constructs. The suggested sequence has worked
well in several instances in producing measures with
desirable psychometric properties (see Churchill et
al., 1974, for one example). Some readers will undoubtedly
disagree with the suggested process or with
the omission of their favorite reliability or validity
coefficient. The following discussion, which details
both the steps and their rationale, shows that some
of these measures should indeed be set aside because
there are better alternatives or, if they are used, that
they should at least be interpreted with the proper
awareness of their shortcomings.
The process suggested is only applicable to multiitem
measures. This deficiency is not as serious as
it might appear. Multi-item measures have much to
recommend them. First, individual items usually have
considerable uniqueness or specificity in that each
item tends to have only a low correlation with the
attribute being measured and tends to relate to other
attributes as well. Second, single items tend to categorize
people into a relatively small number of groups.
For example, a seven-step rating scale can at most
distinguish between seven levels of an attribute. Third,
individual items typically have considerable measurement
error; they produce unreliable responses in the
sense that the same scale position is unlikely to be
checked in successive administrations of an instrument.
All three of these measurement difficulties can be
diminished with multi-item measures: (1) the specificity
of items can be averaged out when they are
combined, (2) by combining items, one can make
relatively fine distinctions among people, and (3) the
reliability tends to increase and measurement error
decreases as the number of items in a combination
increases.
The folly of using single-item measures is illustrated
by a question posed by Jacoby (1978, p. 93):
How comfortable would we feel having our intelligence
assessed on the basis of our response to a single
question?" Yet that's exactly what we do in consumer
research.... The literature reveals hundreds of instances
in which responses to a single question suffice
to establish the person's level on the variable of interest
and then serves as the basis for extensive analysis
and entire articles.
. . . Given the complexity of our subject matter, what
makes us think we can use responses to single items
(or even to two or three items) as measures of these
concepts, then relate these scores to a host of other
variables, arrive at conclusions based on such an
investigation, and get away calling what we have done
"quality research?"
In sum, marketers are much better served with
multi-item than single-item measures of their constructs,
and they should take the time to develop them.
This conclusion is particularly true for those investigating
behavioral relationships from a fundamental