Interrupted Time-Series Design
Definition
A design that compares the trend in a crime series before and after an identified intervention date, testing for a change in level and a change in slope against the pre-intervention pattern.
Related terms
- Crime Rate
- The number of recorded offences of a given type divided by the population at risk, usually per 100,000 people, which allows comparison...
- Ecological Fallacy
- The error of applying a statistical association observed at the population or group level to an individual, without verifying that the individual...
- Negative Binomial Regression
- A count regression model that adds a dispersion parameter to absorb overdispersion, the pattern where a crime count's variance exceeds its mean,...
- Poisson Regression
- A regression model for count outcomes, such as burglaries per ward per month, that assumes the variance of the counts equals their...
- Trend Index
- A rate rebased to 100 in a chosen base year so that later years can be read directly as a percentage change...