Using General Loglinear Analysis to Model Accident Rates

An insurance company is interested in studying age and gender risk factors for automobile accidents in a given region. Using the simple loglinear models previously discussed, the analysts can model the cell counts of the cross-classification of Age category and Gender as coming from Poisson distributions and estimate the main and interaction effects.

However, modeling the raw cell counts can be misleading in this situation, because the numbers of drivers in the region, represented by the variable Population at risk, varies from cell to cell. The analysts want to model the accident rates, which can be accomplished in the General Loglinear Analysis procedure by specifying Population at risk as the Cell Structure variable.

This information is collected in accidents.sav. See the topic Sample Files for more information. Use General Loglinear Analysis to model the accident rates.

Next