Abstract
Five uses of the Automatic Interaction Detector (AID) program in the analysis of health survey data are illustrated in application to examples of increasing complexity. The “forced trees” technique is discussed as a solution to the problem of substitution effects, and results of the AID technique are compared with those of regression analysis. Alternative research strategies are proposed for joint use of the two methods.
Full text
PDFSelected References
These references are in PubMed. This may not be the complete list of references from this article.
- Bice T. W., White K. L. Factors related to the use of health services: an international comparative study. Med Care. 1969 Mar-Apr;7(2):124–133. doi: 10.1097/00005650-196903000-00006. [DOI] [PubMed] [Google Scholar]
- Wirick G. C., Jr A multiple equation model of demand for health care. Health Serv Res. 1966 Winter;1(3):301–346. [PMC free article] [PubMed] [Google Scholar]