Microsoft Office Tutorials and References
In Depth Information
Table 6.4 Results t-test scores of prisoner and non-prisoner customer service representatives
6.5.6 How do Scores Differ Among Prisoners of SC and Texas
Before Special Training?
A third and related question of interest is whether the prisoners in SC and TX
have mean scores (before training) that are signiﬁcantly different. To test this ques-
tion, we can compare the two samples of the prisoners, TX and SC, using the SC
prisoners’ scores prior to special training. To include EB trained prisoners would
be an unfair comparison, given that the special training may have an effect on
the scores. Table 6.6 shows the results of the analysis. Again, we perform the t-
Test: two-samples unequal variances and ﬁnd a t-statistic of 0.614666361 (cell B9).
Given that the two-tail critical value is 1.994435479 (cell B13), the calculated t-
statistic is not sufﬁciently extreme to reject the null hypothesis that there is no
difference in mean scores for the prisoners of TX and SC. Additionally, the p-value,
0.540767979, is much larger than the
of 0.05. This is not an unexpected outcome
given how similar the mean scores, 79.083 and 80.361, were for prisoners in both
Finally, we began the example with a question that focused on the viability of
special training—Is there a signiﬁcant difference in scores after special training?