Using Multinomial Logistic Regression to Analyze a 1-1 Matched Case-Control Study

A matched case-control study is a retrospective method of data collection that is often used when the event of interest is unlikely, or the design of a prospective experiment is impractical or unethical.

For example, consider an insurance company that is studying the risk factors that indicate whether a client will have to make a claim on a 10-year term life insurance contract. The chance of a claim on a randomly selected contract is low enough that rather than collecting a large enough sample to ensure there are enough claims in the sample to make valid inferences, the analysts choose to conduct a matched case-control study. Suppose information on current customers is contained in insure.sav. See the topic Sample Files for more information.

Here, cases are contracts in which there was a claim, and controls are contracts in which there was not. Cases and controls are matched on the values of Age and Gender, which might have an effect on the probability of a claim on given contract, but are not currently of interest to the analysts.

Next