CHANGE IN RATES OF CRIME OVER TIME

INTRODUCTION

This section provides an analysis of changes in crime victimisation rates over the period 2008-09 to 2016-17 for selected personal and household crime types in Australia and in each state and territory, using fitted function techniques and significance testing.

Personal crimes include:

• Physical assault
• Face-to-face threatened assault
• Non face-to-face threatened assault
• Robbery (national level only)
• Sexual assault (national level only)

Household crimes include:

• Break-in
• Attempted break-in
• Motor vehicle theft
• Theft from a motor vehicle
• Malicious property damage
• Other theft

Fitted functions

Year-on-year comparisons of crime victimisation rates alone often show little, if any change, and this type of comparison does not reveal patterns unfolding over a longer time period. To provide an overview of the general pattern in victimisation rates over time, functions were fitted to the data. Fitted functions give a better indication of changes over a longer time period.

The fitted functions in this analysis do not take into account the survey error associated with sample estimates, so other criteria were used in deciding when it was appropriate to apply these techniques (see Technical Note for details).

Significance testing

Where there was no fitted function, significance testing was undertaken to examine whether there was a statistically significant difference between the 2008-09 and 2016-17 victimisation rates, with any significant differences highlighted in the text. Whilst fitted functions analyse general patterns in victimisation rates over time, significance testing is a comparison of two time points only, and does not take into account data patterns across the entire time series. Unless functions were able to be fitted to the data, any apparent patterns in victimisation rates over time should be interpreted with caution.

To assist in making comparisons, data series’ that did not meet the inclusion criteria for fitted function analysis have also been included in the line graphs, provided that the estimates were not subject to high survey error.

For more information about survey error, significance testing, and fitted function analysis, refer to the Technical Note.