To what extent does children’s performance on the contrast task differ by linguistic
condition ?
If all children from each linguistic condition were correct on the two target words, the
maximum total score would be 96 each linguistic condition in each post test. Figure 5.17
below shows children’s performance on the contrast task by linguistic condition for both post
tests. As the figure shows, children in the Lexical contrast condition performed better on the
contrast task than the children in the other conditions for both post tests. The differences were
significant for the immediate (Kruskal-Wallis, 1 Way ANOVA: X2= 120.8, df=3, p<.0000)
and the delayed post test (Kruskal-Wallis, 1 Way ANOVA: X2= 46.8, df=3, p<.0000).
Figure 5.17 Total number of correct responses on the contrast task by linguistic
condition for both post tests
I I Immediateposttest
Q Delayedposttest
Linguistic condition differences within each age group were also investigated. Children’s
performance on the contrast task within each age group differed by linguistic condition. The
4 year-olds in the Contrast condition performed significantly better on the contrast task than
the 4 year-olds in the other conditions during the immediate post test (Kruskal-Wallis, 1 Way
AN0VA: X2= 25.6, df=3, p<.0000). A trend for significance for the same pattern was found
for the delayed post test. Significant differences were also found for the 5 year-olds
(immediate: Kruskal-Wallis, 1 Way ANOVA: X2= 44.3, df=3, p<.0000; delayed: Kruskal-
Wallis, 1 Way ANOVA: X2= 19.5, df=3, p<.0005) and the 6 year-olds (immediate: Kruskal-
Wallis, 1 Way ANOVA: X2= 52.03, df=3, p<.0000; delayed: Kruskal-Wallis, 1 Way
ANOVA: X2= 23.6, df=3, p<.0000).
Post hoc analyses indicated that the 4 year-olds in the Contrast condition performed
significantly better on the contrast task than the children in the Inference (Mann-Whitney:
Z=3.6, df=3, p<.0005), Analogy (Mann-Whitney: Z=3.1, p<.005) and Contrast conditions