1974; Crosby, 1979; Juran and Gryna, 1980; Deming, 1986; Surak and
McAnelly, 1992; Bounds et al., 1994), but few for shelf life (Shewfelt,
1986), or acceptability (Land, 1988). In general, chemists tend to view
quality as a function of qualitative and quantitative differences of indi-
vidual compounds present in a food product (Fennema, 1996), engineers
as a bundle of characteristics that can be measured with accuracy and
precision (Taoukis and Labuza, 1996; Hendrickx et al., 1994), posthar-
vest physiologists as a condition to be attained (Kays, 1991), and sen-
sory specialists as attributes that can be described by trained panels (Lyon
et al., 1993). Most of these interpretations view quality as properties in-
trinsic to the product itself. Adoption of a unified language for these
three terms is essential in comparing subsequent studies.
Quantitative assessment of consumer acceptabilityis necessary for
the adoption of a consumer-oriented approach to quality improvement.
None of the groups of scientists listed above has been actively involved
in developing new means of assessing acceptability. All have been more
concerned with the internal validity of their research than external va-
lidity. Quality is viewed as a property that decreases over time, which
at best can be maintained but never improved.
Chemists trace changes of individual compounds such as pigments,
sugars, acids, flavor volatiles (Shewfelt, 1994) or vitamins (Clydesdale
et al., 1991). Biochemists trace changes in enzyme activity or gene ex-
pression (Tucker and Grierson, 1987). Engineers and postharvest phys-
iologists trace changes in quality characteristics such as flavor, color,
texture, or loss in vitamins (Kays, 1991), with engineers particularly con-
cerned with accuracy and precision of results and kinetic modeling of
the data (Taoukis and Labuza, 1996). Sensory specialists measure
changes in descriptive notes (Lyon et al., 1993; Lawless and Heymann,
1998). Consumer panels are used to assess overall acceptability (Mosko-
witz, 1994; Lawless and Heymann, 1998). Scales are employed to ex-
press acceptance on a Hedonic scale, frequently 9 points (Peryam and
Pilgrim, 1957) or on a 100-point basis (Moskowitz, 1994), although some
studies tend to use descriptive panels to assess this trait with dubious
validity.
Chemists, engineers and postharvest physiologists tend to shy away
from any contact with consumers, with the data considered completely
unreliable. A literal interpretation of Land’s (1988) definition of ac-
ceptability (“the level of continued purchase or consumption by a spec-
ified population,” p. 476) suggests that measurement is not possible
outside of the market. Obviously, some premarket predictor of accept-
Quality Principles 147