In order to examine the differences in metaphorical language use of the experimental group and the control group at the beginning of the metaphor awareness treatment, an independent samples t-test was conducted for the data gathered from the participants’ recent writing test. The statistical analysis indicated that there was not a significant difference in the metaphorical language use of experiment group (M = 1.59; SD = .70) and control (M = 1.44; SD = 1.00); t (43) = .58, p = .56. Levene’s test for equality of variances was found to assume the homogeneity of variance F (1, 43) = 1.70, p = .19. The results of this test showed the participants of the both groups came from the same population.
The analysis of the main data shows that …show more content…
The application of Friedman test to the experimental group data showed statistically significant changes in the distribution of metaphorical language use over the three tests x2 = 23.73, df = 2, p = .00. Similarly, the repeated measures comparison of the control group data using Friedman Test showed statistically significant changes in the distribution of metaphorical language use over the three tests x2 = 14.19, df = 2, p = .00. In order to determine where the specific changes lie in the data, post-hoc analysis with Wilcoxon Signed Ranks Test was conducted.
For experimental group data, a pairwise application of the Wilcoxon Signed Ranks Test with Bonferroni corrected levels of observed significance indicated that the median pretest scores, Mdn = 5.13, were statistically significantly higher than the median posttest scores, Mdn = 3.57 (Z = -4.04, p = .00). Similarly, the median delayed posttest scores, Mdn = 4.67, were statistically significantly higher than the median posttest scores, Mdn = 3.57 (Z = -3.55, p = .00), but the test showed no significant change between the median of pretest and posttest scores (Z = -1.00, p =