國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
5
CHAPTER TWO
LITERATURE REVIEW
This study aims to compare the differences of repeated reading and non-repeated reading approaches on EFL young beginners’ oral reading rate
and accuracy rate. This chapter presents relevant literature in three sections.
Section one reviews the importance of reading fluency. Section two elaborates reading fluency approaches and related studies. The last section, section three, explores related studies in L2/ EFL context.
Reading fluency
Although there is agreement toward the importance of reading fluency,
there seems to be no single definition of reading fluency. The Literacy
Dictionary (Harris & Hodges, 1995) defines reading fluency as ―freedom from word identification problems that might hinder comprehension‖ (p.85). It
interprets reading fluency as accuracy and comprehension. Meanwhile,
reading fluency is often referred to oral reading fluency. The National Reading Panel (2000) regards reading fluency as the ability to read with accuracy, speed, and proper expression. In regard with prosody, a number of studies (M.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
6
R. Kuhn & Stahl, 2003; O'Connor, White, & Swanson, 2007) have indicated that students reading with prosody usually show more comprehension on reading. Those students can segment sentences into meaningful phrases and read in word groups instead of word by word (Dowhower, 1987). In addition, as Rasinski and Hoffman (2003) defined reading fluency as the ability to read with accuracy, speed, prosody and automaticity, automaticity is specified as a component of reading fluency. Research (LaBerge & Samuels, 1974; Therrien, 2004) has shown that fast and accurate word recognition leads to automatic reading, which allows readers to save their energy for comprehending the text.
Accordingly, the importance of reading rate and accuracy has been
documented in literature. Meanwhile, due to the lack of clear evidence whether reading with prosody leads to better comprehension or better comprehension leads to better reading prosody (T. Rasinski, Rikli, & Johnston, 2009), the present study focused reading fluency on two components: oral reading rate and accuracy.
Reading automaticity theory
Reading automaticity theory was first introduced by LaBerge and Samuels (1974). They stated, ―Automaticity refers to the ability to perform complex skills
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
7
with minimal attention and conscious effort‖ (LaBerge & Samuels, 1974, p.
107). It is often compared to skill learning (LaBerge & Samuels, 1974). Take learning to swim as an example, at an initial stage. Learners can only focus on every movement trying to maintain floating in the water. After more and more practices, when the basic sub-skills are formed and performed unconsciously, learners’ attention can be gradually shifted to other higher-level skills. They
then can fluently connect all the movements effortlessly, rhythmically and automatically keeping forwarding in the water.
When automaticity is associated with reading, reading automaticity interprets the importance of reading rate. According to Samuels (1974), when one reads, the resources for processing the information is limited. One has to accomplish two tasks at once: word recognition and comprehension. If most of the attention is used for decoding and recognizing the words, there will be only little or no resource for attending the meaning. On the other hand, when the sub-skills of reading are being processed accurately, quickly, and easily, the
brain can be freed to process higher-order thinking (Timothy V Rasinski &
Hoffman, 2003). With reading automaticity, readers’ working memory will not
be packed to capacity for word recognition sub-skills and can be functioned for comprehension (Musti-Rao, Hawkins, & Barkley, 2009). For emergent readers,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
8
the process to activate the reading sub-skills such as letter-naming
correspondences and phonological processing to recognize the words is new and needs much attention. They might therefore concentrate most of their energy on decoding the prints and left only few for understanding the text.
Although readers who recognize words slowly can still shift their attention to get the meaning, the process will be struggling instead of enjoyable.
Nevertheless, for some struggling readers, even though they finally pronounce the words, they often just ―call‖ (Stanovich, 1986) the words without getting its
meaning and hence become word callers. In short, when sub-skills of word recognition are processed automatically, readers are able to focus on the meaning of the texts more.
Reading development stages and reading automaticity
In Chall’s six stages of reading development model (1983), she pointed
out that the first three stages refers to learners who are learning to read, while the last three stages refers to learns who achieve reading automaticity and are able to read to learn. It implies that automatic reading ability is the gateway to reading for learning.
The stages characterize learners from preschool to grade eight based on
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
9
their reading abilities. The first stage, stage 0, is ―logographic stage‖ when
learners recognize words by context and experiences such as pictures and logos instead of the letters (Villaume & Brabham, 2003). The second stage, stage 1, describes learners who can apply their phonological knowledge to decode words. When learns can decode and recognize words easily, they move toward the third stage, stage 2. In stage 2, learners are able to read automatically which includes reading with speed and accuracy (Samuels &
Flor, 1997). Once they can read automatically, they can make use of their energy for upper-level skills to comprehend the text. The focus of this stage is not about learning new reading skills, instead, it’s about sharpening their word
recognition skills until they become automatic (Nichols, Rupley, & Rasinski, 2009). As a result, exposing students to abundant reading is important for
stage 2 readers because it helps students read words accurately and gradually move from reading word by word to fluent reading. In this stage, learners’
focus gradually shifts from word recognition to comprehending the texts. After learners acquire automatic reading skill, they can go through stage 3 to 5 to apply their reading skills to learn more about the world as well as to create and convey their own world through reading.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
10
Reading fluency and reading comprehension
Reading comprehension, as the target goal of reading instruction, is empirically confirmed in great amount of studies to have a high correlation with reading fluency (Jenkins, Fuchs, van den Broek, Espin, & Deno, 2003).
Furthermore, studies have shown that in most research which focuses on beginning readers, the correlation between reading fluency and reading comprehension is strong (Rupley, Willson, & Nichols, 1998; Spear-Swerling &
Sternberg, 1994).
In Samuel’s Automaticity Theory (LaBerge & Samuels, 1974), they
suggested that when word recognition is automatic, the working memory can be freed to comprehend the text. Although fast and accurate word recognition
does not directly lead to perfect comprehension, slow and incorrect word recognition does impede reading comprehension. In National Reading Panel’s
review (2000), the average effect size for the ninety nine research comparing reading fluency, word knowledge, and comprehension was 0.50, which indicated that reading rate and reading comprehension is reciprocal. Related studies (Jenkins, et al., 2003; Perfetti & Hogaboam, 1975) have demonstrated that readers with good comprehension skills read words faster than those who do not perform good comprehension in reading.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
11
Oral reading fluency
Oral reading, compared with silent reading, is usually the primary reading fluency instruction method for readers in the early learning stages during these decades and has always maintained the dominant approach inside classrooms (Timothy V Rasinski & Hoffman, 2003).
In the history of oral reading fluency, the art of oral reading has been one of the focuses and has been integrated into curriculum since the middle of the 19th century (Timothy V Rasinski & Hoffman, 2003). However, during the end of the 19th century, the focus of reading instructional method shifted from oral reading to silent reading, because critics (Hyatt, 1943) at that time assumed that oral reading was a method for practicing mechanic reading skills.
In the late 20th century, a recall of the benefits of oral reading was claimed.
The advantages of oral reading instruction are that students can have feedbacks immediately and the teachers can understand their students’
reading level more clearly. Oral reading has then been combined with silent reading in reading program for checking learners’ word recognition ability.
Approaches like round robin reading is often applied in classrooms at the time
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
12
(Hoffman, 1987). Students take turns to read a short sentence or paragraph and receive feedbacks from the teacher or their peers. Hence, oral reading instruction is not only seen as a mechanical reading instruction, rather, it is seen as a diagnostic indicator of reading ability.
Studies in extensive research have shown that oral reading fluency and reading achievement have a high correlation (M. R. Kuhn & Stahl, 2003;
Melanie, Paula, Robin, Lesley Mandel, & et al., 2006; Therrien, 2004).
Consequently, oral reading fluency can be taken as an indicator of reading comprehension (Jenkins, et al., 2003). In short, we can assume that fluent oral reading can serve as a bridge to reading comprehension.
Reading fluency approaches and related studies
Oral reading approaches include reading aloud approaches such as repeated reading and other non-repeated reading approaches.
Assisted oral reading
Assisted oral reading includes reading aloud approaches which enhance
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
13
students’ reading fluency development by providing students support.
Assistance in oral reading instruction can be aural input from audio devices or guidance from teachers, peers, or parents. Approaches include echo reading, choral reading, partner reading, paired reading, cloze reading, and
listening-while-reading (Welsch, 2006).
In echo reading, students repeat the same lines after the teacher. The teachers demonstrate accurate pronunciation and proper expression for the students. Choral reading, as Heckelman’s (1969) proposal of the neurological
impress method, can be adopted in whole-class activity, small group activity or individual instruction. In whole-class or small group reading, students read along the same text in unison. In individual instruction, the teacher or
assistance presents as a reading model and work together with the students.
Students were encouraged to choose books easier for them and read together with the teacher or assistant. Partner reading, or peer-mediated reading, refers to the approach in which students work in pairs and read the passage to each
other alternatively. In partner reading, students usually provide assistance to each other and monitor each other’ reading at the same time (Musti-Rao, et al.,
2009). A related study (Yurick, Robinson, Cartledge, Lo, & Evans, 2006) was conducted to compare the effect of different groupings to the students’ oral
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
14
reading improvement. The results showed that the students with partner reading approach receiving their peer’s feedback outperformed the students
who worked as a whole-class group. Paired reading is originally designed for parents and children to work together at home. Being adapted into classrooms, it can be operated between learners with different ages and levels. When heterogeneous students are paired, both higher-level students and lower-level students will benefit from co-operative learning (Topping, 1989). Lower level students can receive assistance and feedback by their partners and acquire a more fluent reading manner from their partners. Lower level students in this case can receive demonstration and independent instruction. Results show that with well-structured pairing and material selection, both tutors and tutees
advanced their reading rates. However, pairing isolated students can be challenging. In addition, studies show that if the material is way below tutor’s
competence, the tutor can hardly gain fluent improvement. In cloze reading, the teacher reads the text with intentionally pauses and students have to fill in the blanks orally (Homa, et al., 1993). It can provide students scaffoldings of the difficulty words first. Gradually, the students can read longer phrases and build up sentences. Listening-while-reading is first being mentioned in
Chomsky’s (1978) research. Students are provided readings with audio inputs.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
15
Learners can listen to the audio inputs repeatedly according to their own need until they are comfortable to read the readings alone. In the National Reading Panels’ (2000) review of studies, assisted and guided oral reading contributed
to clear and agreeing improvements in every aspect of reading abilities, including fluency, accuracy, and comprehension.
Traditional repeated reading in L1 context
Repeated reading (RR) is a reading fluency teaching method derived from Samuels and LaBerge’s automaticity theory (1974). In order to achieve
reading automaticity, Samuels (1979) suggested that by repeatedly practicing the elements in the same short passage, students can sharpen their word
recognition skills and gradually speed up their reading rate. Fast and effortless word recognition leads to reading automaticity which enhances students’
reading comprehension improvement. Samuels (1979) argued that
comprehension can be gained through rereading because students can devote most of their attention on comprehending the text rather than on decoding and recognizing the words.
Repeated reading, as Samuels (1979) stated, is "a supplemental reading program that consists of rereading a short and meaningful passage until a satisfactory level of fluency is reached‖ (p.377). During the reading session,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
16
students are provided with a short and interesting story according to their levels. The students practice it independently and take turns to read to the assistant. While they read to the teacher or assistant, the total words they have read in one minute and the number of error words will be charted on one graph.
With the feedback, students go back to their seat to correct their own errors independently and practice the same passage repeatedly for the next record.
Once they achieve the criterion of 85 correct words per minute (wpm), they are provided with another short and meaningful passage (Samuels, 1979).
Repeated reading has been regarded as one of the most effective reading approaches because related studies revealed positive outcomes practicing RR on students regardless of ages and levels. For instance, in NRP’s (2000) review, RR is convincing on improving students’ reading rate and decreased the students’ errors. Interestingly, repeated reading leads to isolated word
recognition improvement, yet isolated word recognition dose not lead to reading fluency improvement (Therrien & Jr., 2007). Therefore, repeated
reading aids students only in reading connect words instead of word list. In Therrien’s (2004) meta-analysis, RR not only benefited students’ reading rate
and accuracy, but also improved their comprehension. Further, the improvements made after RR instruction was transferable to unpracticed
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
17
passages. In each new reading, students needed less time to achieve the reading rate criterion and decrease the miscues on each new story (Dowhower, 1987; Samuels, 1979; Therrien, 2004). Though critics (O'Connor, et al., 2007;
T. V. Rasinski, 1990) pointed out that repeated reading might be tedious and restricted, Samuels (1979) declared that students’ motivation was increased
as they observed their own growth on the recorded graph. Likewise, Amarel (1978) argued that knowing that repeated reading was beneficial for their reading fluency and comprehension, beginning readers were willing to involving the activity. In sum, through repeated reading, word recognition
abilities, reading speed, accuracy, as well as comprehension can be gained.(M.
R. Kuhn & Stahl, 2003).
Assisted repeated reading
For greater ease of use, traditional repeated reading has been later on modified and integrated with assisted oral reading approaches such as
listening-while-reading, paired reading, and choral reading to provide students scaffolding (Timothy V Rasinski & Hoffman, 2003). The benefits of assisted repeated reading are that the assistance scaffolds the students and provides them a reading model. Several studies (Begeny, Krouse, Ross, & Mitchell,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
18
2009a; Dowhower, 1987) in L1 and L2/EFL settings have revealed that assisted RR is beneficial for readers regardless of ages, levels, or learning circumstances. It can be practiced to normal learners from second graders to university students who are behind levels, or with learning disabilities, autism, or low vision. It can also be applied into a whole-classroom activity and can be applied to small-group remedial learning or individual special learning.
Chomsky (1978) conducted a research integrating listening-while-reading strategy with repeated reading in remedial instruction. Fourth graders
individually read the storybooks with audiotapes. They listened to the whole story the first time and then repeatedly listened to a section they would like to read along until they were able to independently read the section. Meanwhile, they can monitor themselves by recording their own readings (Chomsky, 1978).
Over the 10-month intervention, those remedial students gained great
progress in fluency and comprehension though they were still behind their levels. The materials were believed to be high above the learners’ level,
though the difficulty level of the materials used was not reported in the study.
Carbo (1978) replicated Chomsky’s research but carefully selected materials with difficulty levels slightly above the learners’ ability. Comparing to
Chomsky’s study, the stories were modified and the reading rate in the audio
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
19
tape was controlled. Results showed significant improvement on the struggling readers’ word recognition. Critics believed that the difficulty level of materials
was one of the key elements affecting the study (M. R. Kuhn & Stahl, 2003).
Heckelman (1969) incorporated choral reading with repeated reading.
Twenty-four middle school students worked in a remedial program for 7.25 hrs.
Students were empowered to choose the stories they liked and worked with the teacher. With the assistance, the students who were behind their levels for about three years gained 1.9 year growth in average.
Kuhn (2003) reviewed nine studies on assisted repeated reading with control groups. In her review, six studies out of nine revealed significant differences between treatment and control groups.
Comparisons of assisted and unassisted repeated reading
Eldrege (1990) conducted a study to compare repeated reading with assistance and without assistance. The intervention group read the stories along with the teacher and then repeatedly practiced the same story
themselves for several times. The control group received traditional repeated reading approach without assistance. Results showed that the intervention group with oral assistance outperformed the control group on vocabulary and
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
20
comprehension.
Dowhower (1987) compared unassisted repeated reading and assisted repeated reading on 15 students. They focused four elements: reading rate, accuracy, comprehension, and prosody. The students were randomly assigned into two groups. Students in the assisted repeated reading group listened to the tapes while they were reading the basil readers, the textbooks used in schools for enhancing students’ reading abilities, without any assistance until
they were confident to read along without assistance. The unassisted repeated reading group reread the basil readers until they past the criteria of reading rate. Both assisted and unassisted repeated reading improved students’
reading rate, accuracy, comprehension and prosody gained improvements on practiced and unpracticed readings. However, one thing worth noting is that the students in assisted repeated reading outperformed the students in unassisted repeated reading on prosody. Students in assisted
listening-while-reading group read the reading with less improper phrases and with better intonation than students in the unassisted repeated readings (M. R.
Kuhn & Stahl, 2003).
In succession to Dowhower, Rasinski (1990) conducted a similar study to compare assisted and unassisted repeated readings. After the intervention, 20
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
21
third graders in both groups performed significantly better on reading rate and accuracy. However, there was no significance between the two treatments.
Prosody was not administered. Rasinski (1990) concluded that both assisted or unassisted approaches aid learners reading rate and reading accuracy.
With the audiotapes, the teacher can provide learners assistance easily.
Therefore, assisted repeated reading is suggested due to that it is easier to implement listening-while-reading rather than traditional repeated reading approach without assistance.
Comparison of repeated reading and non-repeated reading
Several studies (Homa, et al., 1993; M. Kuhn, 2004; O'Connor, et al., 2007)
has been conducted to compare the effect of RR and Non-RR on improving L1 and L2 /EFL students’ oral reading fluency. Though repeated reading has been
has been conducted to compare the effect of RR and Non-RR on improving L1 and L2 /EFL students’ oral reading fluency. Though repeated reading has been