Frau Dr. Sonja Radas hat am 24. Januar einen Gastvortrag zum Thema "Uncovering Hidden Knowledge in Stated Purchase Intentions: Application of Bayesian Truth Serum" gehalten.

Abstract:
In new product development, purchase intention survey is often used for early sales forecasting and elimination of potentially failing concepts.
Regardless of its usefulness, it is known that there is a systematic discrepancy between stated purchase intentions and purchase incidence, resulting from both temporal and non-temporal causes. The latter, which are most often ascribed to systematic biases, are usually corrected by use of sophisticated statistical models which require large quantities of respondent and product related data.
In this paper we offer a new explanation for this discrepancy: we suggest that those respondents who are not sufficiently knowledgeable about the product report unreliable purchase intentions, which introduces noise in the predictions. So in order to improve forecasting precision, we propose to identify those respondents who are knowledgeable, isolate their purchase intentions and disregard the information from the others. This task is made more difficult by the fact that this knowledge is often tacit, especially for new products with higher degree of novelty.
We address this issue by building a theoretical model that can identify respondents who have tacit product related knowledge. Required data consists of answers to only two questions: (1) "will you buy the product", and (2) "what percentage of your peers will make the same choice as you". Having assumed that respondents themselves have private information about knowledgeable segments (which is invisible to the researcher), we employ the Bayesian Truth Serum methodology to uncover respondents’ belief structure.