Table 5.18 Children’s responses on the sentence			generation task for both post tests
	Immediate post test_____		_____Delayed post test_____________
	%	_______n_______	O/ /O		n____________
Don’t know Idiosyncratic properties Perceptual properties Functional properties Semantic properties______	33.3 6 10.9 41.7 8.1	128 23 42 160 31	27.1 6.3 8.8 53.4 4.4		104 24 34 205 17_____________
N of responses___________		384____________		384

Children’s given sentences were mainly focused on the ‘functional properties'” of the items
for both post tests. Statistical analysis revealed that the children provided significantly more
“functional” than “perceptual” (Immediate; Wilcoxon; Z=6.7, p<.0000; delayed: Wilcoxon:
Z=4.9, p<.0000) and “semantic” properties (immediate: Wilcoxon: Z=7.3, p<.' ^,'00; delayed:
Z=9.08, p<.0000). Focus on “perceptual” properties was another popular choice for both post
tests. Significant differences were found during the delayed post test, where they provided
significantly more “perceptual” than “semantic” properties (Wilcoxon: Z=8.09, p<.0000).
Lastly, fewer children made up sentences which focused on the “semantic” properties of the
target items for both post tests.

Whether the same pattern was evident within each age group and each linguistic condition
was also investigated. The distribution of responses separately by age and linguistic condition
is presented in Appendices 5.27 and 5.28 respectively. The table in Appendix 5.24 shows
that the older children provide fewer “don ^,t know” responses than the younger ones for both
post tests. Furtliermore, the 4- and 5- year olds focus mainly on the “functional” properties
of the items and then on the “perceptual” properties. The 6 year-olds focus on the
“functional” properties but they also start focusing on the “semantic” properties.

The table in Appendix 5.25 shows that children from all the linguistic conditions provided
more “functional” than other properties. “Perceptual” properties were also provided less
frequently. Particularly, from the responses given, it was found that the children in the
Inference condition provided significantly more “functional” than “perceptual" (immediate:
Wilcoxon: Z= 4.1, p<.0000; Delayed: Wilcoxon: Z=3.4, p<.0005) and “semantic” properties
(Immediate: Wilcoxon: Z= 4.5, p<.0000; Delayed: Wilcoxon: Z=5.02. p<.0000).
Furthermore, they provided significantly more “functional” properties than "don’t know”
responses (Wilcoxon: Z=2.1, p<.05). In the delayed post test they provided significantly
more “perceptual” than “semantic” properties (Wilcoxon: Z=4.04, p<.0005j.

165

More intriguing information

1. Improving behaviour classification consistency: a technique from biological taxonomy
2. The name is absent
3. The name is absent
4. A Pure Test for the Elasticity of Yield Spreads
5. Return Predictability and Stock Market Crashes in a Simple Rational Expectations Model
6. Stakeholder Activism, Managerial Entrenchment, and the Congruence of Interests between Shareholders and Stakeholders
7. Discourse Patterns in First Language Use at Hcme and Second Language Learning at School: an Ethnographic Approach
8. NVESTIGATING LEXICAL ACQUISITION PATTERNS: CONTEXT AND COGNITION
9. A multistate demographic model for firms in the province of Gelderland
10. The name is absent