Abstract:

One of the new developments in the field of market research during the last decade is the construction of fusion techniques. We developed a procedure called VENUS. This method was applied in the Netherlands several times, and a cumulation of experience has led to new insights in application possibilities and validation procedures. The data has enabled us to test the adequacy of our fusion procedure, not only at an aggregate level but also at an individual level. Emphasis in this paper is on the evaluation of the procedure. Three concrete cases will be treated: a. injection of psychographic data into a readership survey, b. fusion of psychographics with readership data, c. fusion of tv-watching data and radio-listening data. The presented analyses show that fusion is a very acceptable solution. If single - source measurement is prohibited by financial and practical aspects, fusion and injection are under-utilized tools. Another important conclusion from our research is that the success of simulation is dependent upon: (a) the variables for which scores are simulated and (b) the segment to which donor and recipient belong. Behavior is better to simulate than attitudes, but within behavior the general activities are more difficult to simulate than specific activities. Furthermore it is easier to find a similar donor for lesser educated individuals than for higher educated people.