In the design and perception of new products it is important to specify the contributions made by to different facets or elements. The overall utility and acceptance of such a new product can then be estimated and understood as a possibly additive function of the elementary utilities. Examples are the design of cars, a food article or the program of a political party. For a new type of margarine one may ask whether a change in taste or presentation will enhance the overall perception of the product. The elementary utilities are here the presentation style and the taste (e.g., calory content). For a party program one may want to investigate whether a stronger ecological or a stronger social orientation gives a better overall profile of the party. For the marketing of a new car one may be interested in whether this new car should have a stronger active safety equipment or a more sporty note or combinations of both.
In Conjoint Measurement Analysis one assumes that the overall utility can be explained as an additive decomposition of the utilities of different elements. In a sample of questionnaires people ranked the product types and thus revealed their preference orderings. The aim is to find the decomposition of the overall utility on the basis of observed data and to interpret the elementary or marginal utilities.
car 1: | basic safety equipment | and | low sportiness |
car 2: | basic safety equipment | and | high sportiness |
car 3: | high safety equipment | and | low sportiness |
car 4: | high safety equipment | and | high sportiness |
The elementary utilities here are the safety equipment and the level of sportiness. Conjoint Measurement Analysis aims at explaining the rank order given by the test person as a function of these elementary utilities.
product 1: | low calories | and | plastic pot |
product 2: | low calories | and | paper package |
product 3: | high calories | and | plastic pot |
product 4: | high calories | and | paper package |
These four fictive products may now be ordered by a set of sample testers
as described in Table 16.2.
The Conjoint Measurement Analysis aims to explain such a preference ranking by attributing part-worths to the different elements of the product. The part-worths are the utilities of the elementary components of the product.
In interpreting the part-worths one may find that for a test person one of the elements has a higher value or utility. This may lead to a new design or to the decision that this utility should be emphasized in advertisement schemes.