Using Expert Systems To Model And Improve Survey Classification Processes

Frederick G. Conrad


In order to analyze most survey responses, the data must be assigned to categories. When the response task does not include classification (as in answering open ended questions) responses are classified by coding specialists, usually after the raw data are collected. This chapter distinguishes between simple and complex classification and argues that expert system software is particularly good at supporting the second of these. Simple classification usually involves locating the response in some sort of dictionary where it is explicitly mapped to a particular category. The literature about automated coding of this sort is first reviewed. Then complex classification is presented as a task requiring the coding specialist's expertise about the content area. This is illustrated with two prototype expert systems that embody specialists' knowledge of occupational classification in the Occupational Compensation Survey Program and the review of commodity substitutions in the Consumer Price Index. Finally, some of the practical considerations of implementing expert systems in survey organizations are presented.