The effect of attribute questions on overall liking ratings

In food acceptance tests, it is typical for respondents to rate a product on overall liking and on a series of product attributes. This study was designed to determine if the inclusion of attribute questions can alter a respondent’s rating of overall liking. Five groups of respondents evaluated four variations of a dairy dessert. Each group used a different questionnaire, comprised either of (a) overall liking only; (b) overall liking plus intensity scales, which asked respondents to rate the strength of 10 sensory characteristics; (c) overall liking plus attribute liking scales, which asked respondents to rate their liking of appearance, flavor and texture of the products; (d) overall liking plus just-about-right scales, which asked respondents to indicate whether the level of 10 sensory characteristics was ‘‘too high’’, ‘‘too low’’, or ‘‘just about right’’; and (e) overall liking plus attribute liking and just-about-right scales. The inclusion of just-about-right questions on the questionnaire changed the average overall liking ratings of the products, whereas intensity scales had no such effect. Attribute liking questions also tended to alter overall liking ratings, but to a lesser extent than just-about-right questions. Several hypotheses concerning the source of these context effects are discussed. � 2003 Elsevier Ltd. All rights reserved.