gradually-increasing speech rate on listening comprehension were explicated in section 5.2. Finally, interpretations for the learners’ perception toward speech rate are addressed in section 5.3
5.1. The Effect of Slowed-down Speech Rate on Listening Comprehension
This section first illustrates the treatment of a confounding score. Then, the final outcomes of the effect of slowed-down speech rate on listening comprehension are reported.
The Confounding Score:
Closely examining the raw scores of the listening comprehension test, the researcher found that there was one score greatly influencing the overall results. This score was conspicuously higher than the other scores so that it prompted the
researcher to conduct an interview with the participant who received the score.
The interview was about four weeks after the formal test. Before the interview, the researcher invited the participant to take the listening comprehension test again.
The researcher then examined each answer given by the participant while the
participant explained in retrospection how she got the answers and what she was thinking. The discussion demonstrated that she was not guessing in the listening comprehension test. The researcher inquired about her English learning method, and found that she listened to BBC an hour per day. This practice, according to the participant, having been on-going for almost seven years, gave a reasonable
explanation for her particularly high score. Her outstanding English proficiency was later confirmed by her English listening teacher, who once mistook her as someone who had studied, or at least spent some time, in an English-speaking country.
This participant’s English proficiency is clearly superior to that of the other participants in this study. As a result, it is necessary to exclude her score for the purpose of obtaining a valid outcome. After omitting her score, it is not surprisingly that, as Table 23 displays, a significant difference was found between Group A listening to the slower-than-normal speech rate and Group C listening to the average speech rate (p= .016). The third hypothesis of no difference in listening
comprehension performance between the slower-than-normal speech rate and the average speech rate was therefore not supported.
Table 22: Results of Multiple Comparison measuring listening comprehension test scores without confounding data of Group A, B, and C
Comparison sig
Group A vs. Group B .611
Group A vs. Group C .016*
Group B vs. Group C .123
Note. Group A= Slower-than-normal speech rate; Group B= Moderately slow speech rate;
Group C=Average speech rate
The Final Outcome:
As suggested by the hypotheses concerning the effect of the slowed-down speech rate on listening comprehension, the principal prediction from this study is that the listening comprehension would not increase if the speech rate is too slow and that a suitable slow speech rate, represented by the moderately slow speech rate in this study, could increase listening comprehension effectively. When a slowed-down speech rate is overstretched, as represented by the slower-than-normal speech rate in the study, the listening comprehension would not be improved when compared to results for the average speech rate.
This motivation of examining the possible invalid effect from overstretched slowed-down speech rate, as noted earlier in Chapter One (p.1), grew out of Griffiths’
studies done in 1990a and 1991a. A comparison of the results in these two studies showed that attempting to increase listening comprehension by slowing down the speech rate from the average speech rate was ineffective when the slowed-down speech rate was too slow. In Griffiths’ study done in 1990a, the results showed that the text delivered at a slower speech rate, 100 WPM, was not more comprehensible than the same text delivered at a normal speech rate, 159 WPM. Nonetheless, in 1991a, a significant difference in listening comprehension was obtained between the slower speech rate and the normal speech rate when the slower speech rate was adjusted to 127 WPM and the normal speech rate was adjusted to 188 WPM. It was then hypothesized that when compared to the average speech rate, a slower-than-normal speech rate would not lead to an increase in listening comprehension, but a
moderately slow speech rate would.
The results in this study, though, failed to support the hypotheses. For the first hypothesis, it was predicted that the moderately slow speech rate would lead to better listening comprehension performance compared to the slower-than-normal speech
rate. However, no significant result was obtained to support a difference in comprehension. The second hypothesis was that the moderately slow speech rate would lead to better listening comprehension compared to the average speech rate, but the results established no significant difference in the level of comprehension between these two speech rates. The moderately slow speech rate was therefore not effective in increasing increase listening comprehension compared to the slower-than-normal or the average speech rate. The third hypothesis was also not supported, as illustrated in the previous section.
While it was predicted that there was no difference in listening comprehension between the slower-than-normal speech rate and the average speech rate, a significant difference between these two speech rates in listening comprehension was revealed.
The slower-than-normal speech rate led to a significant increase of listening comprehension compared to the average speech rate, demonstrating that the most adequate slowed-down speech rate for increasing listening comprehension is the slower-than-normal speech rate. In other words, the results indicated that between the two slowed-down speech rates, the slower-than-normal speech rate, and not the moderately slow speech rate, is the most beneficial slowed-down speech rate to increase listening comprehension.
Although running counter to the hypotheses, this finding is plausible for the following reasons. First, the slower-than-normal speech rate in this study, literally around 125-130 WPM, was not as slow as the slow speech rate in Griffiths’ study, 100 WPM. It is, therefore, the researcher’s contention that a slower speech rate of around 130 WPM will result in improved listening comprehension, whereas a rate of 100 WPM will have a negligible effect. 100 WPM was not adopted as the
slower-than-normal speech rate because the speech rate adjustment instrument employed in this study, CoolEdit, can only slow the original speech rate by 20%
without distorting the original pitch. The inherent disadvantage of the instrument was not given sufficient consideration because it is the only validated speech rate
adjustment tool available for rate studies. In summary, the limitation of CoolEdit left a rate of around 130 WPM rate as the slowest usable speech rate. This is strongly believed to be a major factor in failing to achieve the anticipated results.
Another possible cause for the contradictory results is the genre type of the
listening text: the news. News, which is comprised of specialized text full of technical or specialized vocabulary and filled with information, can be daunting for language learners, especially in the case of EFL students. The slower-than-normal speech rate when applied to other text types might not be facilitative; however, when it is applied to the news, the slower-than-normal speech rate is more likely to effectively enhance listening comprehension for language learners.
To conclude, the students listening to the slower-than-normal speech rate attained a significantly higher level of listening comprehension compared to the students listening to the average speech rate. However, the students listening to the moderately slow speech rate did not outperform the students listening to the average speech rate.
This shows that the slowed-down speech rate intended to increase the listening