The third research question discusses whether frequency adverbs modifying the subtypes of telic events (i.e., ET1-1, ET1-2 and ET1-3) displayed a different degree of
80
acceptability. ET1-1 is a resultative construction; ET1-2 is a verb-particle construction;
ET1-3 is a dative/double object construction. Section 4.3.1 reports the subjects’ overall acceptability ratings of frequency adverbs when they modified the three subtypes of events. Section 4.3.2 discusses whether it is necessary to differentiate the subtypes of telic events with regard to their co-occurrence with frequency adverbs.
4.3.1 Overall Findings
Table 4-9 shows the subjects’ acceptability ratings for frequency adverbs when they modified the subtypes of telic events. According to Table 4-9, frequency adverbs modifying these subtypes displayed a significantly different degree of the acceptability (F (2,1197) = 8.68, p < .001). Among them, ET1-2 (M = 1.47) received the highest rating, followed by ET1-3 (M = 1.34), and ET1-1 (M = 1.23) obtained the lowest rating of acceptability.
Table 4-9 Subjects’ Overall Acceptability Ratings for Frequency Adverbs Modifying the Three Subtypes of Telic Events
Type M SD F p-value
ET1-1 1.23 0.87 8.68 0.000181 ***
ET1-2 1.47 0.77
ET1-3 1.34 0.86
Note: *** indicates that the p-value is smaller than .001.
Figure 4-4 presents each group’s acceptability ratings for frequency adverbs when
81
they modified the three subtypes of telic events.
Figure 4-4 Each Group’s Acceptability Ratings for Frequency Adverbs Modifying the Three Subtypes of Telic Events
In Figure 4-4, it can be seen that frequency adverbs modifying ET1-2 received the highest ratings of the acceptability among the three subtypes for each group except for the kindergartners and Grade 6. Actually, the acceptability rating Grade 6 gave to ET1-2 was rather high, which demonstrates that they actually accepted frequency adverbs modifying ET1-2. ET1-1 and ET1-3, on the other hand, were generally rated lower than ET1-2. Though the kindergartners gave ET1-3 a higher rating than the adult controls did, the difference was not significant. With regard to the within-group differences, Table 4-10 presents the acceptability of frequency adverbs when they modified the three
Kindergartner Grade 2 Grade 4 Grade 6 Control
ET1-1 0.46 1.1 1.43 1.54 1.61
ET1-1 SD 0.75 0.88 0.74 0.78 0.68
ET1-2 1.25 1.56 1.49 1.44 1.63
ET1-2 SD 0.89 0.65 0.71 0.82 0.68
ET1-3 1.53 1.29 1.24 1.25 1.4
ET1-3 SD 0.78 0.84 0.83 0.95 0.85
0
82
subtypes.
Table 4-10 The p-values for the within-group Differences in the Acceptability of Frequency Adverbs Modifying the Three Subtypes of Telic Events
Kindergartner Grade 2 Grade 4 Grade 6 Control
F 37.24 6.78 2.24 2.34 2.32
p .000*** .001** .10 .10 .10
Note: ** indicates that the p-value is smaller than .01 and *** is smaller than .001.
As indicated in Table 4-10, it turns out that frequency adverbs modifying the three subtypes of telic events displayed a significant difference only for the younger children (Kindergartner: F (2,237) = 37.24, p < .001; Grade 2: F (2, 237) = 6.78, p < .01). The other three groups, Grade 4 , Grade 6 and the adult controls, accepted frequency adverbs modifying the three subtypes in a similar degree (Grade 4: F (2, 237) = 2.24, p > .05;
Grade 6: F (2,237) = 2.34, p > .05; Control: F (2,237) = 2.32, p > .05).
As for the interaction effects, two-way ANOVA indicated that there was a significant difference for groups (F (4,1185) = 11.15, p < .001), and event types (F (2,1185) = 9.54, p < .001). Also, the interaction effects beween groups and event types were significant (F (8,1185) = 10.69, p < .001), as provided in Table 4-11.
83
Table 4-11 The Interaction Effects in the Acceptability of Frequency Adverbs Modifying the Three Subtypes of Telic Events
F p
Group 11.15 6.97e-09 ***
Event Type 9.54 7.80e-05 ***
Interaction 10.69 1.34e-14 ***
Note: *** indicates that the p-value is smaller than .001.
Since there were interaction effects, the Tukey test was administered. Table 4-12 shows that there was a significant difference between Grade 2 and the kindergartners (p < .05), Grade 4 and the kindergartners (p < .001), Grade 6 and the kindergartners (p < .001), the adult controls and the kindergartners (p < .001) and the controls and Grade 2 (p
< .05). In short, all the age groups accepted frequency adverbs modifying the three subtypes of telic events significantly differently from the kindergartners, and the controls also accepted frequency adverbs modifying the three subtypes significantly differently from Grade 2. Based on the findings, it can be inferred that by Grade 4, the children had accepted frequency adverbs when they modified the three subtypes in a similar degree.
84
Table 4-12 The p-values for the between-group Differences in the Acceptability of Frequency Adverbs Modifying the Three Subtypes of Telic Events
Grade 2 Grade 4 Grade 6 Control
Kindergartner ** *** *** ***
Grade 2 - .89 .71 *
Grade 4 - - 1.00 .17
Grade 6 - - - .32
Note: * indicates that the p-value is smaller than .05, ** is smaller than .01 and *** is smaller than .001.
The differences of between-event types were also examined. Frequency adverbs modifying ET1-2 differed from those modifying either ET1-1 (p < .001) or ET1-3 (p
< .05) in their degree of acceptability. ET1-1 and ET1-3 displayed a similar degree of acceptability (p > .05), indicating that ET1-1 and ET1-3 were considered as the same type of events for the subjects.
With regard to the interaction effects of Groups and Event types, the results indicate that frequency adverbs modifying ET1-1 had a significant interaction effect with the other two subtypes for the kindergartners. For ET1-1 versus ET1-2,
significantly differently from the kindergartners, all the age groups accepted frequency adverbs when they modified the two subtypes. Similarly, for ET1-1 versus ET1-3,
significantly differently from the kindergartners, all the age groups accepted frequency adverbs when they modified the two subtypes. Apparently, the kindergartners had a
significantly different acceptability rating of frequency adverbs modifying ET1-1
85
compared with the other groups. For ET1-1 versus ET1-2, Grade 2, significantly differently from the controls, also accepted frequency adverbs when they modified the two subtypes of events, but not for ET1-1 versus ET1-3.
4.3.2 General Discussion
Previous studies (e.g., Kayne 1984 and 1985, Hoekstra 1988, Snyder 1995) have identified three telicity constructions. We are interested in exploring whether frequency adverbs modifying these three constructions display a different degree of acceptability.
According to the findings, the acceptability ratings for frequency adverbs modifying the three subtypes of telic events were different. A closer examination revealed that the significance was mainly caused by the younger children’s (the kindergartners and Grade 2) responses. For them, they were still not able to figure out the same function for the three constructions, and this explained their different acceptability of frequency adverbs modifying the three subtypes of telic events. For the older children (Grades 4 and 6) and the adult controls, the three constructions all denoted telic events (i.e., with the same function but in different forms) and thus received similarly higher ratings of acceptability when being modified by frequency adverbs.
In Section 4.2, it was found that all the age groups, except for the kindergartners, accepted frequency adverbs modifying telic events significantly differently from atelic
86
events. In other words, Grade 2 had already acquired the concept of telicity and known that frequency adverbs could not modify all events. However, according to the findings presented in this section, although they had known that some events could not be modified by frequency adverbs, they had not yet fully acquired adult-like acceptability of frequency adverbs modifying all subtypes of telic events. From the findings of Section 4.2 and Section 4.3, it can be inferred that the children was dimly aware of telicity at first (the kindergartners), and when they grew older (Grade 2), they became aware that frequency adverbs were limited to modify all events. Finally, they accepted frequency adverbs modifying the three subtypes of telic events in a similar degree (Grade 4).