S. Issanchou

INRA, UMR1129 FLAVIC, F-21000 Dijon, France

Hedonic measurements are frequently used in the first stages of product development. However, there are more and more criticisms related to the fact that most of these measurements are conducted in a laboratory or in a central location, i.e. in situations which differ largely from real consumption situations. Thus it raises the question of the external validity of data collected in such situations.

During this talk, some possible limits related to three components of these measurements, i.e. to the respondents, the product and the external context, will be reviewed.

Concerning respondents, a key point is the inter-individual variability. This raises the question of panel size but also of how well the panel is representative of the target population. Besides inter-individual variability, there is also intra-individual variability. We will examine the possible factors that can induce such instability at the individual level on a very short term or after repeated consumption and in particular those which seem the most neglected while they can impact the external validity of the laboratory measurements.

Concerning product, besides the way the product is presented the impact of the quantity tested by the consumers has been raised but seems still often neglected.

Concerning context, we will first consider the questions asked to the consumers and the declarative nature of the answers which can be consequently loose or, sometimes biased by social desirability. We will examine recent data collected thanks to a combination of hedonic measurements and methods which have the advantage to increase participants’ commitment.