knowledge" questions was found to differ significantly during post test 2 (Kruskal-Wallis:
X² = 13.07, df=2, p<.005) and post test 3 (Kruskal-Wallis: X² = 65.2, df= 4, p<.0000).

Post hoc analysis revealed that the Lexical contrast group performed significantly better than
the Ostensive definition group during post test 2 (Wilcoxon: Z=3.2, p<.005). Moreover, the
Definition group performed significantly better than the Ostensive definition group during
post test 2 (Wilcoxon: Z=2.7, p<.05). During post test 3 the Lexical contrast group performed
significantly better than the Control (Wilcoxon: Z=5.2, p<.0000), the Phonological control
(Wilcoxon: Z=5.3, p<.0000) and the Ostensive definition group (Wilcoxon: Z=4.3, p<.0000).
The Definition group also performed significantly better than the Control (Wilcoxon: Z=5.5,
p<.0000) the Phonological control (WilcoxomZ=5.7, p<.0000) and the Ostensive definition
group (Wilcoxon: Z=4.9, p<.0000). Diagram 7.6 presents the significant differences between
the groups.

Diagram 7.6 Significant group differences in the “world knowledge questions” by group

Does children's performance on the short questions improve with increased exposure to the
lexical items?

Categorisation questions

Children’s performance on the categorisation questions improved significantly over time
(Friedman Two Way Anova: X² = 31.1, df=2, p<.05). The same pattern was evident for the
Ostensive definition group (Friedman Two Way Anova: X² = 7.7, df=2, p<.05), and the
Definition group (Friedman Two Way Anova: X² = 31.4, df=2, p<.0000).

229

<<   226   227   228   229   230   231   232   >>

More intriguing information