Microsoft Office Tutorials and References

In Depth Information

Table 6.3
Results of z-test for training of customer service representatives

The results of our analysis is shown in Table 6.3. Note that a z-statistic of approx-

imately –2.74 has been calculated. We reject the null hypothesis if the test statistic

(z) is either:

z>

=

critical two-tail value (1.959962787)

......

see cell B12

or

z<

=

- critical two-tail value (-1.959962787).

Note that we have two rules for rejection since our test does not assume that

one of the sample means is larger or smaller than the other. Alternatively, we can

compare the p-value

=

α =

0.006191794 (in cell B11) to

0.05 and reject if the p-

value is <

. In this case the critical values (and the p-value) suggest that we
reject

the null hypothesis that the samples means are the same; that is, we have found

evidence that the EB training program at SC has indeed had a signiﬁcant effect on

scores for the customer service representatives. EB is elated with this news since

it suggests that the training does indeed make a difference, at least at the

= α

0.5

level of signiﬁcance. This last comment is recognition that it is still possible, in

spite of the current results, that our samples have led to the rejection of a true null

hypothesis. If greater assurance is required, then run the test with a smaller

α =

,for

example 0.01. The results will be the same since a p-value 0.006191794 is less than

0.01.

α

Search JabSto ::

Custom Search