Analysis and Results
A single factor ANOVA shows a significant difference between
group variance for both metrics, product quality score, and
quality issue rate (see Exhibit 4). Since the p-values (p=0.000) are
less than the selected level of significance (α= 0.05), we conclude
that at least one mean for each quality performance metric is
significantly different.
An analysis of means was performed to determine where the
differences lie. Corresponding ANOM charts are shown in Exhibit
5. The horizontal center lines indicate the overall mean. The top
and bottom horizontal lines represent the upper and lower critical
limits respectively. Since the pretest sample sets for company
B have smaller sample sizes than the other sample groups, the
acceptance limits for company B’s pretest datasets are wider. The
only mean product quality score that falls above the critical limit
is that of company B (B posttest), demonstrating that the site
with a KM/QM strategy is significantly different than the three
other sites in a positive way. To the contrary, company B prior
to KM/QM strategy implementation (B pretest) shows the mean
product quality score below the overall average in Exhibit 5,
indicating an inferior product quality. The mean product quality
score for company A falls below the critical limit and is, therefore,
significantly different in a negative way than the mean product
quality score for the other three sites. The reason for site A’s
inferior mean product quality score is not part of this research. In
practical terms, the reader may inquire about the implication of
having a mean product quality score of 6.77 (company B posttest)
versus 6.60 (company B pretest).
To illustrate the difference with an arbitrarily selected
product, such as a household appliance, the higher mean product
quality score of 6.77 indicates that the customer can expect no
functional defects upon installation of the product. The appliance
should work as intended; however, the population with this score
can still include a small number of appliances with minor defects
that are not always noticed by the customer and do not affect the
function of the appliance. Such minor defects can include scratch
marks or small dents on surfaces not visible to the customer at
first sight. The lower mean product quality score of 6.60 is an
indication that the population can also have a small number of
appliances that do not work upon installation and may require
repair or a product exchange to resolve the quality problem for
the customer, e.g., a washing machine spin cycle does not work
upon installation, a washing machine does not properly drain
water despite correct installation, et