User profiles, which personalize applications, are an important factor in designing on-line computer systems. This study targets user profiles to better understand the interactive process of users and consultants in social science research studies. Definition of such profiles should be a basis for designing an electronic consulting system. The results of an empirical survey of both users and statistical consultants are presented. Questionnaires and interview were used to identify the different perceptions of using statistics, analysis tools, typical hypotheses, interpretation of results, and related topics. The findings shed light on the users of statistical systems and the role of the consultant in supporting the conduct of a research study. We found that the majority of users possess only a basic course in statistics, find difficulty in stating their problem, and two-thirds of them use consultants to define and design their study. The implication for an electronic consulting system are that such systems should pay attention to the initial research problem definition and should use a system-initiated question/answer approach to elicit problem information from the users. Based on the findings, design guidelines are presented for an electronic consulting system that can bridge the gap between the user and the wide range of statistical methods available in commercial packages.
Berg, A., & Spiegler, I. (2000). User Profiles as a Basis for an Electronic Statistical Consulting System. Communications of the Association for Information Systems, 3, pp-pp. https://doi.org/10.17705/1CAIS.00312