line with Statistics NZ standards and classifications
- Better meet users information needs
A summary of the main changes can be found on the Comparability between survey years page. More detail is given in the NZCASS Technical manual.
NZCASS Technical manual [PDF, 5.3 MB]
Weighting
Three types of weighting was applied: household weights and individual weights ensure results are representative of the New Zealand population, while incident weights adjust for the fact that detailed information was only collected for up to six incidents (‘Victim form information’).
Imputation
A summary of the imputation applied in the NZCASS can be found on the Imputation page. More detail is given in the NZCASS Technical manual. [PDF, 5.3 MB]
NZCASS Technical manual [PDF, 5.3 MB]
This page was last updated:
