Factors intrinsic to the productmust be separated from those intrin-
sic to consumers.One of the severest criticisms of the use of consumer
data is that the consumer is not a monolith but a confederation of many
different tastes and preferences. Current sensory practice tends to pro-
vide answers for the average consumer without providing an under-
standing of the population. Segmentation of the population is a way to
overcome this problem (Land, 1988). Segmentation by demographic
variables has been rejected by Moskowitz (1994), who prefers to seg-
ment by quality preferences. For this approach to work, it must be as-
sumed that there is a finite (preferably four or less) number of distinct
quality-preference segments for a given product. In addition, it should
be recognized that some “factors that drive acceptability” (p. 107), are
in the realm of sensory characteristics intrinsic to the food, while oth-
ers are “image characteristics” (p. 447), which are related to marketing
and other intangibles that are beyond the capability of most sensory lab-
oratories (Moskowitz, 1994, p. 447). Segmentation of the consuming
population can be achieved by evaluating for overall acceptability using
a 3-point scale. Acceptability can be expressed as a probability distri-
bution for each segment (Shewfelt et al., 1997).
In summary, consumer-based quality evaluation is possible. Such
studies require the use of a common set of terms, must be designed to
be both externally and internally valid, and must differentiate between
characteristics that are based on defects and those that are based on ex-
cellence. Confining consumers to narrow choices on acceptability alone
and expressing the data in the form of a probability distribution within
a market segment provides an assessment tool of acceptability that is
quantitative. When combined with sensory descriptive analysis, this ap-
proach (known as quality enhancement) can relate product-based qual-
ity characteristics to consumer acceptability.
QUALITY ENHANCEMENT—THE MODEL
Quality enhancement (QE) is a theoretical model that has been de-
scribed previously (Shewfelt, 1994; Shewfelt and Phillips, 1996; Shew-
felt et al., 1997; Malundo et al., 1997), but there have been only a few
application studies. The model was developed on the premise that it is
the consumer, not the researcher or the commercial distributor, who de-
fines quality. It seeks to address concerns described in the previous sec-
tion, particularly with respect to achieving both internal and external
Quality Enhancement—The Model 149