This section begins with reporting the results of the pilot study, followed by presenting the data collection procedure for answering the first and the second research question respectively.
3.3.1. The Pilot Study
Before the formal experiments, a pilot study was conducted to examine the validity of the listening comprehension test designed by the researcher. Eighteen students in the continuing education center of National Taiwan Normal University were asked to take adapted TOEIC, and then the listening comprehension test. With two sets of scores from each student, the researcher performed Pearson Correlation analysis between the adapted TOEIC and the researcher-designed listening
comprehension test (r= 0.463). Although the correlation result was not satisfying, it was demonstrated that the researcher-designed listening comprehension test had a moderate correlation with the standard listening comprehension test. This outcome is essential since the listening comprehension test designed by the researcher shared
more characteristics with “in-house” tests, which decrease the generalizability of the results compared to a standardized measure (Berne, 2004). By verifying this
researcher-designed listening comprehension test, the interpretation of the final outcome would be strengthened.
To sum up, the results of the pilot study showed that it was legitimate enough to use the researcher-designed listening comprehension test as a test instrument to assess the participants’ listening comprehension performance. This researcher-designed listening comprehension test is addressed as the listening comprehension test in this thesis.
3.3.2. Phase one for first research question
To investigate the effect of slowed-down speech rate on listening comprehension, the first phase of the experiments was conducted. The following provides detailed steps of the data collection procedure.
Sixty participants first filled out the background information questionnaire. Then, the participants took the adapted TOEIC. After taking the test, the researcher listed the participants starting with the highest scorer and ending with the lowest scorer. Based on this score-based name list, the participants were assigned from Group A, to Group B, to Group C, and then back to Group A and so forth. A closer listening proficiency level among these three groups could be resulted from this assignment. After the grouping, five participants did not attend the listening comprehension test due to personal reasons. Therefore, the formal experiment of speech rate treatment was conducted with 19 participants in Group A, 19 participants in Group B, and 17 participants in Group C
The participants in Group A took the test with the listening passages delivered at the slower-than-normal speech rate, ranging below 130 WPM. Group B received the
test at the moderately slow speech rate, ranging from 130 to 150 WPM. As for Group C, the test was delivered at the average speech rate, ranging from 150 to 170 wpm.
While administering the listening comprehension test, reading the questions before or during the listening was not permitted, but note-taking during the listening was allowed. Ten minutes were given for answering the 10 questions in the listening comprehension test. At the end of the test, the participants were asked to fill out the background knowledge questionnaire inquiring about their prior knowledge of the news. The scores from six participants were excluded due to their report of content familiarity. Finally, the data collected in the first phase of experiments was Group A of 15 participants, Group B of 17 participants, and Group C 17 of participants.
3.3.3. Phase two for the second research question
The effect of gradually-increasing speech rate on listening comprehension was explored in the second phase of the experiments. Each step of data collection procedure in this phase is described as below.
Thirty-four participants first filled out the background information questionnaire, and then took the adapted TOEIC. The researcher assigned the participants to two groups, which were Group D of 18 participants and Group E of 16. However, four of the participants in Group D had already taken part in the experiments for the first research question, and three of the participants in Group E quit in the middle of the training. Therefore, the final number of the participants was fourteen in Group D, and thirteen in Group E.
Both Group D and Group E received a four-week training in which Group D listened to the listening texts presented with the gradually-increasing speech rate, while Group E received the same listening input delivered with authentic average speech rate. In the four-week training, the participants were asked to listen to the news
on a Web log set up by the researcher. On the blog, the participants were asked to leave a message on the blog before and after they listened to the news, allowing the researcher to make sure of their participation in the training. Every day during the training, the researcher checked their messages on the blog and gave them a call if any names were missing on the blog.
Once a week during the four-week training, the participants in Group D and Group E were invited to class on a group basis to recall one entry of news, whose main idea and detail were recalled separately on the free recall protocol (Appendix F).
The raters, accordingly, assigned two scores on the protocols. Therefore, at the end of the four-week training, there were four free recall protocols, eight scores, from each participant.
At the end of the training, the participants took the listening comprehension test.
Based on the results of the listening comprehension test, four participants from each group were invited to attend the semi-structured interview, which concluded the data collection procedures for examining the effect of gradually-increasing speech rate on listening comprehension.
CHAPTER FOUR RESULTS
This chapter contains two sections, which respectively report findings to the two research questions posed previously in Chapter One (p.1). Section 4.1 presents the results concerning the first research question inquiring the effect of slowed-down speech rate on listening comprehension. Section 4.2 reports the results regarding the second research question constructed to investigate the effect of the
gradually-increasing speech rate on listening comprehension.
4.1. The effect of the slowed-down speech rate on listening