For marketing research, many decisions, evaluations, or purpose of evaluation are done by the survey or questionnaire to seek for people's consensus. The commonly used methods are the surveys or questionnaires: asking people to think in binary logic way from a multiple choice design. However these processes often ignore the fuzzy thinking perceived in human logic and recognition. Thus, employing the membership function and statistical fuzzy analysis will be a more heuristic way. This paper aims at developing a more realistic decision framework for marketing research. The motivation stems from the fact that in many practical problems such as price of a new merchandize, intelligent property, pollution penalty, people's selection for a party… etc. During the survey, people are often encountered with uncertainty or imprecision problems. In this paper we will provide the definitions of fuzzy mode, fuzzy median, and their related properties. We hope that these soft computing methods will make the corresponding fuzzy techniques more practical and reasonable in the future study of management science.