Overall Findings

Table 4-1 illustrates the overall mean scores of our child groups (3-, 4-, 5-, 6-year-old) and our adult group on the five meanings of hui:

Table 4-1 Overall Mean Scores of Five Meanings of Hui

Dynamic Deontic Epistemic Generic Futurity

M SD M SD M SD M SD M SD p-value

Child groups 0.7 0.21 0.67 0.18 0.56 0.17 0.65 0.17 0.55 0.16 <2e-16***

Adult group 0.89 0.16 0.75 0.13 0.9 0.16 0.97 0.09 0.94 0.11 1.25e-05***

Our child groups scored higher on the dynamic hui (M=0.7) and the deontic hui (M=0.67) than on the epistemic hui (M=0.56) and on the generic hui (M=0.65). The lowest score was found on the futurity hui (M=0.55). The p-value comparison showed that the overall scores of the five meanings of hui by the child groups were significant (p < 2e-16). With regard to the adult group, the mean scores of the dynamic hui, the epistemic hui, the generic hui and the futurity hui were high and quite close (M=0.89, 0.90, 0.97, 0.94, respectively). Nevertheless, the adults’ score on the deontic hui was lower (M =0.75) than those on the other four meanings of hui. The p-value comparison showed that the adults’ overall interpretations of the five meanings of hui were also significant (p < 1.25e-05).

Table 4-2 shows that after the One-way ANOVA test, the p-values for some types of hui within each group were significant:

51

Table 4-2 Overall P-values for the within-group between-type Differences Dynamic Deontic Epistemic Generic Futurity Child

groups

F 3.777 0.780 0.486 0.214 5.638

p 0.015** 0.509 0.692 0.885 0.001***

Adult group

F 3.249 0.636 29.939 0.094 0.937

p 0.09303 0.438 6.921e-15*** 0.7627 0.3494

As can be seen in Table 4-2, the child groups interpreted the dynamic hui (p = 0.015) and the futurity hui (p = 0.001) significantly differently from the other three types of hui (i.e., deontic, epistemic, and generic). The adult group was found to interpret only the epistemic hui (p = 6.921e-15) differently from the other four types of hui at a significant level. This finding shows that the interpretations of different types

of hui for both subject groups indeed varied.

Figure 4-1 illustrates the overall mean scores and standard deviations of the dynamic, deontic, epistemic, generic and futurity hui’s by each child group and the adult group:

52 Figure 4-1 Subjects’ Overall Interpretations of Different Meanings of Hui

Concerning the mean scores, what stands out from the graph is that the 3- and 4-year-old groups had a similar tendency (i.e., scores and lines), whereas the results of the 5- and 6-year-olds were different from those of the other child groups. What is more, the adult group scored the highest on the five meanings of hui than the child groups.

With respect to the interpretation of the 3- and 4-year-old groups, the dynamic hui got 0.61 and 0.63, respectively, and there was a decrease in the deontic hui (3-year-olds, M=0.56; 4-year-olds, M=0.60). These two child groups’ interpretation of the epistemic hui (3-year-olds, M=0.53; 4-year-olds, M=0.61) and of the generic hui (3-year-olds,

0

53

M=0.64; 4-year-olds, M=0.67) gradually improved, but their interpretation of the futurity hui (M = 0.45, 0.5) did not. If we look at the interpretation of the 5- and 6-year-old groups, we may find that their mean scores dropped from the dynamic hui (5-year-olds, M=0.77; (5-year-olds, M=0.80) to the deontic hui (5-year-(5-year-olds, M=0.64; 6-year-olds, M=0.66) and the epistemic hui (5-year-6-year-olds, M=0.56; 6-year-6-year-olds, M=0.54)

radically. Their mean scores rose gradually on the generic hui (5-year-olds, M=0.66; 6-year-olds, M=0.63) and on the futurity hui (5-6-year-olds, M=0.61; 6-6-year-olds, M=0.67).

As for the adult group, their scores declined from the dynamic hui (M = 0.89) to the Deontic Hui (M = 0.75) dramatically, and then rose to 0.91, 0.97, and 0.94 on the epistemic hui, the generic hui and the futurity hui, respectively.

Let us turn to the within-type between-group comparisons of hui, as shown in Table 4-3:

54

Table 4-3 Overall P-values for the within-type between-group Differences of Hui Dynamic Deontic Epistemic Generic Futurity interpreted it similarly (p = 0.816). It is also clear that the interpretation of the 5- and 6-year-olds were similar (p = 0.64287) and adult-like (5 vs. Adult, p = 0.06648; 6 vs.

Adult, p = 0.16661). However, it was found that the younger children’s interpretation of the dynamic hui was significantly different from those of the older child groups and the adult group (3 vs. 5, p = 0.022; 3 vs. 6, p = 0.00661; 3 vs. Adult, p = 7.5e-05; 4 vs.

5, p = 0.03954; 4 vs. 6, p = 0.01246; 4 vs. Adult, p = 0.00017). Second, regarding the deontic hui, it can be seen that the younger child groups and the older child groups interpreted it in a similar way (3 vs. 4, p = 0.7635; 3 vs. 5, p = 0.9896; 3 vs. 6, p = 0.6205; 4 vs. 5, p = 0.9542; 4 vs. 6, p = 0.9993; 5 vs. 6, p = 0.9993). However, they interpreted this hui differently from the adults (3 vs. Adult, p = 0.0088; 4 vs. Adult, p = 0.0135). An interesting point is that the 5- and 6-year-olds’ interpretation of the deontic

55

hui were adult-like (5 vs. Adult, p = 0.8791; 6 vs. Adult, p = 0.1083). Third, concerning

the epistemic hui, the 3- and 4-year-olds’ mean scores were (p = 0.1945), and so were the mean scores of the 5- and 6-year-olds (p = 0.6021). If we compare the interpretation of this hui by the younger child groups and the older child groups, the p-values were found insignificant on the epistemic hui (3 vs. 5, p = 0.6021; 3 vs. 6, p = 1.0000; 4 vs.

5, p = 0.4346; 4 vs. 6, p = 0.1945). This is to say that the interpretation of the younger child groups and the older child groups were compatible. However, the interpretation of the four child groups were incompatible with the adults’ (3 vs. Adult, p = 2.0e-08; 4 vs. Adult, p = 4.0e-06; 5 vs. Adult, p = 1.7e-07; 6 vs. Adult, p = 2.0e-08). Fourth, as for the generic hui, our four child groups’ interpretation were similar to those of the epistemic hui. For instance, both the younger child groups and the older child groups interpreted this hui alike (3 vs. 4, p = 0.5832; 3 vs. 5, p = 0.7153; 3 vs. 6, p = 0.9021; 4 vs. 5, p =0.7837; 4 vs. 6, p = 0.4111; 5 vs. 6, p = 0.5832). However, the four child groups and the adult group interpreted the generic hui differently (3 vs. Adult, p = 1.6e-07; 4 vs. Adult, p = 1.5e-06; 5 vs. Adult, p = 4.8e-1.6e-07; 6 vs. Adult, p = 5.0e-08). Finally, regarding the futurity hui, the younger child groups interpreted it similarly (p = 0.75077), while the older children’s interpretation was quite the same (p = 0.20619).

Nevertheless, the interpretation of the futurity hui between the younger children and the older child groups were incompatible (3 vs. 5, p = 0.00534; 3 vs. 6, p = 8.9e-05; 4 vs.

56

5, p = 0.01280;). Moreover, both the younger child groups and the older child groups’

interpretation were different, compared with the adult group’s (3 vs. Adult, p = 1.3e-14;

4 vs. Adult, p = 0.00017; 5 vs. Adult, p = 3.5e-09; 6 vs. Adult, p = 7.0e-07).

According to Table 4-3, it is noticeable that the 3- and 4-year-old groups interpreted the five meanings of hui alike. The 5- and 6-year-old groups had a similar

interpretation of the Chinese modal hui, and their interpretation was more adult-like than the younger child groups’.