Offences
The number of offences is useful for measuring the volume of crime in an area. Police use this information for operational purposes. The percentage change between the number of offences in the current 12 month period compared with the 12 months prior shows the change in volume of crime across time.
Rates per 100,000 population
Comparisons between areas cannot be made using the number of offences, because areas with larger populations are likely to experience higher volumes of crime. The rate of offences per 100,000 population allows comparisons to be made across different geographical areas and across time, because the population size is taken into account in the calculation. Crime rates also show the likelihood of experiencing a crime in a given area.
Ratios
The ratio of a Police Service Area's (PSA) crime rate to the Victorian crime rate can be used to compare a PSA's relativity to the state crime rate.
The 'ratio to VIC rate' is calculated as the ratio of a PSA crime rate per 100,000 population to the Victorian crime rate per 100,000 population. This enables comparison between the PSA's crime rate and the state rate. For example, a ratio of 1 indicates that the PSA's crime rate is equal to the Victorian rate. A ratio of 2 signifies that the PSA's crime rate is double the State's; a ratio of 0.5 indicates tha the PSA's rate is half of the State's.
It should be noted that crime rates in areas with large transient populations (eg. tourists or commuters) should be interpreted with caution. Rates are calculated using the estimated resident population, and so areas that experience a large transient population, such as Melbourne CBD, are likely to have high crime rates.
Minimum values to report percentage changes
My Place will not display percentage changes when the number of 'offences recorded' for any category in a PSA is considered to be too small (less than 23) for any statistical significance.
This minimum value (23) was derived by statistically analysing approximately 175,000 rolling 12 month samples of Victoria Police data for the period 2000 to 2010.
The 175,000 samples consisted of 120 rolling 12 month periods for each of the 27 offence categories recorded in Law Enforcement Assistance Program (LEAP) data base for all the 54 Police Service Areas (PSA).
The following three statistical techniques were used to derive the value (23):
- an 'Outlier Detection' algorithm was used to reject any 12 month sample which contained a monthly value that was more than three (3) standard deviations from the mean
- a non-parametric significance test called a 'Kolmogorov-Smirnov test' was then performed on each sample to test for normality
- a weighted 'Logistic Regression Model' was then fitted to the results from the normality tests to determine the rolling twelve month value which gave more than 95% confidence of being a normal distribution
The following graph depicts the result.
Produced by Corporate Statistics, Victoria Police, November 29, 2011
Updated