numerous opportunities to practice reading text at their instructional level ..... daily goal for students reading at grade level was set similar to the end-of-year goal ...
Effects of Repeated Readings on the Oral Reading Fluency of Urban Fourth-Grade Students: Implications for Practice Shobana Musti-Rao, Renee O. Hawkins, and Elizabeth A. Barkley
ABSTRACT: Oral reading fluency, which includes the effortless and smooth oral production of text, is an essential part of deriving meaning from text. Fluency is a particularly important skill in upper elementary school grades when students are exposed to higher level reading materials. The authors evaluated the effects of a repeated readings intervention with urban 4th-grade students who showed at-risk markers for reading failure. The authors trained the classroom teacher to implement repeated readings classwide and collected treatment fidelity data on the extent to which the teacher adhered to the procedures. Results indicated that peer-mediated repeated reading improved students’ oral reading rate across baseline levels; however, by the end of the study, the students did not reach benchmark goals in reading. Integrity data suggested that the teacher was able to implement the intervention with a high degree of fidelity classwide. The authors discuss important implications for practice and directions for future research.
offering less advanced instructional material. For example, when students are grouped by ability levels, the lower performing, struggling readers only have other poor readers to serve as models. There is merit in Stanovich’s (1986) biblical analogy as it relates to reading (i.e., the Matthew effect). Students who begin schooling with rich literacy experiences and receive effective instruction are more likely to become good readers than are those who enter school while lacking meaningful literacy experiences and are provided with insufficient experiences. If students are to succeed, then schools should make the teaching of reading a priority in the early grades. There is general consensus among researchers and practitioners on the components that constitute effective reading instruction (i.e., phonemic awareness, alphabetic principle, fluency, vocabulary, comprehension; National Reading Panel [NRP], 2000). As students transition from the learning-to-read stage to the reading-to-learn stage during the upper elementary school years, the need for explicit instruction in reading fluency and comprehension cannot be overstated.
KEYWORDS: reading fluency, repeated readings, treatment integrity THE READING REPORT CARD of fourth-grade students in the United States depicts a failing grade. The problem is particularly acute in urban schools in which 60% of African American fourth-grade students read below basic levels (Cartledge & Lo, 2006). Students from low socioeconomic and culturally diverse backgrounds are less likely to speak standard English (Talbert-Johnson, 2004), and many of these children begin school with insufficient preliteracy and oral language skills in comparison with their more affluent peers (Musti-Rao & Cartledge, 2007). The intersection of language and literacy dramatically increases the probability of urban students having difficulty with reading, and, in turn, performing below grade level (Klingner & Artiles, 2006). Further, these students are disproportionately referred for special education services and classified as having a learning disability because of their poor reading skills (Hitchcock, Prater, & Dowrick, 2004). Despite the urgency to prevent reading failure and provide urban learners with quality reading instruction, schools exacerbate the problem by placing students in lower tracked curriculum classes and
Fluency Instruction The NRP (2000) defined fluency as the ability to read with accuracy, speed, and proper expression. Fluency is synonymous with quick and effortless reading of text—a skill that struggling and dysfluent readers lack. Automaticity theorists (e.g., LaBerge & Samuels, 1974) contend that if more attention is spent on decoding each word, less attention Preparation of this article was supported by a grant from the University of Cincinnati University Research Council Faculty Research Support Program. Address correspondence to Renee O. Hawkins, University of Cincinnati, College of Education, Criminal Justice and Human Services, PO Box 210068, Cincinnati, OH 45221, USA; renee.hawkins@uc .edu (e-mail). Copyright © 2009 Heldref Publications 12
Fall 2009
is available for comprehension (Kuhn, 2005). Therefore, automatic decoding of text is an essential attribute of fluent readers. The automaticity theorists maintain that this effortless decoding of text allows readers to focus on the meaning of the text and hence aids in comprehension. There is agreement that one of the ways to transition struggling readers to becoming fluent readers is to provide them with numerous opportunities to practice reading text at their instructional level (Kuhn; Kuhn et al., 2006). Despite the documented link between fluency and comprehension, explicit fluency training is a missing component and a forgotten goal in many classrooms today (Cartledge & Lo, 2006). A research-validated approach most often used to improve reading fluency is repeating readings (Kuhn; Kuhn & Stahl, 2003). Repeated readings require students to reread a passage of text several times for a particular period of time (e.g., 10 min) or until a fluency criterion is met (e.g., 100 words per minute [wpm]). In a recent study, Kuhn (2005) evaluated the effectiveness of repeated reading and nonrepetitive reading for secondgrade students in a small group setting. Students were randomly assigned to one of three conditions: repeated reading, wide reading, or listening only. The study also included a control group that did not receive any intervention outside of the regular language arts curriculum. The repeated reading condition included modeling, repetition, and positive feedback from adults and peers, whereas the wide reading condition included choral reading. Students in the listening only condition listened to text read by the researcher. The intervention took place for 18 sessions across 6 weeks. Students received instruction three times per week for 15–20 min each. Kuhn reported that students in the repeated reading and wide reading groups showed more substantial gains in word recognition and fluency than did students in the listening only or control groups. Repeated readings can be implemented in several formats. One format is the small group approach in which groups of students take turns at reading a passage (Kuhn). Another format is the peer-mediated approach in which students work in pairs and read a passage of text repeatedly (Yurick, Robinson, Cartledge, Lo, & Evans, 2006). Peer-Mediated Repeated Readings The effectiveness of using a peer-mediated approach to building oral reading fluency by using repeated reading is documented in the literature. Strong, Wehby, Falk, and Lane (2004) studied the effects of a repeated reading intervention used in conjunction with a structured reading intervention on six middle school students’ (four were African American, two were Caucasian; M age = 13.0 years, SD age = 0.9 years) oral reading rates and comprehension. Using a multiplebaseline-across-subjects design, a research assistant conducted repeated reading sessions with student pairs in the
Musti-Rao, Hawkins, & Barkley
13
library. The students chorally read a passage two times with the research assistant and then alternated reading the passages to each other. Results indicated that the repeated reading resulted in an increase in oral reading rates for four of the six students; however, the oral reading rates that the target students showed by the end of the study were below grade-level expectations. Yurick et al. (2006) conducted three experiments to evaluate the effects of peer-mediated repeated reading on the oral reading fluency and comprehension of urban learners. Eight fifth-grade students (six were African American, two were Caucasian), ranging in age from 10 to 11 years, participated in paired repeated reading for 10 min. The performance of students in the repeated reading condition was compared with that of the sustained silent reading condition. Results indicated that the target students demonstrated a mean increase of 68 wpm with repeated reading over the sustained silent reading condition. All but one of the students mastered the sixth-grade fluency criterion and moved to more advanced reading materials during the course of the study. In the second experiment, repeated reading was implemented on a classwide basis; however, data were collected on eight third-grade students (M age = 8 years, 9 months; range = 8 years, 4 months to 9 years, 5 months) who displayed reading difficulties. Of these target students, five were African American, and three were Caucasian. In addition to recording the reading rate on practiced passages, the students were given an unpracticed passage each week as a measure of generalization. The data showed that the students’ performance on unpracticed passages was higher than their performance during baseline (78.5 wpm vs. 58.7 wpm, respectively), but none of the students met the fluency criterion of 145 wpm on the unpracticed passages. The third experiment involved six fourth-grade target students (five were African American, one was Hispanic American) with a mean age of 10 years, 2 months. The students were pulled out of their classroom for the repeated reading sessions. Generalization of increased fluency rates as a result of repeated reading on unpracticed passages was evaluated, comparing the following: covertly timed, overtly timed, and overtly timed with a charting component. Yurick et al. reported that the greatest improvements were when students were overtly timed and were required to chart their performance. Collectively, the results of the aforementioned studies point to the effectiveness of repeated readings in increasing students’ oral reading rates. The Gap Despite the apparent need for explicit reading interventions in the upper elementary grades and the documented positive effects of repeated readings in increasing students’ oral reading rates, a common complaint in education is that teacher educators conduct most research with limited
14
Preventing School Failure
Vol. 54, No. 1
collaboration with classroom teachers (Greenwood & Maheady, 1997). This has not only widened the researchto-practice gap, but has also resulted in the limited sustainability of effective instructional strategies at the completion of a research study. With the increased variability in instructional practices in urban schools (Foorman, 2007), further research on the link between teacher instructional practices and student outcomes is needed. This is particularly true with the No Child Left Behind Act’s (2001) emphasis on teacher accountability and need for teachers to use research-validated strategies in their classrooms. Thus, theoretical and practical questions on how to best implement fluency instruction in classrooms remain unanswered. Similarly, although Yurick et al.’s (2006) study has shown that repeated reading is effective with urban learners, questions remain regarding the classroom teacher’s skill for implementing such a strategy in the classroom with a high degree of fidelity. The present study extends the research on repeated readings by looking at student outcomes and teacher implementation. The purpose of the study was to determine the effects of peer-mediated repeated reading on the oral reading fluency of urban fourth-grade students. We measured students’ performance on unpracticed passages on a weekly basis to determine whether gains in reading fluency during repeated reading practice would generalize to unfamiliar reading. In addition, we asked to what degree of integrity can classroom teachers implement repeated readings in their classroom. The researchers trained the classroom teacher to
implement repeated reading on a classwide basis, and treatment fidelity data were collected on the extent to which the teacher adhered to the procedures. Method Participants Students. Participants were 12 African American students ranging in age from 9 years, 3 months to 12 years, 5 months (M = 10.6 years, SD = 0.9 years) from a classroom of 32 fourth-grade students who were identified as target students for the study. Although the entire class participated in classwide repeating readings, we collected periodic progressmonitoring data only on the 12 students we selected. Of these target students, 5 were boys, and 7 were girls. A total of 6 students (Students 3, 4, 7, 8, 9, and 12) were identified as having a disability and received special education services in the resource room. Table 1 presents demographic information for target students. In all, 3 students (Students 4, 9, and 10) had repeated a grade in school, and Student 1 repeated Grades 3 and 4. We selected the students on the basis of the results of screening for at-risk markers in reading on the Dynamic Indicators of Basic Early Literacy Skills (DIBELS; Good & Kaminski, 2002) instrument and on the basis of teacher nominations. The report generated by the DIBELS data system indicated that all 12 target students needed intensive instruction to reach benchmark goals for fourth grade. Teacher. The fourth-grade teacher was a 57-year-old White woman, “Ms. C.,” with 35 years of experience in teaching
TABLE 1. Demographic Information for Target Students, With Fall and Spring Benchmark Assessment Scores DIBELS benchmark scores Student number 1 2 3 4 5 6 7 8 9 10 11 12
Age (year, month)
Gender
Disability
11, 9 9, 11 9, 3 12, 5 10, 10 11, 1 10, 11 10, 2 11, 4 11, 1 10, 4 10, 3
Female Male Female Female Female Male Male Male Female Female Female Male
None None OHI CD None None SLD SLD SLD None None SED
Number of grade retentions Raw score 2 — — 1 — — — — 1 1 — —
57 54 26 19 69 55 35 46 60 65 67 9
Fall
Spring Status
Raw score
Status
At risk At risk At risk At risk At risk At risk At risk At risk At risk At risk At risk At risk
122 — 32 32 99 59 55 72 73 98 92 18
Low risk — At risk At risk Some risk At risk At risk At risk At risk Some risk At risk At risk
Note. All students were African American. Raw score and status were not reported for the spring semester for Student 2 because he was absent during the last few weeks of school when we administered the benchmark assessments. OHI = other health impaired; CD = cognitive disability; DIBELS = Dynamic Indicators of Basic Early Literacy Skills (R. H. Good & R. A. Kaminski, 2002); SLD = specific learning disability; SED = severe emotional disability.
Fall 2009
elementary and middle school students and was in her 4th year of teaching in this school. Ms. C. was a National Board Certified teacher and volunteered to participate in the study. Setting We conducted the present study in an urban charter school in the U.S. Midwest. The school had an enrollment of 605 students from Kindergarten to Grade 9. Regarding the race of the 605 students, 581 (96%) reported that they were African American, 8 (1.3%) reported Caucasian, 2 (0.3%) reported Hispanic, and 14 (2.4%) reported multiracial. A majority (86.5%) of the students qualified to receive free lunch at school, 6.4% qualified to receive reducedprice lunch, and 7.1% paid full-price lunches. All students in the fourth grade were scheduled to engage in silent reading after they returned from lunch and recess. The silent reading session took place from 12:40 to 1:10 p.m. every day. On 3 of these days, students were engaged in repeated reading during the course of the study. During the first two tiers of the multiple baseline, the researchers pulled the target students out of their classroom during repeated reading sessions while the rest of the students in the class read silently. Target students were taken to an unoccupied classroom in the school building. Once we introduced intervention to the third tier of students, the study took place in the fourth-grade general education classroom, and the teacher implemented procedures. This fourth-grade class was the lower achieving class of the two fourth-grade classes in the school. According to the results of diagnostic testing conducted by the school, students in this class were considered to be performing below grade level and were placed on an accelerated track. Dependent Variable In this study, we focused on students’ oral reading fluency— correct words per minute (cwpm)—as the primary dependent variable. Students were administered fourth-grade-level passages from the DIBELS oral reading fluency (DORF; Good & Kaminski, 2002). These data served as progressmonitoring data for students throughout the study. Good, Gruba, and Kaminski (2001) indicated that the alternateform reliability of the DIBELS measured as adequate, ranging from .65 to .93. The predictive validity correlations with outcomes 1 year later ranged from .36 to .82 in a series of studies conducted to evaluate the reliability, validity, and sensitivity of the DIBELS measures. Further, the test–retest reliabilities of elementary students ranged from .92 to .97. The alternate-form reliability of different reading passages drawn from the same level ranged from .89 to .94. Experimental Design We used a multiple-baseline-across-participants design to evaluate the effects of repeated reading sessions on the
Musti-Rao, Hawkins, & Barkley
15
oral reading fluency of students. The advantage of using this design is that it does not require withdrawal of treatment and can be used for more than one student needing the same intervention (Cooper, Heron, & Heward, 2007). For the purpose of the present study, a three-tiered multiple baseline with two conditions was in effect. Experimental Conditions and Procedures Baseline. During this condition, all students were engaged in sustained silent reading. The students were allowed to select a book from their classroom or from their school library and engaged in independent reading activity. At the end of the 30 min, the researcher administered a DORF progress-monitoring passage to students individually. We assessed the target students once per week and the nontarget students in the class once per month. Student training. Before peer-mediated repeated reading procedures began, the first two pairs of students (i.e., Students 1, 2, 3, and 4) were trained at one time during two 15-min sessions. The first training session consisted of an overview discussion of the entire process followed by modeling of the correction procedure and the role of the reader and listener. The second training session consisted of independent student practice with feedback. Students then entered the paired repeated reading condition. The same training was provided to the next two pairs of students to begin paired reading procedures. Teacher training. Before the classwide application of the paired repeated reading procedures, the researcher explained the intervention session format to the fourth-grade teacher. The teacher was given a script that included the various steps involved and specific instructions to read aloud to the students. The researcher modeled the steps by conducting the first classwide session, which included training of the class. The teacher was allowed to conduct the rest of the sessions independently, with periodic feedback and coaching from the research team. Paired repeated readings. The first group of students to be introduced to the repeated readings intervention comprised Students 1, 2, 3, and 4. Whereas Students 1 and 2 worked well as partners, Students 3 and 4 struggled to correct their partner and read the fourth grade-level passage at the instructional level. Therefore, Students 3 and 4 were given third- and second-grade-level passages, respectively, and were paired with two of their classmates who were reading at grade level. The repeated readings sessions for Students 3 and 4 did not include the reciprocal reading. The second group (Students 5, 6, 7, and 8) and third group (Students 9, 10, 11, and 12) were introduced to the repeated readings condition in a staggered manner approximately 4 weeks apart. When the third group of students entered the
16
Preventing School Failure
intervention condition, the repeated reading sessions were moved to the fourth-grade classroom where all students engaged in classwide repeated reading. All students were assigned a 120–150-word passage from their grade-level language arts book. During classwide repeating reading sessions, students read 120–150-word passages from the book Charlotte’s Web (White, 1952). The daily goal for students reading at grade level was set similar to the end-of-year goal as indicated in DIBELS (i.e., 118 wpm). As previously mentioned, because of their low levels (i.e., below grade level) of reading, we made an exception to the reading passages for Students 3, 4, and 12. These 3 students were given passages at the second and third gradelevels from the Aimsweb series.1 (The daily reading goal for Students 3, 4, and 12 was similar to the end-of-year goal for their reading grade level.) For example, if a student was reading passages at the second-grade level, then the daily reading goal was 90 wpm. Similarly, the daily reading goal for a student reading at the third-grade level was 110 wpm. If Students 3, 4, and 12 met goal on three consecutive reading (practiced) passages, then they were given the next grade-level passage. Although these students were practicing reading with below grade-level passages, they were assigned weekly passages at the fourth-grade level. Each student had a repeated reading folder that included a copy of the reading passage, exam copy of the reading passage with the word count indicated at the end of each line, good tutor card, correction card, and reading log. The good tutor card was part of a reinforcement component in which students earned stamps for good tutoring behaviors. Each student was provided with a 4 × 6-in. index card labeled good tutor card. Students placed the good tutor card on their desk during repeated reading so it would be easily accessible to the teacher. As students were engaged in tutoring, the teacher walked around the room and stamped the good tutor cards of students who demonstrated appropriate tutoring behavior. In their reading log, students recorded the number of stamps they had received during each session. The purpose of the reading log was to record the date, number of words read correctly at the end of the session, and number of stamps the students received during repeated reading sessions. At the beginning of each session, students sat across from one another and placed their reading passage, good tutor card, and correction card in front of them. The students took turns at reading each paragraph of the passage for 10 min. The students were instructed to reread the passage for the 10-min duration. The correction procedure used in this study was similar to the procedure used in Yurick et al.’s (2006) study. At the end of 10 min of repeated readings, students were asked to individually read the practiced passage for 1 min. Then, the student recorded in the reading log the number of words read correctly. The reading goal
Vol. 54, No. 1
for the fourth grade was set at 118 wpm. If a student met the goal on the third day of repeated readings, he or she was given a tangible reward (e.g., mechanical pencil, gel pen, key chain). Data Analysis In addition to using the visual analysis method (Kazdin, 1982), we analyzed data in terms of individual effect sizes and percentage change (Gresham, 2005). As Parker et al. (2005) recommended, we used effect size estimates descriptively rather than inferentially in the present study. We calculated effect sizes using the following formula: d = (MI – MB) / SDp, where MI is the mean score for intervention, MB is the mean score for baseline, and SDp is the pooled standard deviation for the baseline and intervention conditions. According to Cohen’s (1988) guidelines, .20 is a small effect size (d), .50 is a medium effect size (d), and .80 is a large effect size (d). We calculated percentage change for each student by comparing the mean level of performance during baseline with the mean level of performance during intervention. We also used the dual-criteria method (Fisher, Kelley, & Lomas, 2003) to increase the accuracy of interpreting the data. We drew baseline mean and a second criterion line on the basis of the baseline mean for each target student’s data. On the basis of the dual-criteria method, a particular number of data points in the treatment phase must be greater than both criterion lines to conclude that there has been a reliable treatment effect. Interobserver Agreement A second observer independently scored the administration of the DIBELS probes to obtain interscorer reliability. The percentage of interobserver agreement was calculated by dividing the number of agreements between the two observers with the total number of agreements and disagreements multiplied by 100. We collected interobserver agreement data on 15.9% of probes (27 of 170). There was a mean agreement of 97.4% (SD = 3.0%; range = 88–100%) between the two observers. Social Validation At the completion of the study, the fourth-grade classroom teacher was asked to complete the Teacher PostIntervention Acceptability and Importance of Effects Survey (Lane & Beebe-Frankenberger, 2004). The researcher interviewed the students regarding their experience with the present study. The questionnaires assessed the consumer’s opinion about the efficacy of the intervention. The questionnaires also solicited consumers’ opinions about specific components of the intervention that they liked or disliked and whether they would recommend the intervention for future instruction.
Fall 2009
Results CWPM Figures 1 and 2 show the number of cwpm on the progressmonitoring probes during baseline and intervention conditions. The graphs include the baseline mean and trend line for each student’s data. It should be noted that the students were unfamiliar with the progress-monitoring passages. We used these probes as a measure of fluency generalization from practiced passages. Table 2 shows the mean number of cwpm for each student along with individual effect size estimates and percentage change in performance (reading fluency) from baseline to intervention. During baseline, all students—with the exception of Students 1 and 9—showed stable responding with no trends. With the introduction of the repeated reading intervention, there was an increase in the fluency rates of all students. Except on one probe in which Student 1 read 126 wpm (data point 11), none of the students met the benchmark goal of 118 wpm. Among the 3 students (Students 3, 4, and 12) reading passages at below grade level, 2 students (Students 3 and 4) advanced one grade in reading level. Student 3 started with passages at the third-grade level and after meeting the goal of 110 wpm on three different passages, progressed to reading passages at the fourth-grade level during the classwide repeated readings sessions. Student 4 started with the passages on the second-grade level and by the end of the study read 110 wpm on three passages at the third-grade level. Students 3 and 4 were engaged in repeated readings sessions for 17 weeks of the study. Students 5 and 7 made gradual yet steady progress with the introduction of repeated readings. Student 12, who received intervention for only 6 weeks, reached goal on three second grade-level passages and ended the study reading at the third-grade level. Overall, the intervention data indicated a mean percentage change of 39.8% (SD = 29.6%; range = 9.7–107.6%) for the group. Effect size estimates indicated that the intervention had a moderate to large effect on students’ oral reading fluency. According to the dual-criteria method, only 4 students’ (Students 3, 4, 5, and 7) data met criterion to indicate a reliable treatment effect. That is, the treatment-phase data points for these 4 students were higher than the two criterion lines. As shown in Table 1 before intervention, all 12 target students showed at-risk markers for reading. At the end of the study, 8 (66.7%) students continued to show at-risk markers, 2 (16.7%) showed some risk, and 1 (8.3%) showed low risk for reading failure. Data were not available on 1 student (8.3%) because of absenteeism at the time of assessment administration. Procedural Integrity We used treatment integrity checklists to evaluate the accuracy of students’ behavior during repeated readings sessions
Musti-Rao, Hawkins, & Barkley
17
and of the teacher’s behavior in implementing the steps of the intervention during classwide repeated reading sessions. Student implementation. At the student level, the researchers developed a checklist containing the steps involved in the implementation of the repeated reading sessions to record whether the intervention has been implemented as intended. Students were evaluated for reader and listener behaviors. The researcher observed each student pair for 3 min and recorded whether each of the steps was present and applied correctly by both students. Also, the researchers recorded whether the steps listed were being followed. Fidelity was measured by dividing the number of steps performed correctly by the total number of required steps multiplied by 100. Integrity checks were completed on 29 students during 7 of 48 (14.6%) repeated reading sessions. Results indicated a mean compliance of 93.1% (SD = 13.2%, range = 50–100%). Teacher implementation. Before starting the classwide repeated reading sessions, we provided the teacher with a script for conducting the sessions in the classroom. The teacher was also provided with a copy of the integrity checklist that listed eight implementer behaviors. Data were collected on the accuracy with which the teacher completed the eight listed steps, which included providing students with directions to start repeated reading, methods of error correction, and student feedback. We collected integrity data on 9 of 17 classwide repeated reading sessions (i.e., 52.9%) over a 6-week period. The teacher implemented the sessions with a mean accuracy of 98.6% (SD = 4.2%, range = 87.5–100.0%). In one of the checks conducted during the 1st week of classwide repeated reading, the researcher recorded that the teacher did not provide students with feedback while observing students read. Social Validity On a 7-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree), the fourth-grade teacher strongly agreed that the repeated reading intervention “supplemented classroom instruction, was easy to implement, will have lasting positive effects, and improved students’ overall performance.” The teacher indicated that this is an intervention she would use again and recommend to others. The teacher shared that she particularly liked the partnerreading format and was able to run peer-mediated repeated readings in science and social studies classes. In addition, the teacher wondered how fluency would transfer to social studies, but social studies was discontinued in lieu of more reading and math instruction. The teacher also shared that she wished the program started at the beginning of the year with a “more balanced focus on fluency and comprehension.” All target students indicated that they liked the repeated reading sessions a lot and agreed that they were learning
Preventing School Failure
Vol. 54, No. 1
Baseline
Correct Words Per Minute
18
120 100 80 60 40 20 0 120 100 80 60 40 20 0 120 100 80 60 40 20 0 120 100 80 60 40 20 0 120 100 80 60 40 20 0 120 100 80 60 40 20 0
Repeated reading intervention
Student 1
Student 2
Student 7
Student 8
Student 11
Student 12 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 Probes
FIGURE 1. Multiple baseline data on oral reading fluency (R. H. Good & R. A. Kaminski, 2002) measures, with baseline mean and baseline trend lines for Students 1, 2, 7, 8, 11, and 12.
Fall 2009
Musti-Rao, Hawkins, & Barkley
Baseline
Correct Words Per Minute
120 100 80 60 40 20 0 120 100 80 60 40 20 0
Repeated reading intervention
Student 3
Student 4
120 100 80 60 40 20 0
Student 5
120 100 80 60 40 20 0
Student 6
120 100 80 60 40 20 0 120 100 80 60 40 20 0
Student 9
Student 10 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 Probes
FIGURE 2. Multiple baseline data on oral reading fluency (R. H. Good & R. A. Kaminski, 2002) measures, with baseline mean and baseline trend lines for Students 3, 4, 5, 6, 9, and 10.
19
20
Preventing School Failure
Vol. 54, No. 1
TABLE 2. Student Baseline and Intervention Means, Effect Sizes for Measures, Percentages of Change, and Reliable Treatment Effects Using Dual-Criteria Method
Student number 1 2 3 4 5 6 7 8 9 10 11 12 Total
Baseline
Intervention
Number of weeks
M
SD
M
SD
d
17 17 17 17 12 12 12 12 6 6 6 6 —
73.0 55.0 24.7 13.3 62.2 51.3 42.7 37.0 55.4 64.0 62.2 8.8 45.8
19.7 1.0 1.2 5.5 8.4 5.7 8.9 6.4 13.8 9.4 7.5 3.5 —
104.0 67.8 34.8 24.9 80.0 63.7 64.6 47.7 60.8 77.6 70.2 18.2 59.5
11.9 11.8 10.2 6.9 12.7 11.2 9.1 8.7 7.5 12.9 4.5 5.7 —
1.9 1.5 1.4 1.9 1.7 1.4 2.5 1.4 0.5 1.2 1.3 2.0 1.5
Change (%) 42.1 23.3 40.9 86.4 28.7 24.1 51.5 28.9 9.7 21.3 12.8 107.6 39.8
Reliable treatment effect No No Yes Yes Yes No Yes No No No No No —
Note. Effect size is for the DIBELS (Dynamic Indicators of Basic Early Literacy Skills Oral Reading Fluency; R. H. Good & R. A. Kaminski, 2002).
important skills. For example, Student 1 commented that she learned “how to read better and how to correct [her] mistakes.” All students indicated that they use their reading skills in other settings such as at home and with other reading assignments. Student 7 shared that he used his reading skills “mostly in reading and [took] more time to figure out words that [he didn’t] know.” All students indicated that they wished the intervention lasted longer, possibly throughout the school year. Discussion We evaluated the effects of a peer-mediated repeated reading intervention on the oral reading skills among 12 urban, fourth-grade students. We used a multiple-baseline-acrossparticipants design in which we introduced 4 students at a time to intervention in a staggered manner, resulting in three tiers. At the end of study, all of the students showed increases in oral reading rate with repeated readings compared with the silent reading (baseline) condition. Although the students showed increases in fluency rates during repeated reading sessions with practice passages, none of the students met end-of-year goals on the spring benchmark assessments for fourth grade. Using the dual-criteria method to interpret the single-case (A-B) design, we found that the data indicated reliable treatment effect for only 4 of the 12 target students. Data collected at the end of each repeated reading session indicated that students were able to meet goal (i.e., 118 wpm) on practice passages; however, this rate did not transfer to unfamiliar passages given during the weekly probe. This limited transfer may be attributed, in part, to the time spent in practicing reading. Students in this study spent 30 min
per week in repeated readings. The time spent may not be sufficient enough for these students to catch up to gradelevel reading. The end-of-year benchmark scores showed improved status for only 3 of the 12 target students. These gains indicated some promise in comparison with nontarget students who continued to show at-risk status without any intervention. Despite the limited gains made by students, results from this study suggest that repeated readings promote reading fluency rates among struggling readers. Although the NRP’s (2000) definition of fluency included accuracy, speed, and proper expression, in the present study, we focused only on accuracy and speed. Kuhn (2005) contended that there is more to fluent reading than only speed and accuracy. Kuhn stated that to “make for an expressive and meaningful rendering of a text” (p. 128), prosody that includes features such as stress, pitch, and intonation is needed. We believe that in addition to providing students with multiple exposures to print, repeated readings inadvertently address these prosodic features to some degree. This is particularly true in classwide implementation of the strategy in which readers range from proficient to poor in skill level. Whereas the poor readers benefit more in terms of automaticity with word recognition and speed in reading, the proficient readers in the class may benefit in developing prosody. In the present study, 42% (9 of 21) of fourth-grade students showed an at-risk status at the end of the study. Instructional practices and grouping patterns in schools contribute to the reading development (or lack thereof) of struggling readers. In the present study, 6 students who had special needs did not receive systematic instruction in oral
Fall 2009
reading before the implementation of the repeated readings intervention. At the beginning of the study, because of their poor reading ability, Students 3 and 4 were unable to catch and correct each other’s errors. They needed to be paired with two other proficient readers in the class who served as good models and were able to provide immediate error correction. Such student pairing is not possible in a resource room setting in which students have only other poor readers with whom to work. Peer-mediated repeated readings in an inclusive setting allows for flexible and more meaningful pairing of students so that they can derive the maximum benefit from this strategy. Considering the increased accountability placed on schools to perform, teachers are faced with a challenge to not only increase students’ scores on proficiency tests but also to teach a curriculum that is higher than what most students can comprehend. For example, although the teacher in this study found merit in the repeated readings and wanted to implement this strategy on a daily basis, she needed to use the time to complete an unfinished lesson or make up quizzes and other assessments. As students advance to higher grade levels, the amount of information to be learned in subject areas and the level of reading will only increase. Although reading fluently is not the sole purpose of reading, higher order reading and comprehension cannot be developed without a strong foundation of accurate and efficient word recognition. Repeated readings aid in building this foundation, and it certainly helped students in this study. As the high procedural-integrity data indicated, repeated readings is an intervention that teachers across different subject areas can easily implement. Using a script and coaching from us, the teacher in the present study was able to perform the steps of the intervention with relative ease. The teacher was able to walk around the classroom, observe students read, and provide reinforcement for good tutoring behavior and good reading behavior. One of the major and more obvious limitations of this study is the lack of focus on comprehension. With the focus on improving students’ oral reading rates, students in this study were not assessed for comprehension. Moreover, with only 10 min allotted for repeated reading intervention, the teacher could not afford more time to conduct comprehension assessments. Any improvements seen in comprehension as a result of repeated readings would simply confirm the assertion made by automaticity theorists that as students become fluent readers, more attention is available for extracting meaning from the text. However, there is a body of research that suggests that repeated readings is a strategy that does not produce gains in comprehension commensurate with gains seen in fluency (Kuhn, 2005; O’Shea, Sindelar, & O’Shea, 1985, 1987). These researchers have suggested that more explicit instruction in focusing students’ attention to text meaning is required to aid in com-
Musti-Rao, Hawkins, & Barkley
21
prehension development. In the future, researchers should include the use of explicit comprehension-building strategies in addition to fluency building strategies (e.g., repeated readings). Although the repeated reading sessions included an error-correction component in which students followed a script to correct their partners, data were not collected on the number of errors made during the weekly reading probes. In their study, Daane, Campbell, Grigg, Goodman, and Oranje (2005) found that a greater number of errors in reading resulted in lower comprehension of the text. Experts have suggested that at least a 95% level of accuracy in word recognition is needed for students to derive meaning from the text (Daane et al.; Kuhn & Stahl, 2003). In future studies, researchers should also analyze the number and type of errors that students make and how these errors may affect their interpretation and the meaning of text. If investigating the effectiveness of repeated readings in subjects such as social studies and science, researchers should also include a vocabulary-instruction component. The performance of students in urban school districts across the United States, the apparent lack of effective instruction that meets the unique needs of this culturally and linguistically diverse population, and the mismatch between instruction provided and students’ needs have become common topics in the education literature today. There is general consensus that to prevent reading failure, schools must intervene as early as possible with explicit, systematic, and intensive reading instruction. Research has supported the use of repeating readings as an effective instruction for building oral reading ability among struggling readers. However, there are several implications for teachers and practitioners as they use repeated readings in their classroom. Implications for Practice Reading time. Several schools allocate time specifically for reading, commonly called sustained silent reading or DEAR (drop everything and read) time, which predominantly entails silent reading. Although silent reading allows students to form a habit and love for reading, if the goal is to improve students’ oral reading rate, then more time for overt reading activities must be established. Teachers can allocate 10–20 min for repeated readings, with additional time for vocabulary- and comprehension-building activities. Grouping practices. Repeated readings can be implemented with a variety of grouping formats in the classroom. Depending on student need, the teacher can divide the class into two groups. While the teacher is instructing one group of students, the other group of students can be involved in repeated readings with the teachers’ option of moving groups. Grouping practices can allow for individualized progress monitoring and decision making. There is a great degree of day-to-day
22
Preventing School Failure
variability in the classroom. For example, some students may reach the goal or criterion sooner than their partners. Some students may be absent for one or more sessions. Such variability is common in classrooms and was present in our study. The teacher in our study had assigned students to their partners but had to make changes on a daily basis. Because on some days there were an odd number of students, one group included three students. Flexible grouping practices would allow students to progress at their own pace and practice reading at their instructional level. Reading model. Some research in the literature has suggested that providing students with a model of fluent reading before independent reading is a strategy that builds fluency (e.g., Cole & Lionetti, 2004; Daly & Martens, 1994; Gilbert, Williams, & McLaughlin, 1996; Rasinski, 1990; Skinner, Cooper, & Cole, 1997). Researchers have used the terms listening while reading (e.g., Rasinski), listening previewing (e.g., Rose, 1984), assisted reading (e.g., Hoskisson & Krohm, 1974), and echo reading (e.g., Homan, Klesius, & Hite, 1993) to describe procedures in which students are first instructed to read along silently while another student provides a fluent reading of a passage. Students then read the passage aloud. When the reading content is the same across students (e.g., in subjects such as social studies, science, and language arts), teachers can model fluent reading before repeated readings. With varying reading levels and limited teacher resources, audiotaped or computer-generated models can be a viable solution. Monitoring progress. One of the ways to measure students’ progress in oral reading ability is to measure the cwpm. In a classroom with 25–30 students, the teacher can make monitoring of progress more manageable by assessing students weekly on a rotational basis. For example, if 5–6 students are assessed every Friday, the teacher can collect data on each student at least once per month. Conclusion In this article, we described a study with 12 fourth-grade urban students who showed at-risk markers for reading failure. Three groups of students received a repeated reading intervention for 17, 12, and 6 weeks, respectively. Results indicate that all students improved their oral reading rates with the intervention over baseline levels; however, none of the students reached benchmark goals for fourth grade at the end of the study. Our results confirm that repeated readings is an effective fluency-building intervention for urban learners. This is especially true in inclusive settings in which students with and without disabilities can be engaged in meaningful reading activities at their instructional levels. If fluent readers read with more accuracy at a faster rate and hence comprehend text more than struggling readers,
Vol. 54, No. 1
then explicit and systematic fluency instruction should be present in urban classrooms. Such instruction complimented with additional instruction in vocabulary and comprehension allows students to integrate the various components of reading into a unified process, resulting in the ultimate goal of reading; namely, reading comprehension. NOTE 1. For further information, visit http://www.aimsweb.com. AUTHOR NOTES Shobana Musti-Rao is an assistant professor in early childhood and special needs education at the National Institute of Education in Singapore. Her research interests are in the area of using researchvalidated strategies in a response-to-intervention model in the general education setting. Renee O. Hawkins is an assistant professor in the Division of Human Services at the University of Cincinnati. Her interests include data-based decision making, response to intervention, and peer-mediated interventions. Elizabeth A. Barkley is a doctoral student in school psychology at the University of Cincinnati. Her research interests include literacy and early intervention. REFERENCES Cartledge, G., & Lo, Y. (2006). Teaching urban learners: Culturally responsive strategies for developing academic and behavioral competence. Champaign, IL: Research Press. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Cole, C. L., & Lionetti, T. M. (2004). A comparison of the effects of two rates of listening while reading on oral reading fluency and comprehension. Education and Treatment of Children, 27, 114–129. Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis (2nd ed.). Upper Saddle River, NJ: Pearson. Daane, M. C., Campbell, J. R., Grigg, W. S., Goodman, M. J., & Oranje, A. (2005). Fourth-grade students reading aloud: NAEP 2002 special study of oral reading. Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics. Retrieved September 20, 2007, from http:// nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2006469 Daly, E. J., & Martens, B. K. (1994). A comparison of three interventions for increasing oral reading performance: Application of the instructional hierarchy. Journal of Applied Behavior Analysis, 27, 459–469. Fisher, W. W., Kelley, M. E., & Lomas, J. E. (2003). Visual aids and structured criteria for improving visual inspection and interpretation of single-case designs. Journal of Applied Behavior Analysis, 36, 387–406. Foorman, B. R. (2007). Primary prevention in classroom reading instruction. Teaching Exceptional Children, 39(5), 24–30. Gilbert, L. M., Williams, R. L., & McLaughlin, T. F. (1996). Use of assisted reading to increase correct reading rates and decrease error rates of students with learning disabilities. Journal of Applied Behavior Analysis, 29, 255–257. Good, R. H., Gruba, J., & Kaminski, R. A. (2001). Best practices in using Dynamic Indicators of Basic Early Literacy Skills (DIBELS) in an outcome-driven model. In A. Thomas & J. Grimes (Eds.), Best practices in school psychology (Vol. 4, pp. 679–700). Washington, DC: National Association of School Psychologists.
Fall 2009 Good, R. H., & Kaminski, R. A. (Eds.) (2002). Dynamic Indicators of Basic Early Literacy Skills (6th ed.). Eugene, OR: Institute for the Development of Educational Achievement. Available from http://dibels.uoregon.edu Greenwood, C. R., & Maheady, L. (1997). Measurable change in student performance: Forgotten standard in teacher preparation? Teacher Education and Special Education, 20, 265–275. Gresham, F. M. (2005). Response to intervention: An alternative means of identifying students as emotionally disturbed. Education and Treatment of Children, 28, 328–344. Hitchcock, C. H., Prater, M. A., & Dowrick, P. W. (2004). Reading comprehension and fluency: Examining the effects of tutoring and video self-modeling on first-grade students with reading difficulties. Learning Disability Quarterly, 27, 89–103. Homan, S. P., Klesius, J. P., & Hite, C. (1993). Effects of repeated readings and nonrepetitive strategies on students’ fluency and comprehension. Journal of Educational Research, 87, 94–99. Hoskisson, K., & Krohm, B. (1974). Reading by immersion: Assisted reading. Elementary English, 51, 832–836. Kazdin, A. E. (1982). Single-case research designs: Methods for critical and applied settings. New York: Oxford University Press. Klingner, J., & Artiles, A. J. (2006). English language learners struggling to learn to read: Emergent scholarship on linguistic differences and learning disabilities. Journal of Learning Disabilities, 39, 386–389. Kuhn, M. R. (2005). A comparative study of small group fluency instruction. Reading Psychology, 26, 127–146. Kuhn, M. R., Schwanenflugel, P. J., Morris, R. D., Morriw, L. M., Woo, D. G., Meisinger, E. B., et al. (2006). Teaching children to become fluent and automatic readers. Journal of Literacy Research, 38, 357–387. Kuhn, M. R., & Stahl, S. (2003). Fluency: A review of developmental and remedial strategies. Journal of Educational Psychology, 95, 1–19. LaBerge, D., & Samuels, S. J. (1974). Toward a theory of automatic information processing in reading. Cognitive Psychology, 6, 293–323. Lane, K. L., & Beebe-Frankenberger, M. (2004). School-based interventions: The tools you need to succeed. Boston: Allyn & Bacon. Musti-Rao, S., & Cartledge, G. (2007). Early reading intervention for urban learners: Implications for practice. Multiple Voices, 10(1–2), 94–106.
Musti-Rao, Hawkins, & Barkley
23
National Reading Panel (NRP). (2000). Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction: Reports of the subgroups. Bethesda, MD: National Institute of Child Health and Human Development. No Child Left Behind Act of 2001. Pub. L. No. 107-110 (2001). O’Shea, L. J., Sindelar, P. T., & O’Shea, D. J. (1985). The effects of repeated readings and attentional cues on reading fluency and comprehension. Journal of Reading Behavior, 17(2), 129–142. O’Shea, L. J., Sindelar, P. T., & O’Shea, D. J. (1987). The effects of repeated readings and attentional cues on the reading fluency and comprehension of learning disabled readers. Learning Disabilities Research, 2(2), 103–109. Parker, R. I., Brossart, D. F., Vannest, K. J., Long, J. R., Garcia De-Alba, R., Roman, B., et al. (2005). Effect sizes in single case research: How large is large? School Psychology Review, 34, 116–132. Rasinski, T. V. (1990). Effects of repeated reading and listeningwhile-reading on reading fluency. The Journal of Educational Research, 83, 147–150. Rose, T. L. (1984). The effects of two pre-preparative procedures on oral reading. Journal of Learning Disabilities, 17, 544–548. Skinner, C. H., Cooper, L., & Cole, C. L. (1997). The effects of oral presentation previewing rates on reading performance. Journal of Applied Behavior Analysis, 30, 331–333. Stanovich, K. E. (1986). Matthew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21, 360–406. Strong, A. C., Wehby, J. H., Falk, K. B., & Lane, K. L. (2004). The impact of a structured reading curriculum and repeated reading on the performance of junior high students with emotional and behavioral disorders. School Psychology Review, 33, 561–581. Talbert-Johnson, C. (2004). Structural inequities and the achievement gap in urban schools. Education and Urban Society, 37, 22–36. White, E. B. (1952). Charlotte’s web. New York: Harper and Row. Yurick, A. L., Robinson, P. D., Cartledge, G., Lo, Y., & Evans, T. L. (2006). Using peer-mediated repeated readings as a fluencybuilding activity for urban learners. Educational and Treatment of Children, 29, 469–506.
