62
CHAPTER FIVE
CONCLUSION, IMPLICATIONS AND SUGGESTIONS
This concluding chapter consists of three parts. The findings and conclusion of this study are presented in 5.1. The pedagogical implications for teaching reading and testing reading comprehension are proposed in 5.2. The limitations of this study are presented in 5.3 along with some suggestions for those who intend to conduct similar or related studies in the future.
5.1 Findings and Conclusion
With the intention to investigate the effect of sound stimulation on Taiwanese senior high school students’ reading comprehension, three different reading
modes—silent reading, self-reading aloud, and teacher’s reading aloud—are applied in this study. 75 second graders, divided into three groups, are required to do 21 reading passages with 67 comprehension questions in total. The reading passages are of two text types—narration and exposition, which differ obviously in their
organizations, for narration has sequencing structures while exposition tends to be non-chronological. The comprehension questions are also of two question
types—main idea questions and detail questions. Main idea questions are designed to test the students’ general comprehension of the reading passages while detail
questions are for specific details; the former pertains to macro reading skills while the latter the micro ones. Ten volunteers from each group are interviewed after the
reading comprehension tests. The interview is employed for gaining insight from the students’ perspectives. The results of the reading comprehension tests and the
interview are summarized as follows:
With regard to the students’ performance in the reading comprehension tests, there are five major findings.
First, according to the means of the students’ correct responses on overall reading comprehension, on two different text types and two different question types, the group of silent reading outperforms the group of self-reading aloud, which in turn
outperforms the group of teacher’s reading aloud. In other words, silent reading is shown to be the most effective reading mode while teacher’s reading aloud is the least effective for the students’ overall comprehension, their comprehension of the two different text types, and their correctness in answering the two different question types.
This result shows that the sound of oral reading somehow hinders the students’
comprehension. Moreover, the teacher’s reading in larger semantic units inhibits the comprehension of the students who tend to read word by word, leading to the poor performance of the teacher’s-reading-aloud group.
Second, after the statistical analysis of one-way ANOVA and post hoc test, the results can be generally synthesized into two kinds:
1. There is no significant difference among the three groups’ performances in the following four aspects: (1) in comprehending expositions, (2) in answering main idea questions, (3) in answering main idea questions of narrative and expository texts respectively, and (4) in answering detail questions of expository texts.
2. There is significant difference among the three groups’ performances in the
following four aspects: (1) in overall reading comprehension, (2) in comprehending narratives, (3) in answering detail questions, and (4) in answering detail questions of narrative texts. In all the above situations, the significant difference falls only on the group of silent reading and the group of teacher’s reading aloud. No significant
difference between the self-reading-aloud group and the teacher’s-reading-aloud group implies that the teacher’s leading the students to read in larger semantic units doesn’t have obvious benefit for the students’ reading comprehension. No
significant difference between the silent-reading group and the self-reading-aloud group implies the students tend to read word by word even when reading silently.
But the teacher’s reading by phrases and clauses makes the students’ in the group unable to read word by word, bringing about the significantly inferior performance of the teachers-reading-aloud group. Table 5.1 summarizes the results of the reading comprehension tests.
Table 5.1 Results of the Reading Comprehension Tests
Mode/Sig.level
Aspect Mean
Silent reading (Group A)
Self- aloud (Group B)
Teacher’s aloud (Group C)
Significant level of difference
Overall highest lowest Sig in A and C
Narration highest lowest Sig in A and C
Exposition highest lowest Non sig
Main idea questions highest lowest Non sig
Main idea questions
of narratives highest lowest Non sig
Main idea questions
of expository texts highest lowest Non sig
Detail questions highest lowest Sig in A and C
Detail questions of
narratives highest lowest Sig in A and C
Detail questions of
expository texts highest lowest Non sig
Third, concerning the students’ comprehension of narrative and expository texts,
the different structures of the two text types are the most possible reason for the three groups’ significantly different performances in narration, but not in exposition.
Narrative texts have sequencing structures but expository texts do not. While reading the narrative texts, the students have to fully comprehend the order of the events to have good performance. Since the silent-reading group takes the advantages of being able to read back or forth the texts, it stands to reason that the group members could better comprehend the series of events in narrative texts and thus outperform the students in the oral-reading groups. However, such superiority of the silent-reading group is not obvious in comprehending the expository texts, which don’t have sequencing structures, leading to similar performances of the three groups. The
students of the teacher’s-reading-aloud group, controlled by the teacher’s reading pace, might not fully comprehend the sequencing events of the narrative texts, consequently couldn’t perform as well as the other two groups.
Fourth, with respect to main idea questions, no significant difference is found among the three groups’ performance. Further investigations on main idea questions of different text types also show no significant difference among the three groups’
performances. The consequences show that the influence from the sound is not obvious for comprehending the general ideas of the texts, no matter the gist is explicitly stated, such as that in exposition, or implicitly presented, such as that in narrative.
Fifth, as for detail questions, the three reading modes manifest different effects on the students’ performance. With various subtypes, detail questions are generally more difficult. The separation of reading texts and their comprehension questions in this study makes answering detail questions even more difficult, for the students have to completely understand or even memorize the details to have good performance.
The silent-reading group reads without sound as distraction and is more likely to have good performance. Further investigation on detail questions of the two text types shows the silent-reading group significantly outperforms the teacher’s reading aloud group in comprehending narratives, but not expository texts. This result further supports the finding that the sequence structure of narrative is an influential factor for the three groups’ different performances. Since silent-reading group takes advantage of being able to read back or forth the text and reading without sound as distraction, it stands to reason that they comprehend the detail questions in narratives even better.
With regard to the interview reports, the major findings are presented as follows.
First, concerning the students’ opinions about the more beneficial reading mode, eight out of the ten interviewees in the self-reading-aloud group and seven out of the ten in the teacher’s-reading-aloud group consider silent reading more helpful to their reading comprehension. These interviewees generally consider the sound and the controlled speed in the oral reading modes inhibit their reading comprehension. The interviewees’ reports correspond to the test result, which shows that silent reading is the most effective mode for the students’ reading comprehension. However, six out of the ten interviewees in the silent-reading group believe reading a word aloud helps recognizing the word and three of them mention individually self-reading aloud enhance their reading comprehension. According to their reports, it seems that the sound would be helpful to comprehension with appropriate personal use, but the sound is a distractor when the students read with controlled pace and without pauses for thought.
With respect to the two different text types, most of the interviewees of the oral-reading groups report that the easier text type, either narration or exposition, is still the easier one to read while accompanied by sound. Only one of them reports it’s
harder to grasp the sequencing events of narratives while performing self-reading aloud. Her report is in agreement with the test result, which demonstrates significant difference with or without the sound in narration, but not exposition. But interestingly, eight out of the ten interviewees in the silent reading group mention their tendency to read back or forth a text when encountering some incomprehensive points. It’s possible that the students have the habit of reading back or ahead a text, but few of them are aware of the different organizations of the texts, which leads to their similar reports regardless of the utterance of the sound or not.
As for the two different question types, most of the interviewees performing the oral reading modes consider answering detail questions easier while under the influence of sound. But they report it easier mostly because they think understanding certain sentences is easier than understanding the whole text. Such impression is very likely to have a gap with their real performance because main idea questions
sometimes can be answered easily by identifying the topic sentences of the texts while answering detail questions might involve integration of numerous sentences. Two interviewees specifically regard the main idea questions of narrative as more difficult, for they have to understand the whole text to make proper inference. Their report is not consistent with the test result, which shows no significant difference among the three groups’ performance in the main idea of both narrative and expository texts. A possible explanation is that the narrative texts applied in this study are not difficult for the students to infer the main ideas. Other interviewees seem to have no awareness of the different forms of the main ideas in the two text types.
According to the above findings, no matter from the statistical results of the reading comprehension tests or the interview reports, silent reading is shown to be the most beneficial mode to senior high school students’ reading comprehension. Oral
reading might be good for providing variation to regular courses or for raising the students’ motivation, but with regard to comprehension, certain time for silent reading is still needed for each individual student. That is, reading had better be an individual activity for senior high school students. By providing them chances to read at their own pace with sufficient time for thinking, the students are more likely to have better comprehension of the texts and superior performance in the reading tests.
5.2 Pedagogical Implications
Based on the analyses and findings of this study, three pedagogical implications are derived for teaching reading in Taiwan high school English classes. They are listed in the following.
First, while introducing a reading text by oral reading, teachers had better
provide time for silent reading beforehand or afterward the sound input, especially for narrative texts, which have sequence structures. With sufficient time for silent reading, students are able to read back or ahead the text and better comprehend the series of events. Students should not be asked the comprehension questions right after listening to the tape or reading the text aloud, especially those questions concerning the details of the text. Focusing on the pronunciation, the students may comprehend only certain fragments of the text. Detail questions directly following the oral reading may make the students feel depressed.
Second, students should be taught to identify the different structures of narrative and expository texts and to use this knowledge to guide their reading process.
Teachers should provide the students with some well-organized reading materials to enhance their awareness of the different text structures. Moreover, students should also receive instruction on how to identify the main ideas of the two text types, no
matter the main ideas are explicitly stated or implicitly implied. With the awareness of text structures and the ability to identify the main ideas of different text types, the students are more likely to gain reading achievements.
Third, although silent reading is shown to be the most beneficial mode in this study, the students are found reading word by word while reading silently. Building reading speed is thus necessary for the senior high school students. Researchers such as Mikulecky (1990) and Witte (1985) point out that learning to read faster leads to better comprehension. When reading faster, students cannot focus on each word;
rather, they focus on groups of words. Reading fast also helps the student think about main ideas, not just single words, and thus understand better. Besides, in large-scale tests, such as the English Subject Ability Tests, the passages in the reading
comprehension section are increasing in length and quantity. Since the students are expected to demonstrate their reading performance in a limited time frame, training and practice in building up the students’ reading speed is quite essential for their better performance of the tests. Moreover, guiding the students to read in large semantic units is also a possible way for building up the students’ reading speed (Witte, 1985). Although teachers’ reading aloud is shown to be the least effective mode for the students’ comprehension, it is possible that the students are not
accustomed to such reading mode and thus have inferior performance. Amer (1997), who finds teacher’s reading aloud beneficial to Arabic six-graders, suggests
employing teacher’s reading aloud at regular basis. Instruction on how to group the words into phrases and clauses before the guided reading is necessary and students’
own practice in identifying semantic units afterward is also important, which may lead to their habitual reading pattern even if they read silently.
5.3 Limitations and Suggestions for Further Research
Some limitations of this study are presented in the following, along with suggestions for those who intend to conduct similar studies in the future. The limitations and suggestions are presented in three main domains: (1) silent and oral reading modes, (2) the text types and the question types, and (3) other suggestions.
With respect to silent and oral reading modes, the limitations with practical suggestions are as follows. First of all, with each group performing one reading mode, only self-reading aloud or teacher’s reading aloud can be compared with silent
reading, which is the students’ habitual reading mode. Similar to the results of the questionnaires in Hsieh’s (1998, 2000) two studies, no interview reports can be obtained from the comparison of the two oral reading modes. One single group performing all the three reading modes can serve that purpose, but the modes and text types should be applied with careful design to avoid the practice effect. Secondly, the reading pace of oral reading is generally slower than that of silent reading and
teacher’s reading aloud, especially when the students of the whole class are reading aloud together. As a result, the time needed for the group of self-reading aloud is generally longer than that for the other two groups. In this study, the time difference for performing the reading modes is about 10-15 minutes. Better ways have to be figured out to minimize the time difference. The students’ attention span is another uncontrollable factor. It’s possible that some of the students of the oral-reading groups read silently by themselves without reading aloud or following the teacher’s reading aloud. Besides, with three of the interviewees showing their preference for
individually self-reading aloud, investigation on the effect of individually self-reading aloud and the whole class’ reading aloud in chorus could be a potentially rewarding topic for further research. Future studies could also try providing training on the two
oral reading modes. The effect of the oral reading modes on comprehension then could be examined easily by comparing the students’ performance before and after the training.
As for the text types and the question types, four limitations with practical suggestions are listed in the following. First, since few students are aware of the different structures of the text types, instruction beforehand rather than explanation afterward in the interview may lead to more meaningful reports from the interview.
Second, the texts applied in this study and the words put in the margins are chosen subjectively by the researcher, and this might influence the results of the study. For example, the narrative texts applied in the study seem to be more difficult than the expository texts, which might be the reason for the significant difference among the three groups’ performance on narration, but not exposition. Future studies had better adapt some objective methods in choosing the texts and the key words. Moreover, in the reading comprehension tests, the passages and questions are separated to prevent the students of the oral-reading groups from reading back to the text silently. But the students’ ability of memorization is thus unavoidable being tested, too. Futures studies have to take all the above factors into consideration to have more reliable results.
Finally, text types aside from narration and exposition, and question types apart from main idea and detail questions could be applied in further studies, but the text types or question types should be carefully chosen to be representative of the types. Other subtypes of detail questions besides specific-answer questions, cause-effect questions, and true-false questions could also be used for future studies. Actually,
sequence-of-event questions are quite essential for checking the subjects’
comprehension of narrative texts. Future studies had better involve sequence-of-event questions to have more representative results of narrative texts.
There are still some other suggestions for those who intend to conduct similar studies in the future. First of all, more subjects or subjects of other senior high schools could be involved in the experiment, which will bring about more representative results. Subjects of different proficiency levels also could be included, for they might have dissimilar performances under the influence of the sound. Similarly, subjects of apparently diverse ages, such as adults and children, would be a good try for further research. Children, having less experience with the print, might find oral reading beneficial. Moreover, further studies could have subjects of different L1 backgrounds, especially those subjects whose languages belong to different writing systems. The subjects, influenced by their reading habits in L1, might have different reaction to silent reading and oral reading.
