Using SPSS to Understand Research and Data Analysis.

10.3 Interpreting the Output

Several statistics are presented in the first table, Paired Samples Statistics (Figure 10.3). The most relevant for our purposes are the two means. Examination of these means suggests that the average leadership performance score was higher after the workshop. However, the t-test will determine whether or not this difference is real or due to chance.

The paired-samples t-test procedure automatically computes the correlation between the two sets of scores (Figure 10.4). As discussed in the previous section, we can see that there is, indeed, a significant positive correlation between the scores before and after the workshop (r = .29, p < .001), indicating that those who scored high on social skills before the workshop also tended to score high on social skills after the workshop.

The next table presents the results of the t-test (Figure 10.5). The first column shows the actual difference between the two means (-.40), which is the numerator of the t-test formula. This is a negative number because the mean after the workshop (a larger value) is subtracted from that before the workshop (a smaller value).

The second column presents the standard deviation of the difference scores (2.43). The third column shows the standard error of the mean (.16), which is the denominator of the t-test formula. The 95% Confidence Interval presents the lower and upper limits of the mean difference (i.e., we can be 95% confident that the difference between the means falls somewhere between -.71 and -.08)

The most important columns for our purposes are the one presenting the obtained t-value (-2.45) and its probability, or significance (.02). Since the probability is less than .05, we reject the null hypothesis of no difference in favor ob the alternative hypothesis that the difference is real. To report the results in APA format, we would state the following:

A paired-samples t-test revealed a significant differences in the social skills scores before and after the workshop, t (227) = -2.45, p < .02. This indicates that the mean social skills score after the workshop (M = 5.86) was significantly higher than the mean before the workshop (M = 5.46).

This analysis provides evidence that the leadership training workshop was, indeed, effective in leading to increased social skills in leadership situtations. This type of research is essential to evaluate the effectiveness of any new program or procedure implemented, and as such, it is frequently called program evaluation research. In the exercise at the end of the chapter you will be asked to determine whether or not task skills also changed significantly after the workshop. You might also be curious to determine whether or not leader performance changed as well.