3. Importance measuresThe second major difficulty facing IPA, and the focus of this paper, is the measurement of importance and performance.In this regard, performance has generally been less controversial than importance: the usual measurement procedure has been to take the mean of the performance ratings obtainedfrom an appropriate group of people by means of a metric or Likert scale. However, a variety of means exist to perform importance measurement. In particular, two quite differentkinds of measure are common in IPA applications: 1) direct measures based on Likert scale, k-point scale or metric ratings obtained in the same way as for performance; and 2) indirectmeasures obtained from the performance ratings, either by multivariate regression of an overall product or service rating on the ratings given to the individual attributes (Danaher andMattsson, 1994; Dolinsky, 1991; Neslin, 1981; Taylor, 1997;Wittink and Bayer, 1994) or by means of conjoint analysistechniques (Danaher, 1997; DeSarbo et al., 1994; Ostrom andIacobucci, 1995).Although a recent review of these methods (Bacon, 2003)supports earlier studies (e.g. Alpert, 1971; Heeler et al., 1979) in finding that direct measures capture the importance of attributes better than indirect measures, the former have serious drawbacks.In particular, Bacon (2003) himself emphasizesthat direct importance assessment is often misleading becauseratings are uniformly high. The main source of this problem isinherent to Martilla and James' procedure, in which the first step should be to identify themost salient attributes of a product or service by qualitative studies (focus groups and/orunstructured personal interviews) or by reviewing previous research. This procedure has a natural tendency to record high importance ratings on a metric or Likert scale for all theattributes selected for evaluation, with the result that they all crowd together at the top of the IPA grid. This defeats the purpose of the exercise, since the objective of the IPA diagramis not to act simply as a record of absolute importance and performance values, but to discriminate among attributes as