國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
CHAPTER FOUR RESULTS
This chapter provides the experimental findings and discusses the participants’
responses to the different teacher feedback types. This chapter includes four sections.
The first section analyzes the effect of IDF by comparing the mean scores and the scores of each targeted error type between the pre- and the post-EWPT, the post and the retention EWPT. The second section investigates the effect of DF by comparing the mean scores between the pre- and the post-EWPT, the post- and the retention EWPT, and then compared two feedback groups on the post- and retention EWPTs.
Lastly, the fourth section examines the participants’ attitudes toward different teacher feedback types.
The Effect of Indirect Teacher Feedback (IDF)
In order to see how teacher feedback influence on the sentence writing accuracy of the IDF group, a paired samples t-Test was conducted; this section compared the mean scores of the IDF group and the score of each error type between the pre- and the post-, the post- and the retention EWPT.
Table 3 below is the statistic result of mean scores on the pre-EWPT and the post-EWPT for the IDF group. The mean score increased from 24.54 on the
pre-EWPT to 29.24 on the post-EWPT, thus, the mean score reached the significant improvement after the experiment (t = -4.596, p < .01). The result showed that IDF worked effectively with the participants and the improvement has been showed on the mean score of the post-EWPT (as shown in Table 3).
69
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 3
Paired T-Test on Pre- and Post-EWPT of IDF Group
Comparing the mean score of each targeted error type on the pre- and
post-EWPTs in the IDF group, the scores on the three targeted errors were increased significantly as follow: plural/singular noun, verb tense, and spelling. Specifically speaking, in the category of plural/singular noun, the mean score increased from 4.7 on the pre-EWPT to 5.73 on the post-EWPT, that is, the mean score reached the significant improvement after the experiment (t = -4.032, p < .01). In the category of verb tense, the mean score was improved from 3.97 on the pre-EWPT to 5.91 on the post-EWPT. This result showed that the mean score improved the most amongst the five error types after the IDF treatment (t = -4.103, p < .01). As for the category of spelling, the mean score also had a progress from 4.88 on the pre-EWPT to 6 on the post-EWPT. This means the improvement on spelling had reached the significant level after the experiment (t = -3.450, p < .01). In brief, IDF showed the significant
effectiveness not only on the mean score of EWPT but also on these three specific aspects (see Table 4).
Source N Mean SD t df
Pre-test 33 24.54 9.61 -4.596** 32
Post-test 33 29.24 10.78
** p<.01
70
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 4
Paired T-Test of Each Error Type on Pre- and Post-EWPT of IDF Group
**p<.01
In order to know the retention effect of IDF over three weeks, a paired t-Test was utilized and the data of the post- and retention EWPT were compared in Table 5. The mean score of retention test slightly dropped by 0.48 points which did not reach the significance (t = .704), and this result meant the relapse on the mean score of the retention EWPT was insignificant over three weeks after the IDF treatment. In other words, since the mean score of the retention EWPT had merely dropped a little, the test result showed that the participants could maintain what they learned from IDF treatment over three weeks. In effect, the IDF group could still keep the improvement from the post-test to the retention test even after three weeks.
Category Pre-test Post-test t df
SVA
(Subject -verb Agreement )
6 6.48 -1.449 32
Plural/Singular noun 4.70 5.73 -4.032** 32
Preposition 5 5.12 -.399 32
Verb Tense 3.97 5.91 -4.103** 32
Spelling 4.88 6 -3.450** 32
Total 24.54 29.24 -4.596** 32
71
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 5
Paired T-Test on Post- and Retention EWPT of IDF Group
Take the mean score of each targeted error type on the post- and the retention test into consideration, one error type, preposition, had improved significantly (t = -2.772, p < .01). The rest of error types did not change largely in three weeks after the
treatment (see Table 6). From the post- to the retention EWPT, the subjects improved significantly on the aspect of preposition. After consulting the English teacher of the IDF and DF groups, the improvement was the result of one review activity in the class.
Since their teacher figured out that students were not able to understand the usage of preposition, their teacher lectured and did some related exercises on this specific error type. The review activity was not related to the experiment and it was done in both the IDF and DF groups, thus, the similar improvement of preposition score should be seen in both two groups.
Source N Mean SD t df
Post-test 33 29.24 10.78 .704 32
Retention test 33 28.76 10.87
72
‧
Paired T-test of Each Error Type on Post- and Retention EWPT of IDF Group
**p<.01
In sum, based on the analysis of the IDF group, IDF worked effectively not only on the mean score of EWPT but also on three targeted errors: plural/singular noun, verb tense, and spelling; the regress of IDF on the retention test was insignificant in three weeks later, but in the aspect of preposition, the score has greatly improved.
The Effect of Direct Teacher Feedback (DF)
To know the effect of DF, a paired t-Test was used to compare the mean scores between the pre- and the post-EWPTs. In addition, to know the retention effect of IDF, a paired t-Test was utilized to mean scores between the post- and the retention EWPTs.
The comparison of mean scores on the pre-EWPT and the post-EWPT was listed in Table 7. In the DF group, the mean score increased from 26.03 points on the
pre-EWPT to 27.21 on the post-EWPT. The improvement had not reached the significant level (t = -1.383). In other words, DF did not work effectively with the participants on the mean score of EWPT (as shown in Table 7).
Category Post test Retention test t df
SVA
(Subject –verb Agreement )
6.48 6.27
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 7
Paired T-Test on Pre- and Post-EWPT of DF Group
To detect the retention effect of DF over time, a paired t-Test was conducted; the mean scores of the post- and the retention EWPTs were compared. The mean score increased slightly from 27.21 to 27.24 points but the change was not significant (t = - .043) (see as Table 8). The slight progress might be contributed to the practice effect.
Since the participants had done the post-EWPT three weeks ago, some of them might learn from the post-EWPT and apply what they learned to the retention EWPT.
Although the participants made a little progress, the change of the mean score was insufficient to reach the significant level. The test result meant that DF did not have significant difference on the retention test. In other words, the DF group could maintain the language knowledge they learned well since there was no significant regress on the retention EWPT.
Table 8
Paired T-Test on Post- and Retention EWPT of DF Group
Source N Mean SD t df
Pre-test 33 26.03 9.59 -1.383 32
Post-test 33 27.21 10.29
Source N Mean SD t df
Post-test 33 27.21 10.29 -.043 32
Retention test 33 27.24 9.80
74
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
According to the result, the DF group maintained the similar scores over three weeks and the score of retention EWPT showed that the DF group could maintain what they learned well.
To conclude, firstly, based on the comparison between the pre- and the
post-EWPTs, the participants did not improve significantly after the DF treatment. In other words, DF did not work effectively on the mean scores of the post-EWPT.
Secondly, according to the comparison between the post- and the retention EWPTs, no significant relapse could be seen on the mean score of the retention EWPT. The result showed that the DF group could maintain what they learned well over three weeks.
75
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
The Comparison on Post- and Retention Test of IDF and DF Groups To investigate the effect of IDF and DF, the comparison of IDF and DF on the post- and retention EWPTs was presented in this section.
The Comparison on Post-EWPTs
An independent sample t-Test was conducted to know if there was any difference on the post-EWPTs of the two groups (see Table 9). On the post-EWPTs, the IDF group ( = 29.24) outperformed the DF group ( = 27.21) by 2.03 points but the outperformance did not reach the significant level (t = .783). In other words, the effects of IDF and DF did not show significant difference on the post-EWPTs.
Table 9
Independent Sample T-Test on Post-Tests of IDF and DF groups
The Comparison on Retention EWPTs
On the view of retention EWPTs, an independent sample t-Test was used and the mean scores of two groups were compared in Table 10. The data did not show any significant difference between the IDF and DF groups on the mean scores.
On the mean scores of retention EWPTs (see Table 10), the IDF group scored 28.76 points while the DF group scored 27.24 points, the difference between these two grades was not significant (t = .595).
Post Test N Mean SD t df
IDF 33 29.24 10.78 .783 64
33 27.21 10.29
76
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 10
Independent Sample T-test on Retention Tests of IDF and DF groups
To sum up, comparing the post-EWPTs from two groups, the mean scores of the post-EWPTs did not show any significant differences. Furthermore, on the retention EWPTs, no significant difference between the IDF and DF groups could be seen on the mean score.
The Comparison of the Student Attitudes toward Two Feedback Types The Comparison of the Seven Sections
This part examines the participants’ attitudes toward different teacher feedback types by analyzing the participants’ responses toward the ELAQ; an independent sample t-Test was firstly conducted to compare the data of 7 sections and secondly to analyze 21 questions.
Three sections of the questionnaires showed the significant difference between the two groups (see as Table 11). Specifically speaking, the significant different sections were interest in foreign countries (t = -2.116, p < .05), motivation toward English learning (t = -2.538, p < .05), and attitudes about the treatment (t = -2.698, p
< .01). The statistic results showed that the DF group had higher motivation than the IDF group in these three aspects. In addition, the DF group presented a more positive attitude than that of the IDF group toward the treatment. That is to say, the IDF group seemed to have lower interest and motivation toward English learning and also
Retention test N Mean SD t df
IDF 33 28.76 10.866 .595 64
DF 33 27.24 9.79
77
‧
showed less preference to the treatment than the DF group.
One important point to be noted is that the IDF group ( =8.42) showed more nervousness on the section of anxiety than the DF group ( =8.21). Though the t-value has not reached the significant level, the IDF subjects still expressed the feeling of being anxious about the new format of the teacher feedback.
Table 11
Independent Sample T-Test on Main Section of IDF and DF groups
*p<.05; **p<.01
The Comparison of the Specific Questions
To probe the student attitude toward particular questions, the student responses on specific questions were compared with an independent sample t-Test by sections.
Table 12 listed out the statistic data of section 1.
Specifically speaking, Question 2 showed a significant difference between the two groups; the DF group ( = 2) presented significantly higher motivation than the IDF group ( = 1.52) on contacting with foreign friends via e-mail. (t = -2.617, p
No. Section IDF DF SD t df
1. Interest in foreign countries 4.88 5.94 1.55 -2.116* 64 2. Motivation toward English
learning
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
< .05). On question 3, the score of the DF group was ( = 2.27) higher than the IDF group ( = 1.85) and reached a significantly higher level (t = -2.015, p < .05). The results of question 2 and 3 implied that the DF group had higher motivation on being willing to write complete and correct English sentences because they wanted to communicate with foreign friends with written words (see as Table 12).
Table 12
Independent Sample T-Test on Section 1 of ELAQ of IDF and DF groups Section 1: Interest in foreign countries
No. Question IDF DF SD t df
Q1. I would like to travel to many countries. 1.52 1.67 .71 -.746 64 Q2. If it is possible, I would like to write
e-mail and contact with foreign friends. 1.52 2 .90 -2.617* 64
Q3. I want to write the complete and correct English sentences since I want to communicate with foreign friends with written words.
1.85 2.27 1.01 -2.015* 64
*p<.05
In section 2, motivation toward English learning, the DF group ( = 2.12) expressed higher motivation than the IDF group ( = 1.67) on question 4; and the difference was significant (t = -2.032, p < .05); that is, more participants in the DF group considered their motivation toward English learning was to help with their employment in the future, but less participants in the IDF group thought their motivation toward English learning was for their employment in the future. When it
79
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
comes to question 6, the DF group ( = 1.88) presented significantly higher motivation than the IDF group ( = 1.42) because more students in the DF group thought learning English now is to make the junior high English easier (t = -2.238, p < .05) (as shown in Table 13).
Table 13
Independent Sample T-Test on Section 2 of ELAQ of IDF and DF groups Section 2: Motivation toward English Learning
No. Question IDF DF SD t df
Q4. Learning English now is for future employment.
1.67 2.12 1.02 -2.032* 64
Q5. I learn English for using it while traveling abroad.
1.55 1.88 .75 -1.602 64
Q6. Learning English now is to make the
junior high English easier. 1.42 1.88 .93 -2.238* 64
*p<.05
In section 3 and 4, there was no significant difference between the IDF and DF groups on particular questions, the statistic data is listed in Table 14. IDF and DF performed alike on the sections of parents’ encouragement and instrumental
motivation. What worth to know was, from question 7 to 11, DF group presented more positive attitude than the IDF group. On the contrary, the IDF group expressed more positive attitude on question 12. To be more specific, the DF group agreed more than the IDF group on question 7: my parents encourage me to work hard on English. On
80
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
question 8, students in the DF group considered their parents have more expectation from them to write complete and correct English sentences. On question 9, the DF group regarded learning English as more important to their families than the IDF group. As for question 10 and 11, the DF group also expressed more preference whilst doing work book and writing activities than the IDF group. From question 7-11, the data showed that the DF group had more encouragement from their families than the IDF group. The DF group also enjoyed writing activities more than that of the IDF group.
However, on question 12, the last question in section 4, the IDF group considered English classes at school were more fun than the DF group. This response implied that the IDF group liked English classes at school better than DF group. Adding the results from question 7 to 12 together, although the DF group represented more
encouragement from parents and liked doing the work book and writing activities, the IDF group showed more preference for English classes at school.
81
‧
Independent Sample T-Test on Section 3 and 4 of ELAQ of IDF and DF groups Section 3&4: Parents’ encouragement & Instrumental motivation
No. Question IDF DF SD t df
Q7 My parents encourage me to work hard on English.
1.36 1.67 .55 -1.715 64
Q8 My parents expect me to write complete and correct English sentences.
1.52 1.73 .56 -1.136 64
Q9 At home, I and my families consider
learning English is very important. 1.7 2.06 .73 -1.805 64 Q10 I like doing the work book.
2.33 2.58 .89 -1.058 64 Q11 I enjoy participating in the writing
activities (ex., filling in the blanks, unscrambling or finding the errors in sentences…etc.) in English classes.
2.24 2.61 .87 -1.609 64
Q12 English classes at school are fun to me.
2.91 2.82 1.07 .359 64
As for section 5, one issue that should be noticed is that the IDF group ( = 2.15) expressed more anxious feelings than the DF group ( = 1.76) on question 13,
correcting the errors on workbooks always makes me nervous. The difference reached the significant level (t = 2.186, p < .05). This result might be attributed to the
implementation of a new feedback style since the DF group showed lower anxiety
82
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
with the traditional correction method on the same question (shown as Table 15).
Table 15
Independent Sample T-Test on Section 5 of ELAQ of IDF and DF groups Section 5: Anxiety
No. Question IDF DF SD t df
Q13 Correcting the errors on workbooks always makes me nervous.
2.15 1.76 .83 2.186* 64
Q14 I feel anxious if there are many errors happen during doing the work book.
2.48 2.61 1.25 -.429 64
Q15 Doing work book always makes me anxious.
3.03 2.79 .98 .925 64
*p<.05
In section 6, on question 18, I can understand the markings and the written
feedback from the teacher; the DF group ( =1.97) expressed more confidence than the IDF group ( = 1.48) on understanding the teacher’s feedback with the significant difference (t = -2.915, p < .01). As for question 19, I can understand the teacher’s correction style, participants in the DF group ( = 1.79) considered that they could understand the style of teacher feedback while participants in the IDF group ( = 1.45) were not sure if they could understand the style of teacher feedback (see as Table 16).
The difference between the two groups was significant (t = -2.058, p < .05).
83
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 16
Independent Sample T-Test on Section 6 of ELAQ of IDF and DF groups Section 6: Attitudes about the treatment
No. Question IDF DF SD t df
Q16 I like the way that teacher correcting my work book.
1.7 2 .68 -1.620 64
Q17 I highly value the feedback that teacher provides on the work book.
1.67 1.94 .69 -1.586 64
Q18 I can understand the markings and the
written feedback from the teacher. 1.48 1.97 .73 -2.915** 64 Q19 I can understand the teacher’s correction
style. 1.45 1.79 .74 -2.058* 64
*p<.05; **p<.01
Lastly, in section 7, on question 20, the teacher feedback is helpful for me to self-detect the errors, the DF group approved more of the help of teacher feedback and they agreed more on the fact that teacher’s feedback was helpful for the ability of error-detection. The DF group ( = 1.94) outperformed the IDF group ( =1.55) with significant difference (t = -2.335, p < .05) on this question (as shown in Table 17).
84
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 17
Independent Sample T-Test on Section 7 of ELAQ of IDF and DF groups Section 7: Self-evaluation after the treatment
No. Question IDF DF SD t df
Q20 The teacher feedback is helpful for me to self-detect the errors.
1.55 1.94 .79 -2.335* 64
Q21 According to the teacher feedback, I understand how to correct the errors.
1.52 1.6 .61 -.628 64
*p<.05
To conclude, in the seven sections, the DF group showed more interest and motivation on 3 main sections: interest in foreign countries, motivation toward English learning, attitudes about the treatment. That is, the DF group had more interest and motivation on English learning than the IDF group.
As for the specific questions, the DF group scored higher than the IDF group almost on most of questions on ELAQ instead of question 12, 13, and 15. Among these three questions, the IDF group surpassed the DF group only on the question 13 and the difference was significant (t = -2.186, p < .05). The result showed that the IDF group felt more anxious receiving the teacher feedback than the DF group. The DF group scored highly on 7 questions with significant difference, such as, question 2, 3, 4, 6, 18, 19 and 20. This outcome presented that the DF group had more motivation on learning English, more confidence on understanding teacher feedback and agreed more on the help of teacher feedback.
85
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
86
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
CHAPTER FIVE