This chapter presents the results of the study in four main sections. Section One discusses the demographic information of the participants, and some descriptive results related to the six measuring instruments employed in the formal study. Section Two presents results of the correlation analyses among all the variables under investigation.
Section Three addresses the quantitative results regarding the explanatory power of English proficiency, creativity, and motivation for creativity in EFL learners’ abilities to interpret and produce creative metaphors in English. Section Four demonstrates the qualitative analysis in order to investigate in greater detail how creativity attitude may cast an impact on EFL learners’ creative metaphoric competences in English.
Descriptive Data
This section discusses the demographic information of the participants in the formal study. Descriptive results related to the six instruments adopted in this study are also reported.
Demographic Information of the Participants
The formal study was conducted on a group of 215 freshmen from six intact classes across the fall and spring semesters. All of the participants were required to complete the six quantitative measures employed in this study. Nevertheless, because these instruments were administered to the participant population separately, a few participants’ absence from class unavoidably led to some missing data. The final sample consisted of 169 participants, who all met two criteria: They were not foreign or overseas students, and they had completed the Recognition Task of Creative Metaphors and the Production Task of Creative Metaphors. Therefore, only the data collected from these 169 participants were analyzed to address the relationships of creative metaphoric competences to English
proficiency, creativity and motivation for creativity.
A background survey embedded on the Questionnaire on Motivation Toward Verbal Creativity in English (QMVC) indicated the participants’ gender, English learning experiences, major field of study, previous experiences of living in any English-speaking countries, and self-perceived proficiency of English and creativity. As shown in Table 21, the data sets were collected from 63 males and 106 females. On average, these
participants had learned English for nearly ten years (M = 9.55, SD = 3.17) at the time of the study. They perceived that their overall English proficiency skills and creativity were moderate (English: M = 3.63 on a 7-point scale, SD = 1.20; Creativity: M = 3.73 on a 7-point scale, SD = 1.28).
Table 21. Demographical Information of the Participants in the Formal Study
Class A B C D E F Total
Number of Students
Total 28 22 24 29 31 35 169
Male 17 14 10 10 4 8 63
Female 11 8 14 19 27 27 106
Average Years of Learning 8.59 7.35 9.57 10.14 10.55 10.20 9.55 Perceived Level of English Ability
M 2.30 2.45 3.70 4.16 4.24 4.37 3.63
SD 1.10 1.05 .93 .72 .69 .73 1.20
Perceived Level of Creativity
M 3.07 3.30 3.78 3.97 4.07 3.97 3.73
SD 1.64 1.34 1.09 1.18 1.10 1.07 1.28
Note. Because five participants did not complete the QMVC, information provided above except for the gender distribution was mainly based on the other 164 participants’ data.
Descriptive Statistics Related to Participants’ Performance on the Measurement Instruments
The following section will present an overview of the participants’ performances in the formal study. As indicated in Chapter Three, the participants came from three
proficiency levels at the target university, including two elementary classes (Classes A and B), one low-intermediate class (Class C), and three high-intermediate classes (Classes D, E and F). For the discussion below, they are divided into two proficiency groups only, with the former three being the lower-achievement (LA) group and the latter being the higher-achievement (HA) group. The differences between these two groups will be presented.
English Reading and Writing Proficiency
The English Reading and Writing Proficiency Tests were adopted to measure participants’ English reading and writing skills. The participants’ performances in the tests demonstrate their general reading proficiency and writing proficiency. First of all, as Table 22 indicates, while 17 students did not participate in the English reading test owing to their absence from class, 152 students completed this test successfully. Of the six participating classes, Classes D, E and F (D: M = 88.15, SD = 9.94; E: M = 93.60, SD = 6.52; F: M = 95.09, SD = 10.21) outperformed Classes A, B and C (A: M = 40.32 , SD = 14.15; B: M = 45.53, SD = 9.32; C: M = 69.71, SD = 12.03). Secondly, as shown in the same table, 163 students completed the English writing test, and the results revealed that Classes D, E, and F (D: M = 68.52, SD = 8.64; E: M = 67.10, SD = 12.96; F: M = 65.45, SD = 15.43) obtained strikingly higher average scores than Classes A, B and C (A: M = 20.38, SD = 22.36; B: M = 22.27, SD = 13.78; C: M = 41.67, SD = 12.39).
Table 22. Results of the English Reading and Writing Proficiency Tests Lower-Achievement Group Higher-Achievement Group
Class A B C Total D E F Total TOTAL
English Reading Test
N 25 17 21 63 26 30 33 89 152
M 40.32 45.53 69.71 51.52 88.15 93.60 95.09 92.56 75.55 SD 14.15 9.32 12.03 17.87 9.94 6.52 10.21 9.41 24.37 English Writing Test
N 26 22 24 72 27 31 33 91 163
M 20.38 22.27 41.67 28.06 68.52 67.10 65.45 66.92 49.75 SD 22.36 13.78 12.39 19.40 8.64 12.96 15.43 12.80 25.12 Overall English Proficiency (z scores)
N 23 17 21 61 26 30 31 87 148
M -1.35 -1.20 -.31 -.95 .63 .73 .73 .70 .02
SD .62 .40 .41 .68 .31 .31 .42 .35 .96
Note. The full score of the English reading test was 120, while that of the English writing test was 100.
Figure 10 demonstrates the mean reading and writing scores of the HA group
(Classes D, E and F) and the LA group (Classes A, B and C). It is straightforward that the HA group obtained much higher English reading and writing scores than the LA group.
Moreover, results of Independent Samples t-tests indicated that the HA group (M = 92.56, SD = 9.41) significantly outperformed the LA group (M = 51.52, SD = 17.87) in their reading scores (df = 86.37, t = -16.67, p = .00). The HA group (M = 66.92, SD = 12.80) also performed better than the LA group (M = 28.06, SD = 19.40) on the English writing test (df = 117.32, t = -14.66, p = .00).
Figure 10. Mean scores of the English reading and writing tests
Furthermore, the English reading and writing scores were transformed into z scores, and further combined and averaged to demonstrate an individual’s overall English proficiency. Results again confirmed that the HA group (M = .70, SD = .35) significantly outperformed the LA group (M = -.95, SD = .68), with the difference reaching the significance level (df = 82.82, t = -17.43, p = .00). These variations are in line with the fact that these classes were recruited from different proficiency levels at the target school.
Creative Thinking Abilities
The Torrance Tests of Creative Thinking (TTCT) were employed to calibrate the participants’ creative thinking abilities in four respects: fluency, flexibility, originality, and general creativity denoting the three abilities combined. As Table 23 shows, Class F obtained the highest scores of fluency (M = 85.67, SD = 12.91), flexibility (M = 85.64, SD = 14.00) and general creativity (M = 88.54, SD = 12.43) among all the participating
classes. However, Class C had the highest originality score (M = 94.32, SD = 10.79), preceding Class F (M = 94.30, SD = 11.75) by only a minute difference.
Table 23. Results of the Torrance Tests of Creative Thinking
Class
Lower-Achievement Group Higher-Achievement Group
TOTAL
A B C Total D E F Total
N 18 18 22 58 27 30 33 90 148
Fluency
M 75.94 75.72 84.55 79.14 80.70 79.63 85.67 82.17 80.98 SD 14.57 7.76 10.50 11.84 13.50 9.00 12.91 12.12 12.06 Flexibility
M 70.00 74.89 84.14 76.88 81.89 79.30 85.64 82.40 80.24 SD 11.62 11.02 15.06 14.02 19.03 13.79 14.00 15.67 15.24 Originality
M 79.06 79.22 94.32 84.90 88.63 82.13 94.30 88.54 87.11 SD 18.60 9.45 10.79 15.10 14.04 11.49 11.75 13.28 14.08 General Creativity
M 75.00 76.61 87.67 80.30 83.74 80.36 88.54 84.37 82.78 SD 14.51 8.87 11.42 12.98 15.07 10.71 12.43 13.09 13.15 Note. The range of each creativity score is as follows: fluency, 40 to 160; flexibility, 40 to 160; originality, 40 to 160; general creativity, 41 to 160.
Figure 11 indicates that the HA group and the LA group were very similar to each other in terms of the four aspects of creative thinking abilities. Results of the Independent Samples t-test indicated that the HA group (M = 82.40, SD = 15.67) was significantly better than the LA group (M = 76.88, SD = 14.02) only in the aspect of flexibility (df = 146, t = -2.18, p = .03 < .05), with no significant differences in fluency, originality and general creativity.
Figure 11. Mean scores of creativity thinking abilities
Motivation Toward Verbal Creativity in English
The Questionnaire on Motivation Toward Verbal Creativity in English (QMVC) was designed to investigate the participants’ perceptions regarding two major motivational constructs: expectancy for success and subjective task value. Expectancy for success consisted of two dimensions: ability/expectancy beliefs and perceived task difficulty, whereas subjective task value included four dimensions: interest/importance, extrinsic utility value, required effort, and anxiety. On the one hand, with respect to expectancy for success, Table 24 shows that Class E (M = 3.89, SD = .94) had the highest expectancy that they would succeed in learning how to use English creatively, while Class A (M = 3.08, SD = 1.06) had the lowest expectancy. Likewise, as Class E demonstrated the
highest scores in ability/expectancy beliefs (M = 3.91, SD = .90) and task difficulty (M = 3.87, SD = 1.13), participants in this class indicated greater belief that they would
succeed learning how to use English creatively as well as lower difficulty in expressing creativity in English. By contrast, Class A had the weakest ability/expectancy beliefs (M
= 2.79, SD = 1.15) and Class B perceived the greatest difficulty in expressing creativity in English (M = 3.25, SD = 1.00).
On the other hand, with regard to subjective task value, although Class E (M = 4.44, SD = .56) had the highest subjective task value, all the six classes indicated that learning how to use English creatively was a valuable skill because all of their means were slighter larger than the midpoint, 4, on the seven-point scale. To be specific, first of all, Class F (M = 5.78; SD = 1.00) showed the greatest belief that it was interesting and important to learn how to use English creatively, while Class A had the weakest belief in this aspect (M = 4.83; SD = 1.16) among all. Secondly, Class E (M = 5.57; SD = .79) held the strongest perception that learning this skill was useful, while Class A again showed the weakest belief in this aspect (M = 4.94; SD = 1.10). Thirdly, in terms of required effort and anxiety, after participants’ responses were coded reversely, Class D had the highest scores in both required effort (M = 3.03; SD = .97) and anxiety (M = 3.76; SD = 1.10), meaning that they found learning this skill was easier and less anxious than the other classes did. In contrast, because Class B had the lowest scores in both required effort (M
= 2.19; SD = .94) and anxiety (M = 3.45; SD = 1.20), it implies that the participants in this class found learning this skill more laborious and worrisome than their counterparts.
Table 24. Results of the Questionnaire on Motivation Toward Verbal Creativity
Perceived Cost I: Required Effort
M 2.86 2.19 2.68 2.61 3.03 2.82 2.74 2.86 2.75
SD 1.31 .94 1.00 1.13 .97 1.06 .93 .98 1.05
Perceived Cost II: Anxiety
M 3.60 3.45 3.59 3.56 3.76 3.66 3.58 3.66 3.62
SD 1.37 1.20 .70 1.13 1.10 1.44 .72 1.09 1.11
Note. All the items on each subscale were evaluated on a seven-point scale.
While this questionnaire was evaluated primarily on the basis of the higher rating the stronger the belief, participants’ responses to the subscales of Perceived Task Difficulty, Required Effort and Anxiety were coded reversely, so higher ratings on these three subscales indicate less effort demand and less anxiety.
Figures 12 and 13 demonstrate the comparisons between the HA and LA groups in expectancy for success, subjective task value, and their components. As shown in both figures, the HA group obtained higher scores in all of the motivational constructs.
Independent Samples t-test was also performed to investigate whether there were any significant differences between the HA and LA groups in the abovementioned
motivational variables. Results revealed that the HA group obtained significantly higher scores in expectancy for success (HA: M = 3.75, SD = .88; LA: M = 3.28, SD = .93; df = 161, t = -3.31, p = .00), and its subscales, ability/expectancy beliefs (HA: M = 3.78, SD
= .93; LA: M = 3.16, SD = 1.00; df = 161, t = -4.07, p = .00) and perceived task difficulty (HA: M = 3.72, SD = .95; LA: M = 3.39, SD = 1.15; df = 161, t = -1.98, p = .05) than the LA group. Similarly, the HA group also indicated higher ratings than the LA group in subjective task value (HA: M = 4.39, SD = .55; LA: M = 4.12, SD = .47; df = 161, t = -3.20, p = .00) and one of its components, intrinsic interest and importance (HA: M = 5.64, SD = 1.02; LA: M = 5.16, SD = 1.01; df = 161, t = -2.98, p = .00). Nevertheless, both groups did not differ in extrinsic utility value and perceptions of required effort and anxiety. To sum up, the HA group possessed higher expectancy for success and subjective task than the LA group. To be specific, the HA group indicated more confidence in expressing creativity in English. They also stated greater interest in and importance to express creativity in English and considered expressing verbal creativity in English a useful skill.
Figure 12. Mean scores of motivational constructs related to expectancy for success
Figure 13. Mean scores of motivational constructs related to subjective task value
for Success
Subjective Task Value
Recognition of Creative Metaphors
The Recognition Task of Creative Metaphors (RTCM) aimed to measure the participants’ ability to interpret creative English metaphors accurately. As the 14 target items with a correct metaphoric expression were rated on a scale ranging from 0 to 5, the full score of this task was 70. As shown in Table 25, it is straightforward that the three classes in the HA group (D: M = 56.07, SD = 7.65; E: M = 56.81, SD = 5.77; F: M = 57.83, SD = 5.02) received higher scores than their counterparts (A: M = 45.93, SD = 9.10; B: M = 47.09, SD = 7.90; C: M = 53.29, SD = 4.51). Results of the Independent Samples t-test further showed that the HA group (M = 56.96; SD = 6.14) performed significantly better than the LA group (M = 48.66; SD = 8.10) in this recognition task (df
= 132.35, t = -7.32, p = .00 < .05). To sum up, the participants with higher English proficiency were more capable of interpreting creative metaphors accurately than those with lower proficiency.
Table 25. Results of the Recognition Task of Creative Metaphors
Class
Lower-Achievement Group Higher-Achievement Group
TOTAL
A B C Total D E F Total
N 28 22 24 74 29 31 35 95 169
M 45.93 47.09 53.29 48.66 56.07 56.81 57.83 56.96 53.33 SD 9.10 7.90 4.51 8.10 7.65 5.77 5.02 6.14 8.17 Note. The full score of the Recognition Task of Creative Metaphors was 70.
Production of Creative Metaphors
The Production Task of Creative Metaphors (PTCM) was employed to assess the participants’ four verbal creativity skills: metaphoric fluency, metaphoric flexibility, metaphoric originality, and metaphoric elaboration. There were a total of 1,318 metaphoric expressions produced by the entire participant pool: 499 metaphors in
Activity One, 363 metaphors in Activity Two, and 456 metaphors in Activity Three (see Table 26). This result implies that the second task, “My school is…,” was perhaps more difficult to the participants than the other two items. In addition, as Table 27 indicates, the participants in the HA group on average (M = 10.64, SD = 6.24) produced considerably a lot more metaphors than their counterparts in the LA group (M = 4.15, SD = 3.81).
Results of the Independent Samples t-test also show that the differences reached the significance level (df = 158.60, t = -8.34, p = .00 < .05). It is clear that English
proficiency had a significant influence on the participants’ fluency in producing creative metaphors in English.
Table 26. Statistical Summary of Creative Metaphors
Class A B C D E F Total
Activity One 33 27 47 94 157 141 499
Two 27 29 28 81 106 92 363
Three 34 38 44 85 135 120 456
Total 94 94 119 260 398 353 1,318
Note. Activity One = the life metaphor; Activity Two = the school metaphor; Activity Three = the love metaphor
All the metaphors were categorized by the researcher into different thematic categories (see Appendix P). The 499 life metaphors were classified into 29 thematic groups; the 363 school metaphors were grouped into 27 categories; the 456 love metaphors were classified into 28 categories. These classification systems were later reviewed by the same senior student majoring in English Instruction at University B mentioned earlier. Based on these classifications, participants whose metaphors involved a wider range of thematic categories were deemed having higher metaphoric flexibility.
As shown in Table 27, the metaphors created by the HA group (M = 9.46, SD = 4.98)
on average involved considerably more categories than those produced by the LA group (M = 3.86, SD = 3.27). The differences also reached the significance level (df = 162.66, t
= -8.80, p = .00). In other words, English proficiency also influenced an individual’s ability to produce a wide variety of creative metaphors.
Table 27. Results of the Production Task of Creative Metaphors
Class
Overall Productive Metaphoric Skill (in z score)
M -.56 -.42 -.24 -.41 .19 .55 .23 .32 .00
SD .61 .57 .57 .59 .43 .56 .45 .50 .65
With respect to the evaluation of metaphoric originality, metaphoric categories were assigned an originality score ranging from zero to two points based on the evaluation method discussed in Chapter Three (see Appendix Q). With all the scores assigned, the average originality score that a participant received was used to demonstrate his/her
metaphoric originality. As shown in Table 27, there were no obvious variations in the originality scores across the six classes. Statistical analysis further confirmed that the difference between the HA group (M = .37, SD = .25) and the LA group (M = .34, SD
= .45) was not significant (df = 106.97, t = -.40, p = .69 > .05). In other words, based on the result here, it can be inferred that English proficiency did not affect the originality of metaphors an individual creates.
Metaphoric elaboration is concerned with the amount of detail provided in each metaphoric expression. Participants who received a higher average elaboration score had a greater ability to produce highly elaborate metaphors than those with a lower score. As revealed in Table 27, the HA group (M = .95, SD = .36) received noticeably higher elaboration scores than the LA group (M = .60, SD = .57), and the difference was statistically significant (df = 115.24, t = -4.68, p = .00 < .05). In other words, metaphors produced by learners with higher English proficiency were more elaborate than those created by learners with lower English proficiency.
In summary, the findings discussed above demonstrated the differences between the HA and LA groups in the four aspects of productive metaphoric skills. As shown in
Figures 14 and 15, both groups showed greater variation in metaphoric fluency, flexibility, and elaboration. Results of statistical analyses confirmed that their differences reached the significance level.
Figure 14. Mean scores of metaphoric fluency and metaphoric flexibility
Figure 15. Mean scores of metaphoric originality and metaphoric elaboration
All the fluency, flexibility, originality and elaboration scores were further transformed into z scores, which were aggregated and averaged on a person-by-person basis to
demonstrate a participant’s overall ability to produce creative metaphors. Results showed that the HA group in general had higher competence in producing creative metaphors (M
= .32, SD = .50) than the LA group (M = -.41, SD = .59), with the difference being significant (df = 142.30, t = -8.50, p = .00). Taken together, all the results mentioned earlier support that the participants with higher English proficiency possessed better ability to produce creative metaphors than those with lower English proficiency. They produced more metaphors that came from a wider variety of ideational categories as well as involving a greater amount of detail than those with lower proficiency. Nevertheless, in spite of these differences, both groups did not differ in the originality of their metaphors.
Correlations Among All the Variables
Correlation analyses on the scores of all the variables including English proficiency, creativity, creativity motivation, creative metaphoric competence and their
subcomponents are discussed below to present a more comprehensive picture showing the associations among the variables under investigation in the current study. Table 28 indicates the correlational relationships among all the 12 variables measured in this study.
Five important findings are worth highlighting here. First of all, the participants’ ability to interpret creative metaphors was moderately correlated with their English reading skill (r
= .62, p < .01) and writing (r = .60, p < .01). It was also positively related to their expectancy for success (r = .32, p < .01) and subjective task value (r = .32, p < .01). In addition, the participants’ flexibility in creative thinking was also associated with their ability to interpret creative metaphors (r = .16, p < .05).
Secondly, the participants’ fluency in producing creative metaphors yielded strong associations with their English reading (r = .55, p < .01) and writing (r = .53, p < .01)
skills. It was also significantly related to their fluency (r = .25, p < .01) and flexibility (r
= .22, p < .01) in creative thinking, and their expectancy for success in verbal creativity (r
= .27, p < .01). Thirdly, with regard to the participants’ flexibility in producing creative metaphors, it was also largely correlated with their reading (r = .58, p < .01) and writing (r = .56, p < .01) skills, and their three creative thinking abilities (fluency, r = .27, p < .01;
flexibility, r = .26, p < .01; originality, r = .19, p < .05). Moreover, it was also
significantly associated with their expectancy for success regarding verbal creativity (r
= .30, p < .01).
As shown in Table 28, the participants’ originality in producing creative metaphors was only correlated with their originality in creative thinking (r = .17, p < .05). In other words, learners with greater originality in creative thinking were generally more likely to produce metaphors containing higher originality. This result also suggests that metaphoric originality was related particularly to an individual’s originality in creative thinking.
Lastly, in terms of metaphoric elaboration, it was positively correlated with English reading (r = .41, p < .01) and writing (r = .41, p < .01) skills as well. It was also associated with the participants’ expectancy for success and subjective task value, indicating that those with higher motivation toward verbal creativity tended to provide more elaborate metaphors. In contrast, it is interesting to note that metaphoric elaboration was not related to any of the three creative thinking abilities discussed in this study. To put this in another way, an individual’s creative thinking abilities were not associated with how well he/she could create highly elaborate metaphors. Instead, English
proficiency and motivation toward verbal creativity yielded much stronger associations with metaphoric elaboration.
Table 28. Correlations Among All the Variables Measured in This study
Pearson Correlation Coefficient
Variables 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
1. English Reading ―
2. English Writing .84** ―
3. TTCT_Fluency .16 .20* ―
4. TTCT_Flexibility .21* .24** .87** ―
5. TTCT_Originality .17* .21** .88** .85** ― 6. Expectancy for
Success
.34** .35** .12 .08 .11 ― 7. Subjective Task
Value
.25** .27** -.01 -.02 .01 .51** ―
8. RTCM .62** .60** .16 .16* .15 .32** .32** ―
9. PTCM_Fluency .55** .53** .25** .22** .16 .27** .08 .33** ―
10. PTCM_Flexiblity .58** .56** .27** .26** .19* .30** .09 .36** .98** ―
11. PTCM_Originality .02 .01 .08 .14 .17* .09 .03 -.05 .10 .12 ―
12. PTCM_Elaboration .41** .41** .07 .12 .07 .21** .27** .36** .08 .10 .03 ― Note. TTCT = Torrance Tests of Creative Thinking; RTCM = Recognition Task of Creative Metaphors; PTCM = Production Task of Creative Metaphors
*p < .05
** p < .01
Predictability of English Proficiency, Creativity, and Creativity for Motivation in Abilities to Interpret and Produce Creative Metaphors
This section presents results on the predictability of EFL learners’ English proficiency, creativity, and motivation for creativity in their abilities to interpret and produce creative metaphors in English. To examine the predictability of the former three independent variables, data were analyzed by means of multiple regression analysis. Five primary scores that were put under investigation here included the following: the average score of the reading z score and the writing z score (as an indicator of overall English
This section presents results on the predictability of EFL learners’ English proficiency, creativity, and motivation for creativity in their abilities to interpret and produce creative metaphors in English. To examine the predictability of the former three independent variables, data were analyzed by means of multiple regression analysis. Five primary scores that were put under investigation here included the following: the average score of the reading z score and the writing z score (as an indicator of overall English