CHAPTER 4 EXPERIMENTAL RESULTS
4.2 Comparisons of Listening Comprehension Performance between Two Groups . 46
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
46
performing the instructional experiment.
Table 4.1 The independent-samples t-test results of the pretest for both groups Assessment Item Groups (N) M (SD) t Sig.(2-tailed) Sectional examination VALRM (18) 82.778 (15.36)
.082 .615
SDLRM (21) 84.738 (8.17)
4.2 Comparisons of Listening Comprehension Performance between Two Groups
To assess whether there were significant differences in learners’ listening comprehension performance between the experimental group using the VALRM and the control group using the SDLRM, one-way analysis of covariance (ANCOVA) was conducted. Results of descriptive statistics and inferential statistics of the four listening comprehension tests were addressed as follows.
4.2.1 Comparisons of Four Listening Comprehension Performance between Two
Groups
To conduct ANCOVA for comparisons of four listening comprehension performances between two groups, the first step was to analysis the homogeneity of regression coefficients, and learners’ sectional examination scores were used as covariates in the analyses. The F test result (F=0.717, Sig=.403; F=0.087, Sig=.769;
F=0.015, Sig=.902; F=0.205, Sig=.653) did not reach the significant level, thus it
means the regression slopes of two groups is equivalent. This result confirmed the
‧
ANCOVA. As shown in Table 4.2, the mean score of the experimental group during the first experiment was not significantly different from that of the control group (F=0.589, Sig=.447>.05). Similarly, the mean score of the experimental group during the third experiment was not significantly different from that of the control group (F=1.93, Sig=.170>.05), either. These results indicated that learners in both groups had similar listening comprehension performance.However, the mean score of the experimental group was significantly higher than that of the control group (F=4.22, Sig=.046<.05) during the second experiment. In addition, the mean score of the experimental group was also significantly higher than
that of the control group (F=5.427, Sig=.025<.05) during the fourth experiment.
These results indicated that learners’ listening comprehension performance of the
experimental group was significantly superior to that of the control group after the
‧
4.2.2 Comparisons of the Fifth Listening Comprehension Performance between
Two Groups
The fifth listening comprehension test was to assess each group’s listening comprehension performance after the use of different listening review mechanisms for four times. Both groups were asked to complete the final comprehension test without reviewing the listening material.
To conduct ANCOVA for comparison of the fifth listening comprehension performance between two groups, the first step was to analysis the homogeneity of regression coefficients, and learners’ sectional examination scores were used as covariates in the analyses. The F test result (F=1.792, Sig=.189) did not reach the significant level, thus it means the regression slope of two groups is equivalent. This result confirmed the assumption of homogeneity of coefficients, and so this study further preceded the ANCOVA. As shown in Table 4.3, the mean score of the experimental group was significantly higher than that of the control group (F=4.399,
‧
Sig=.043<.05). This result indicated that learners’ listening comprehension performance of the experimental group was significantly superior to that of the
control group after four experiments.
Table 4.3 Results of one-way ANCOVA in the fifth listening comprehension
4.2.3 Comparisons of Average Listening Comprehension Performance between
Two Groups
To conduct ANCOVA for comparison of the fifth listening comprehension performance between two groups, the first step was to analysis the homogeneity of regression coefficients, and learners’ sectional examination scores were used as covariates in the analyses. The F test result (F=0.082, Sig=.776) did not reach the significant level, thus it means the regression slope of two groups is equivalent. This result confirmed the assumption of homogeneity of coefficients, and so this study further preceded the ANCOVA. As shown in Table 4.4, the mean score of the experimental group was significantly higher than that of the control group (F=6.692, Sig=.013<.05). This result indicated that learners’ average listening comprehension performance of the experimental group was significantly superior to that of the
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
50
control group.
Table 4.4 Results of one-way ANCOVA in average listening comprehension performance for both groups
Assessment Groups (N) M (SD) F Sig.(2-tailed)
Average Listening comprehension test
VALRM (18) 76.667 (19.3438)
6.692 .013*
SDLRM (21) 71.238 (12.8604)
4.3 Comparisons of Listening Comprehension Performance within and between
Two Groups with Different Learning Styles
To assess whether there were significant differences in learners’ listening
comprehension performance within and between the experimental group learners with different learning styles and the control group learners with different learning styles, the independent-samples t-test analysis was applied to compare differences in the posttest scores for both groups.
4.3.1 Comparisons of Listening Comprehension Performance within Two Groups
of Learners with Different Learning Styles
As shown in Table 4.5, the mean posttest score of the experimental group with FI learning style was not significantly different from that with FD learning style (t=.114, p=.142>.05). Similarly, the mean posttest score of the control group with FI learning
style was also not significantly different from that with FD learning style (t=.870,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
51
p=.150>.05). The results indicated that learners’ listening comprehension
performance within both groups of learners with different learning styles who respectively used the VALRM and the SDLRM for English listening learning did not
differ significantly.
Table 4.5 The independent-samples t-test results of the posttest within both groups of learners with different learning styles
Groups Learning Styles (N) M (SD) t Sig.(2-tailed)
VALRM FI (11) 78.182 (14.7093)
.114 .142
FD (7) 64.286 (23.7045)
SDLRM FI (11) 71.818 (17.2152)
.870 .150
FD (10) 60.000 (18.8562)
4.3.2 Comparisons of Listening Comprehension Performance between Two
Groups of Learners with Different Learning Styles
As shown in Table 4.6, the mean posttest score of the FI learners in the experimental group was not significantly different from those in the control group (t=.320, p=.362>.05). Similarly, the mean posttest score of the FD learners in the experimental group was also not significantly different from those of the control group (t=.381, p=.684>.05). The results indicated that learners’ listening comprehension performance between both groups of learners with different learning styles who respectively used the VALRM and the SDLRM for English listening learning did not differ significantly.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
52
Table 4.6 The independent-samples t-test results of the posttest between both groups of learners with different learning styles
Learning Styles Groups (N) M (SD) t Sig.(2-tailed)
FI VALRM (11) 78.182 (14.7093)
.320 .362 SDLRM (11) 71.818 (17.2152)
FD VALRM (7) 64.286 (23.7045)
.381 .684 SDLRM (10) 60.000 (18.8562)