Conjoint Analysis

Conjoint analysis is a statistical technique used to calculate the value – also called partial utility – attached by consumers to varying levels of physical characteristics and/or price. Conjoint analysis is conducted by showing respondents a set of fictitious products called profiles – each profile having a specific price and specific levels of a limited number of attributes – and asking respondents to sort these profiles by decreasing order of preference (or to rate them). By analyzing the preference data and the combination of attributes and price for each product, the methodology evaluates the partial utility (preference) attached by respondents to each individual characteristic making up the product (from global preferences expressed by respondents about each profile).

This study is rather complex and expensive and is therefore not always made available to participants. Check with your professor if this study will be available in your course. The complexity of the study increases dramatically with the number of attributes and the number of levels included in the study. Hence, only price and the three physical characteristics that are perceived as most important are studied; four levels are tested for each attribute. For instance, the four prices $288, $367, $446 and $525 will be tested.

Relative importance of price and physical characteristics. The chart depicted in

Figure 39 – Conjoint Analysis – Relative importance of attributes             

 shows the relative importance of price and the three physical characteristics that are perceived as most important in the market. Note that importance ratings for a given segment sum to 100%.

Utility charts. The charts depicted in Figure 40 show the partial utilities attached to four levels in each dimension included in the study. Partial utilities are measured on a scale from 0% (very low utility) to 100% (very high utility): the higher the partial utility, the higher the consumer’s preference for the corresponding level of this characteristic. The four levels have been chosen in the feasible range for the dimension (e.g.: from 5 to 100 for Proc. Power) to test varying levels of interest. Results are broken down by consumer segment.

It Is important to note that the level with the highest utility is not necessarily the ideal level, as given in the semantic scales or MDS studies. For instance, in Figure 40, the ideal price may be anywhere between $288 and $525.