NAME : DWI RATNA INDRIANI
CLASS : TBI-6b
NIM : 3213103055
Title of the article: The Effect of Assisted RR on Fluency and Comprehension in Chinese FL Classrooms
Name of the author (s): Huifen Chen and Ding Ying
Name of Journal: The Asian EFL Journal Quarterly December 2009 Volume 11, Issue 4
Background of the study
Though there is no general consensus as to the definition of reading fluency, fluency is often referred to as learners' ability to automatically recognize an increasing number of words and phrases, which is a necessary step to comprehension of texts (Grabe, 1991; Paran, 1996; Perfetti, Van Dyke, and Hart, 2001). As reading rate and automaticity of word recognition are two most important components of reading fluency, we often use reading rate for reading fluency.
In Chinese FL classrooms, that whenever they read, many learners would read word for word slowly. As a result, they not only fail to achieve appropriate comprehension but also fail to develop interest in reading in English.
Therefore, more efficient ways of improving learners‘reading fluency need to be developed in Chinese FL classroom. While Extensive Reading (ER) which is quite popular in Chinese College English classrooms has long been thought to be a useful approach to increasing L2/FL learners' reading fluency, assisted Repeated Reading(RR) is rather a new instructional approach in FL reading contexts in China. The main objective of the present study is to find out whether and how an assisted repeated reading model enhances Chinese College English learners ‘reading fluency and consequently improve their reading comprehension.
Research Problem (s)
1. Is RR effective for developing reading fluency of Chinese college students?
2. Is RR also effective for improving reading comprehension of Chinese college students?
3. Is RR model more effective than the traditional ER model in developing reading fluency and comprehension of Chinese college students?
4. Is RR model as helpful to the weaker readers as to the stronger readers?
5. How do Chinese college students perceive the effectiveness of RR?
§ Research Design
The method of the research is experiment. The study was aimed to revealing and describing the effectiveness of assisted RR on fluency and comprehension in Chinese FL classrooms. The design of this research is a true experiment design with pre-test and post-test procedure.
The participants were 90 first-year Chinese students in Nanjing University of Finance And Economics, who were from 3 classes from the school of Food Science and Engineering. We called the 3 classes Group 1(G1), Group 2(G2) and Group 3(G3). As part of their academic program, they had four 45-minute English lessons a week designed to improve reading, writing, speaking and listening, and grammar. The participant ranged in age from 18-20 years old; in each group, there are 18, 17, 15 females and 12, 13, 15 males.
§ Research instruments
Pretest and posttest
The pretest and the post-test respectively comprise two sections, Section 1 and Section 2. Both Section 1s include: Reading Comprehension, Vocabulary and Structure, and Cloze. And Section 2s comprise one passage for fast reading as well as four passages for careful reading comprehension.
The two programs were conducted from October, 2005 to the end of May, 2006, with one-month interval of winter holidays between two semesters, totally twenty-five sessions. As part of their academic program, the freshmen had four regularly scheduled periods of English class a week (totally three hours). Every week, about one hour was spent on the program for all the groups.
Before the RR and the ER treatments began, the author introduced the program in details to the students, and some reading skills and positive reading habits, especially for fast reading were also introduced to the students of all groups. Both of the RR program and the ER program were carried out by the same teacher in the hope that the procedures are observed consistently.
§ Data Analysis
In this study, SPSS13.0 was used to analyze the data statistically, including Independent-Samples T Test, Paired-Samples T Test.
The experimental groups and the control group were chosen based on the results of a pretest, which shows that G1 and G3 were comparable, because their mean scores on the pretest total were 55.32 and 56.35 respectively, while on the pretest reading subtest (the reading comprehension parts of the test) were 56.53 and 56.80 and their WPM averages 147.29 and 159.82 respectively. To ensure that G1 and G3 participants were not statistically different at the outset of the treatments, the Independent-Samples T tests were applied to their scores on the pretest total, pretest reading subtest, as well as to their pretest WPM averages.
There were no significant differences between G1 and G3 on any of the measures at p = 0.665, 0.937, 0.288 >0.05. Therefore, we considered G1 as the experimental group, G3 the control group. The results also indicate that G2 had lower scores than the other two groups in the pretest, with the mean score of the pretest total being 51.13, the pretest reading subtest 49.50 and the WPM average 130.94. A comparison of the scores on the pretest total, pretest reading subtest and the pretest WPM between G2 and G3, we found a significant difference at p = 0.018, 0.016, 0.006 < 0.05. Therefore we regard G2 as comparatively weaker readers.
About the development of reading fluency
It can be seen from table 2 and 3 that both RR groups‘ reading speed on the initial reading of the last session (the 25th) and on the post-test is significantly higher than on the initial reading of the first session and on the pretest, which strongly supports the previous findings that RR is effective in increasing fluency gains not only within each session but also over the whole RR treatment.
Though the difference in reading speed between G2 and G3 on the post-test is not significant, but if we take their significant difference on the pretest into consideration, we feel justified to believe that the two RR groups‘fluency gains over the RR treatment are significantly greater than the control group over the ER treatment,which does not conform with Taguchi, Takayasu-Maass & Gorsuch (2004)‘s finding that RR is as effective as ER in developing reading fluency.
About the development of reading comprehension
From the pretest to the post-test, the average comprehension scores of G1 and G2 increased, and the comparison of their comprehension gains on the pre-test with those on the post-test shows great significance, which suggests that RR is effective in improving the reading comprehension of Chinese college students.
However, as our study is more conclusive than Taguchi and Gorsuch‘s concerning the RR groups‘ improvement in reading comprehension, at least based on the results of within-group analysis, it is necessary for us to look into some of the possible factors that might have led to the difference in our findings.
RR for weaker readers
With the precondition that G2 got much lower scores on the pretest WPM and comprehension than G3 did, while on the post-test the difference was not statistically significant, the effects of improving reading fluency and comprehension RR treatment has worked on the relatively weaker group is evident
About participants' view of the effectiveness of the RR treatment
As for the effectiveness of repetition, most participants believe that repetition enables them to read faster and comprehend better. Their extemporaneous comments are that reading a familiar passage not only saves their attention resources so that they can read faster within each session but also facilitates and enhances their understanding of new linguistic forms and unfamiliar content so that they are better equipped for the next session, and consequently they feel more and more motivated and confident to read.
In terms of the effectiveness of timing, the RR participants‘attitude is very contradictory. As for the input, most participants think the materials used for reading passages over the sessions deal with a variety of topics and are fit for their English level and background knowledge.
Based on the results of the present experiment, it can be concluded that an RR model can play an effective role in developing learners ‘reading fluency and comprehension in Chinese FL classrooms. Furthermore, we tend to believe that the degree of the effectiveness of RR may be influenced by the elements involved in the RR model such as the times of repetition and transfer to new passages, the kind of input and the number of sessions.
Firstly, the appropriate times of repetition and transfer may be responsible for the possibility that adequate transfers may help the readers to adapt to new passages faster and better while too many repetitions may cause the readers to feel bored. Secondly, reading is an interaction between readers and texts, so text factors such as language, rhetorical devices, text types and culture will affect readers ‘comprehension.
The present study is not mature enough in that the significant difference of the fluency and comprehension gains between the experimental groups and the control group might not entirely result from the application of different treatments; there may be other factors that we are unaware of. Besides, we do not have other efficient way to record the WPM for each reading.
In all, this study has revealed to us that the Assisted RR model is an efficient solution to our concern that many Chinese College English students cannot read fluently mainly due to poor word recognition. In addition, the RR treatment is applicable because it is convenient for the teachers to design and operate in class. And we suggest that RR treatment is best suitable for first-year College English students, who often complain about their reading speed and tend to follow teachers‘direction strictly in the treatment, which is important for the effectiveness of the training.