significantly increase listening comprehension. In short, the slower-than-normal speech rate was proved to effectively slowed-down speech rate that effectively increases listening comprehension of authentic news with the average speech rate.
5.2. The Effect of Gradually-increasing Speech Rate on Listening Comprehension
This section is divided into six parts. The findings obtained from the main idea recall are discussed first, followed by the discussion on detail recall. Then, a
comparison of scores distribution between the main idea recall and detail recall are addressed. Following the comparison, the result of the listening comprehension test is examined. Afterwards, the results from the semi-structured interview are discussed.
Finally, a short summary of the results is presented
Main Idea recall:
As displayed in Table 13, the mean scores of main idea in Group D were higher than Group E in the first two weeks of the training. In the third week, however, Group D dropped from a mean score of 8.888 in the second week to a mean score of 4.444 whereas Group E surged from a mean score of 3.461 in the second week to a mean score of 6.92. This contrast continued to the fourth week when Group D climbed to a mean score of 6.111, which was closer to the score obtained in the first week of the training, while Group E plummeted to a mean score of 2.884, the lowest score in the training.
The improved listening comprehension level of Group E from the second week to the third week was probably due to the fact that Group E had constantly listened to the listening texts with average speech rate, which fostered their ability to ignore the unfamiliar or unknown words, thus contributing to the listening comprehension improvement. This jump of score was then followed by a huge decline of score at the end of the training. One possible explanation for the sudden drop postulated by the researcher is the fatigue effect of the four-week training where the participants listened to the texts with authentic average speech rate five days a week.
As for Group D, the drop of listening scores in the third week might suggest a speech rate threshold period when the language learners still relied on word-by-word processing due to the previous slower speech rate, and plunged due to the
comprehension breakdown attributed to increased speech rate. In fact, the sacrificed
listening comprehension is part of the process of adjusting to the faster speech rate.
This is necessary for the listeners to begin to comprehend the yet-to-be digested speech rate, and finally to proceed to the later stage of comprehending texts delivered at a faster speech rate at ease. In this study, the participants in Group D entered the threshold period when the speech rate reached around 150WMP, as evidenced by the decrease of listening comprehension. Eventually, the relatively higher mean score of Group D in the last week of the training shows the end of the threshold period. This result suggests that, for the use of the gradually-increasing speech rate as a training device, a decrease in listening comprehension is indispensible to attain a higher level of listening comprehension for the gist.
As previously discussed, the group with the training of steady average speech rate could receive high mean scores during the four-week training, but it also
experienced a tremendous drop in the last week of the training. As for the group with the training of the gradually-increasing speech rate, a decreased listening
comprehension on main idea would be observed, and yet was followed by a rather satisfying listening comprehension performance. In addition, considering the significant difference between Group D and Group E in the last week of the
four-week training (p= .035), the training group with the gradually-increasing speech rate experienced a stronger facilitative effect on listening comprehension than the training group with steady average speech rate. This result is aligned with the fourth hypothesis.
Detail recall:
There was no significant difference of detail recall between Group D and Group E throughout the four-week training. The fourth hypothesis that, regarding listening for detail, the training with the gradually-increasing speech rate does not have a
stronger facilitative effect on listening comprehension than the training with steady average speech rate is therefore not supported.
In fact, all the mean scores of Group D were higher than those for Group E throughout the four-week training. It is plausible that the participants in Group D could catch more detail than Group E at the beginning of the training since Group D started with a slower speech rate compared to Group E. It is worth noting that when the speech rate was gradually increased to the average speech rate, the participants in Group D could still maintain their ability to process faster input and recall relatively more detail compared to Group E. This shows that the gradually-increasing speech rate could be used as a training device to develop listeners’ rapid word recognition, or automatic processing device, already discussed in the literature review chapter (p. 12) and described as one of the important traits a competent second language speaker should acquire.
Comparison of Score Distribution between Main Idea Recall and Detail Recall:
A diverse score distribution between main idea recall and detail recall of Group D and Group E was observed. Considering the detail scores, Group D and Group E showed a similar detail score distribution throughout the four-week training, whether the score was rising or dropping. This was in contrast to the score distribution of main idea recall where a high score in Group D was always accompanied by a low score in Group E throughout the training, and vice versa.
One possible explanation for this diverse difference between the score
distribution is that the recall of detail is less affected by the change of the speech rate, which, nonetheless, seriously influences the main idea recall. Listeners could score high on the detail section regardless of the speech rate for two reasons. First, note-taking is allowed during the listening. Good note-takers could jot down
everything they hear to get a fairly high score. Second, the jotting down of information only takes one-way processing, bottom-up processing, which merely requires the listeners to pay attention to the smallest units. On the contrary, grasping the main idea of the listening takes more effort since the note-taking doesn’t guarantee the incoming signals can be reconstructed as a whole. To listen for the gist, the
listeners have to take in the incoming spoken signals while making sense of the signals as related information, and finally building up the elements into a complete message. This two-way processing, being more mentally challenging, is probably more controlled by the change of the speech rate. The diverse degree of effect of speech rate on retrieving detail and main idea was clearly revealed by comparing the main idea scores and detail scores of Group D and Group E in the beginning of the training. In terms of the main idea scores, Group D starting with a slowed-down speech rate could score higher than Group E beginning with the average speech rate.
The detail scores, though, were approximately the same between these two groups. In summary, the effect of the speech rate on listening comprehension might vary
depending on what kind of listening comprehension performances were observed.
Listening Comprehension Test:
As presented in the results section, there was no significant difference between the mean scores of Group D and Group E on the listening comprehension test. This suggests that the training with the gradually-increasing speech rate does not have a stronger facilitative effect on listening comprehension than the training with steady average speech rate.
Semi-structured Interview:
The results from the semi-structured interview showed that the listening training, whether with or without the gradually-increasing speech rate, has a
confidence-building effect on the participants’ listening comprehension ability, supported by the fact that the participants from both groups respond positively to the training, including the improvement of listening comprehension performance and the willingness to do listening self-training using news as their listening material.
Summary:
As discussed above, in terms of the listening comprehension performance, only the results from listening for main idea demonstrate a stronger facilitative effect from the training with the gradually-increasing speech rate than the training with steady speech rate. No difference between these two training conditions was found, though, regarding the other two measurements of listening comprehension, including the free recall test of listening for detail and the final listening comprehension test.
Concerning the participants’ feedback from the semi-structured interview, both uses of the rate-controlled training led to a positive perception from the participants that their listening comprehension performance was improved after the training.
Given these results from both quantitative and qualitative measurements of listening comprehension, the only significant strength of the gradually-increasing speech rate training is the enhanced listening comprehension level of listening for main idea.