Fluency and Comprehension Gains as a Result of Repeated Reading: A Meta-Analysis

Abstract

Repeated reading is an evidenced-based strategy designed to increase reading fluency and comprehension. The author conducted a meta-analysis to ascertain essential instructional components of repeated reading and the effect of repeated reading on reading fluency and comprehension. This analysis indicates that repeated reading can be used effectively with nondisabled students and students with learning disabilities to increase reading fluency and comprehension on a particular passage and as an intervention to increase overall fluency and comprehension ability. Essential instructional components of repeated reading varied as a function of the type of repeated reading (i.e., whether effectiveness was evaluated reading the same passage or different passages). Implications for future research are also presented.

Full text

http://rse.sagepub.com/
Remedial and Special Education
http://rse.sagepub.com/content/25/4/252
The online version of this article can be found at:
DOI: 10.1177/07419325040250040801
2004 25: 252Remedial and Special Education
William J. Therrien
Fluency and Comprehension Gains as a Result of Repeated Reading: A Meta-Analysis
Published by:
Hammill Institute on Disabilities
and
http://www.sagepublications.com
can be found at:Remedial and Special EducationAdditional services and information for
http://rse.sagepub.com/cgi/alertsEmail Alerts:
http://rse.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://rse.sagepub.com/content/25/4/252.refs.htmlCitations:
What is This?
- Aug 1, 2004Version of Record >>
at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

252 REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004, Pages 252–261
Fluency and Comprehension Gains as a
Result of Repeated Reading
A Meta-Analysis
WILLIAM J. THERRIEN
ABSTRACT
Repeated reading is an evidenced-based strategy
designed to increase reading fluency and comprehension.
The author conducted a meta-analysis to ascertain essential
instructional components of repeated reading and the effect of
repeated reading on reading fluency and comprehension. This
analysis indicates that repeated reading can be used effectively
with nondisabled students and students with learning disabilities
to increase reading fluency and comprehension on a particular
passage and as an intervention to increase overall fluency and
comprehension ability. Essential instructional components of
repeated reading varied as a function of the type of repeated
reading (i.e., whether effectiveness was evaluated reading the
same passage or different passages). Implications for future
research are also presented.
ALTHOUGH TEACHING STUDENTS TO READ RE-
mains a major goal of education, many students have extreme
difficulty learning even basic reading skills. At least one in
five students has significant difficulties with reading acquisi-
tion (Lyon & Moats, 1997). In addition, approximately 37%
of fourth-grade students did not achieve at the most basic
reading level on a recent national test (U.S. Department of
Education, 2000). Reading difficulties are even more pro-
nounced for students with special needs, who often struggle
with reading throughout their school careers and into their
adult lives (Lyon & Moats, 1997).
The continuing difficulties students have with reading
has caused the educational community to reevaluate how to
teach basic and higher order reading skills. In 2000, a report
from the National Reading Panel (National Institute of Child
Health and Human Development, 2000) delineated five im-
portant reading skill areas: phonemic awareness, phonics,
vocabulary instruction, text comprehension strategies, and
reading fluency.
In this article, I examine procedures used to increase
reading fluency, which is the ability to read with speed and
accuracy (Samuels, 1979). The importance of reading fluency
began to emerge as early as 1969 (Clay, 1969; Clay & Imlach,
1971). Historically, two theoretical constructs for explaining
the importance of reading fluency and the origins of fluency
deficits have been cited in the literature. LaBerge and
Samuels (1974) theorized that reading fluency problems stem
from readers’ poor decoding skills. When decoding is too
slow, a “bottleneck” is created that impedes the flow of
thought and hampers comprehension. Poor readers often
spend a great deal of their cognitive resources on decoding
and have little left for comprehension. Fluent readers, on the
other hand, decode words quickly and accurately, thus retain-
ing many resources they can use for comprehension.
Conversely, Schrieber (1980) theorized that reading flu-
ency difficulties stem from the absence of prosodic cues in
written language. Schreiber contended that some readers
have difficulty transferring from oral language, where
prosodic markings are explicit, to written language, where
at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

253
REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004
among studies. Second, although the National Reading Panel
(National Institute, 2000) conducted a meta-analysis, it did
not separate out findings for repeated reading from the find-
ings for other fluency strategies. Third, neither review took
into account inconsistencies in conceptualization and imple-
mentation in repeated reading studies, making it difficult to
identify important instructional components within repeated
reading interventions.
For this study, I conducted a meta-analysis of repeated
reading studies to address the following questions:
1. Is repeated reading effective in increasing
reading fluency and comprehension?
2. What components within a repeated reading
intervention are critical to the success of the
program?
3. Do students with cognitive disabilities benefit
from repeated reading?
METHOD
To answer the three questions, I followed a six-step process.
First, I formed eligibility requirements for the studies that
would be considered for the review. Articles must have
(a) been published after Dahl’s (1977) chapter on repeated
reading and before June 2001, (b) been experimental and
quantitative, and (c) used school-age participants (i.e., ages
5–18 years).
Second, I located studies in two ways. I searched the
Educational Resources Information Center (ERIC) and Psy-
chological Information (PsycInfo) databases, using the fol-
lowing words: repeated reading, reading fluency, reading
automaticity, reading speed, reading accuracy, and reading
rate. Second, I conducted an ancestral search using the refer-
ence lists of the electronically identified articles and the ref-
erence list in the fluency chapter of The National Institute of
Child Health and Human Development’s (2000) report. This
search produced a total of 33 articles that met the criteria.
Third, I reviewed articles to determine if they were
amenable to meta-analysis methodology. Nine articles were
excluded because they lacked sufficient quantitative data to
calculate effect sizes (i.e., Daly, Martens, Hamler, Dool, &
Eckert, 1999; Freeland, Skinner, Jackson, McDaniel, & Smith,
2000; Gilbert, Williams, & McLaughlin, 1996; Kamps, Bar-
betta, Leonard, & Delquadri, 1994; Layton & Koenig, 1998;
Neill, 1980; Samuels, 1979; Tingstrom, Edwards, & Olmi,
1995; Weinstein & Cooke, 1992).
Fourth, I reviewed the remaining articles to determine
which effect size calculation would allow as many of the
studies to be analyzed as possible. Eighteen articles provided
information needed to calculate standard mean gain effect
sizes (Becker, 1988); therefore, I chose this calculation as the
metric for the meta-analysis. Six articles were excluded be-
prosodic markings need to be inferred. Readers who fail to
generate appropriate prosodic markings do not divide sen-
tences into meaningful phrases and therefore have difficulty
comprehending written text, regardless of their ability to
decode individual words.
Fluency difficulties may in fact stem from problems in
decoding or dividing sentences into meaningful phrases.
Logan (1997) contended that reading from text is complex
and requires integration across all levels of processing—from
decoding individual words to acquiring meaning from sen-
tences, paragraphs, and the text as a whole. Failure at any one
of these levels may result in reading fluency difficulties.
Fluency strategies have been designed and empirically
tested. One of the first empirically evaluated strategies to
focus on fluency was the neurological impress method
(Hollingsworth, 1970, 1978; Langford, Slade, & Barnett,
1974; Lorenz & Vockell, 1979). The goal of this strategy is to
increase fluency by having students and teachers read aloud
simultaneously. Although preliminary findings for the
method were encouraging, subsequent studies did not pro-
duce significant results (Langford et al., 1974; Lorenz &
Vo ckell, 1979).
Three fluency strategies have evolved out of the neuro-
logical impress model; they are assisted reading, reading
while listening, and paired reading. The first two are similar
to the neurological impress method, except that students read
along with an audiotape instead of with a speaking person. In
paired reading, which was designed as a home-based inter-
vention, students read along with a model, such as a teacher,
until they feel comfortable enough to read alone. Reports
regarding the effectiveness of these strategies are mixed (Law
& Kratochwill, 1993; Mefferd & Pettegrew, 1997; Miller,
Robson, & Bushell, 1986; Reitsma, 1988; Shany & Biemiller,
1995; van Bon, Boksebeld, Font Freide, & van den Hurk,
1991; VanWagenen, Williams, & McLaughlin, 1994; Winter,
1988).
Multicomponent interventions that include a fluency
component have also been developed. RAV-O (Wolf & Bow-
ers, 1999) is an intervention that includes instruction in word
retrieval, vocabulary, orthography, and fluency. Shared book
experience (Eldredge, Reutzel, & Hollingsworth, 1996) in-
cludes instruction in prosodic reading, conventions of print,
comprehension strategies, and fluency. These strategies
appear promising; however, they have limited empirical sup-
port.
One fluency strategy that has an extensive research base
is repeated reading, “a supplemental reading program that
consists of re-reading a short and meaningful passage until
a satisfactory level of fluency is reached” (Samuels, 1979,
p. 404). Two recent literature reviews concluded that repeated
reading has the potential to improve students’ reading fluency
(Meyer & Felton, 1999; National Institute, 2000).
Drawing definitive conclusions from these reviews is
difficult, however, for three reasons. First, Meyer and Felton
(1999) did not take into account sample size differences
at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

cause they did not provide the pertinent information needed
to compute mean gain effect sizes (i.e., Carver & Hoffman,
1981; Dahl, 1977; Faulkner & Levy, 1994; Levy, Nicholls, &
Kohen, 1993; Rashotte & Torgesen, 1985; Taylor, Wade, &
Yekovich, 1985). (Articles included in meta-analysis are
noted in the Reference section.)
Fifth, fluency and comprehension effect sizes were cal-
culated. Fluency measures were operationalized as number of
correct words per minute, words per minute, or reading
speed. Comprehension measures were operationalized as
either story retell measures or comprehension questions. Ef-
fect sizes were calculated using the formulae that follow:
ES =M2
−M1
Sp
SE = 2(1 −r) + ES 2
n2n
w=1
SE2
“M1is the mean at time one, M2is the mean at time two, Sp
is the average standard deviation of M1and M2,nis the com-
mon sample size at Time 1 and Time 2 and ris a correlation
estimate between time 1 and time 2 scores” (Lipsey & Wil-
son, 2001, p. 44). When studies did not include correlation
values, conservative estimates of .6 for fluency and .5 for
comprehension were used.
Sixth, I coded effect sizes to allow the studies to be ana-
lyzed. For each effect size, I coded the following variables:
(a) intervention length in sessions, (b) population (i.e., stu-
dents without disabilities, students with cognitive special
needs, or both students with and students without disabili-
ties), (c) dependent variable type (i.e., fluency or comprehen-
sion), and (d) repeated reading intervention components.
Cohen’s (1988) criteria for interpreting the strength of effect
sizes (small ES < .20, medium ES = .50, large ES > .80) were
used to gauge the magnitude of the findings in this analysis.
RESULTS
Dependent Variables
A preliminary inspection of effect sizes revealed a difference
in the nature of dependent measures. Effect sizes were either
nontransfer measures (i.e., measures of students’ ability to
fluently read or comprehend the same passage after reading it
multiple times) or transfer measures (i.e., measures of stu-
dents’ ability to fluently read or comprehend a new passage
after having read different passages multiple times). Due to
this difference, nontransfer and transfer effect sizes were ana-
lyzed separately.
Effectiveness of Repeated Reading: Nontransfer
A total of 28 nontransfer effect sizes were calculated (16 flu-
ency, 11 comprehension). Across all nontransfer measures,
the mean fluency ES increase was .83 (SE = .066) and mean
comprehension ES increase was .67 (SE = .080).
Component Analysis of Repeated
Reading: Nontransfer
Nontransfer studies varied in the instructional components
used within the interventions. Although all nontransfer
repeated reading interventions had students read passages
aloud to an adult, three components (cued reading, corrective
feedback, and performance criteria) often varied among stud-
ies and therefore were analyzed (see Note 1). See Table 1 for
the nontransfer component analysis.
Cued Reading. Prior to reading, students were cued to
focus on speed or comprehension. Four effect sizes did not
indicate the cue given and therefore were not included in the
analysis. Students cued to focus on speed obtained a mean
fluency ES of .72 (SE = .185) and a mean comprehension ES
of .66 (SE = .197). Students cued to focus on comprehension
obtained a mean fluency ES of .81 (SE = .096) and a mean
254 REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004
TABLE 1. Nontransfer Intervention Component Analysis
Cue type Corrective feedback Number of times passage read
Fluency Comp. Fluency & comp. Yes No 2 3 4
Fluency 0.72 0.81 0.94 0.68 0.88 0.57 0.85 0.95
n28 3 313 2103
Comprehension 0.66 0.75 0.67 — — — 0.66 0.71
n26 3 —— —10 2
Note. n indicates number of effect sizes. Dash indicates effect size not calculated or available.
Dependent
variable
at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

comprehension ES of .75 (SE = .127). Students cued to focus
on speed and comprehension obtained a mean fluency ES of
.94 (SE = .135) and a mean comprehension ES of .67 (SE =
.136).
Corrective Feedback. Corrective feedback was part of
the intervention represented by 3 of the 16 nontransfer flu-
ency effect sizes. Corrective feedback consisted of correcting
mispronunciations as they occurred or when students re-
quested assistance. Students who received corrective feedback
obtained a mean fluency ES of .68 (SE = .119), whereas stu-
dents who did not receive corrective feedback obtained a
mean fluency ES of .88 (SE = .075). None of the nontransfer
interventions looked at the effect of including corrective feed-
back on students’ comprehension ability.
Performance Criteria. A fixed number of readings
was the performance criterion used in 27 of 28 nontransfer
effect sizes (see Note 2). Collectively, these interventions
obtained a mean fluency ES of .81 (SE = .066) and a mean
comprehension ES of .66 (SE = .08). Interventions in which a
fixed number of readings were used had students read the
passage two, three, or four times (see Note 3). Mean fluency
effect sizes based on the number of readings were as follows:
two times, ES = .57 (SE = .141); three times, ES = .85 (SE =
.088); and four times, ES = .95 (SE = .145). For none of the
studies in which comprehension was measured had students
read the passage twice. Mean comprehension effect sizes
based on the number of readings were as follows: three times,
ES = .66 (SE = .089); and four times, ES = .71 (SE = .181).
Repeated Reading: Transfer
I calculated 27 transfer effect sizes (16 fluency, 11 compre-
hension). With the exception of Vaughn, Chard, Bryant, Cole-
man, and Kouzekanani (2000), whose interventions lasted 1
year to 3 years, mean intervention length was 36 sessions.
Across all transfer measures, mean fluency ES increase was
.50 (SE = .058) and mean comprehension ES increase was .25
(SE = .067).
Component Analysis of Repeated
Reading: Transfer
Instructional components within interventions that measured
transfer varied more than those in nontransfer studies. I ana-
lyzed six components: adult or peer instructor, modeling, cor-
rective feedback, performance criteria, comprehension, and
charting (see Note 4). See Table 2 for the transfer intervention
component analysis.
Adult or Peer. Interventions were conducted by adults
or peers. Students in interventions conducted by adults
obtained a mean fluency ES of 1.37 (SE = .177) and a mean
comprehension ES of .71 (SE = .265). Students in interven-
tions conducted by peers obtained a mean fluency ES of .36
(SE = .062) and a mean comprehension ES of .22 (SE = .070).
One peer-initiated intervention obtained a higher comprehen-
sion effect size than the average adult-run program. Simmons
and colleagues’ (1995) intervention used peers and obtained
a comprehension ES of .75 (SE = .377). Because there was a
discrepancy between adult- and peer-run programs, I sepa-
rated their effect sizes when examining the remaining re-
peated reading components.
Modeling. Eleven transfer effect sizes were based on
interventions that provided a model of fluent reading. Model-
ing consisted of the tutor reading the passage prior to having
the tutee read it. All of the interventions that contained mod-
eling were conducted by peers. Interventions that included
modeling obtained a mean fluency ES of .40 (SE = .077) and
a mean comprehension ES of .10 (SE = .104). Peer interven-
tions that did not include modeling obtained a mean fluency
ES of .30 (SE = .104) and a mean comprehension ES of .45
(SE = .119). One intervention that contained modeling
255
REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004
TABLE 2. Transfer Intervention Component Analysis
Corrective Number times
Tutor Modelingafeedback passage read Comp.aCharting
Adult Peer Yes No Yes No Fixed Perf. crit. Yes No Yes No
Fluency 1.37 0.36 0.40 0.30 0.51 0.46 0.38 1.70 0.39 0.33 0.57 0.40
n510 64 142 11 5 6 4 9 7
Comprehension 0.71 0.22 0.10 0.45 0.23 0.52 — — 0.28 0.14 0.11 0.44
n2954 92—— 6356
Note. Perf. crit. = performance criteria. nindicates number of effect sizes. Dash indicates effect size not calculated or available.
aModeling and comprehension mean effect sizes are based on studies conducted by peers only.
Dependent
variable
at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

obtained a fluency effect size higher than the average non-
model fluency effect size. Rasinski, Padak, Linek, and Sturte-
vant’s (1994) intervention contained modeling and obtained a
fluency ES of 1.0 (SE = .319). It should be noted that this
intervention lasted 120 days, compared to the average inter-
vention length of 32.4 days.
Corrective Feedback. Corrective feedback was part of
the intervention for all but four transfer effect sizes. Correc-
tive feedback consisted of correcting mispronunciations or
omissions while students were reading or after they had read.
Students were either provided with the correct pronunciation
or prompted to sound out or reread the word. Students who
received corrective feedback obtained a mean fluency ES of
.51 (SE = .06) and a mean comprehension ES of .23 (SE =
.07). Students who did not receive corrective feedback
obtained a mean fluency ES of .46 (SE = .227) and a mean
comprehension ES of .52 (SE = .234). When interventions
conducted by peers were excluded from the analysis, inter-
ventions that provided corrective feedback obtained a mean
fluency ES of 1.37 (SE = .177; see Note 5). The impact of
adult feedback on comprehension could not be determined
because only one adult-run intervention measured compre-
hension.
Performance Criteria. Transfer interventions used
either a fixed number of readings or a performance criterion
to determine when to move from one passage to the next. Per-
formance criteria consisted of either reading until a fixed
number of correct words per minute was reached or reading a
passage within a predetermined time period. Interventions
that used a performance criterion obtained a mean fluency ES
of 1.70 (SE = .188). Interventions that used a fixed number of
readings obtained a mean fluency ES of .38 (SE = .061). A
comparison between studies using the two types of criteria in
regards to gains in comprehension could not be made because
only one study that measured comprehension used a perfor-
mance criterion.
Transfer interventions that used a fixed number of read-
ings had students read passages two or three times (see Note
6). Mean fluency effect sizes for number of readings were as
follows: two readings, ES = .37 (SE = .087); three readings,
ES = .42 (SE = .091). Mean comprehension effect sizes for
number of readings were as follows: two readings, ES = .03
(SE = .093); three readings, ES = .49 (SE = .108).
Comprehension Component. Comprehension ques-
tions were asked or a paragraph summary was implemented
in peer-run transfer interventions, which yielded 12 effect
size calculations. Peer-run interventions that included a com-
prehension component obtained a mean fluency ES of .39
(SE = .084) and a mean comprehension ES of .28 (SE = .092).
Peer-run interventions that did not include a comprehension
component obtained a mean fluency ES = .33 (SE = .091) and
a mean comprehension ES of .14 (SE = .106).
Charting. Charting student progress was part of the
intervention for 14 effect sizes. Interventions that included
charting obtained a mean fluency ES of .57 (SE = .075) and a
mean comprehension ES of .11 (SE = .088). Interventions
that did not include charting obtained a mean fluency ES of
.40 (SE = .091) and a mean comprehension ES of .44 (SE =
.105). Adult-implemented interventions that charted student
progress obtained a mean fluency ES of 1.58 (SE = .208). No
adult-implemented intervention that charted student progress
measured comprehension.
Students with Disabilities
Students with learning disabilities (LD) were the only stu-
dents with disabilities who participated in the studies exam-
ined. Study reports indicated that students with LD were
identified based upon school, school district, or state guide-
lines. Four studies (Bryant et al., 2000; Mathes & Fuchs,
1993; Mercer, Campbell, Miller, Mercer, & Lane, 2000; Sin-
delar, Monda, & O’Shea, 1990) that reported the results for
students with LD separately explicitly indicated that a dis-
crepancy formula between achievement and IQ was used to
identify students with LD.
Students with LD and students without disabilities were
compared to ascertain if repeated reading increased fluency
and comprehension for both groups (see Note 7). For non-
transfer measures, the mean fluency ES for students without
disabilities was .85 (SE = .075), and the mean comprehension
ES was .64 (SE = .094). The mean fluency ES for students
with LD was .75 (SE = .161), and the mean comprehension
ES was .73 (SE = .152).
For transfer measures, the mean fluency ES for students
without disabilities was .59 (SE = .11), and the mean com-
prehension ES was .18 (SE = .126). The mean fluency ES for
256 REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004
TABLE 3. Comparison of Nondisabled
Students and Students with LD
Nondisabled Students with
Dependent variable students LD
Nontransfer
Fluency 0.85 0.75
n12 4
Comprehension 0.64 0.73
n84
Transfer
Fluency 0.59 0.79
n65
Comprehension 0.18 0.41
n42
Note. n indicates number of effect size. LD = learning disabilities.
at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

students with LD was .79 (SE = .124), and the mean compre-
hension ES was .41 (SE = .173). See Table 3 for comparisons
between nondisabled students and students with LD.
DISCUSSION
Conclusions
As was found in previous literature reviews (Meyer & Felton,
1999; National Institute, 2000), findings from this analysis
indicate that repeated reading improves the reading fluency
and comprehension of both nondisabled (ND) students and
students with LD. All students obtained a moderate mean
increase in fluency (ND students: ES = .76, SE = .06; students
with LD: ES = .77, SE = .09) and a somewhat smaller mean
increase in comprehension (ND students: ES = .48, SE = .07;
students with LD: ES = .59, SE =. 11; see Note 8).
Unlike previous literature reviews, this analysis sepa-
rated results into nontransfer measures and transfer measures.
Nontransfer results (i.e., measures of students’ ability to flu-
ently read or comprehend a passage after reading it multiple
times) from this analysis indicate that repeated reading is an
effective strategy for improving reading fluency and compre-
hension on a passage that is read repeatedly. Across all non-
transfer studies, the mean fluency increase was large (ES =
.83, SE = .066), and mean comprehension effect size was
moderate (ES = .67, SE = .080). When students reread a pas-
sage, they read it more fluently and comprehended it better.
Transfer results (i.e., measures of students’ ability to flu-
ently read or comprehend new passages after having pre-
viously reread other reading material) from this analysis
indicate that repeated reading may also improve students’
ability to fluently read and comprehend new passages.
Although most interventions lasted 45 sessions or less, stu-
dents across all transfer studies obtained a moderate mean
fluency effect size increase (ES = .50, SE = .058) and a
smaller, but still significant, mean comprehension effect size
increase (ES = .25, SE = .067). For transfer interventions con-
ducted by adults, the mean fluency effect size increase was
large (ES = 1.37, SE = .177), and the mean comprehension
effect size increase was moderate (ES = .71, SE = .265). Con-
sequently, it appears that repeated reading has the potential to
improve students’ overall reading fluency and comprehension
abilities in regards to new material.
This analysis also offered clarification as to essential
instructional components of repeated reading. Regardless of
purpose, all repeated reading interventions should ensure that
students read passages aloud to an adult. Adult implementa-
tion is recommended because the fluency and comprehension
effect sizes for students in transfer interventions conducted
by adults were more than three times larger (mean fluency
ES = 1.37, mean comprehension ES = .71) than those ob-
tained by students in interventions conducted by peers (mean
fluency ES = .36, mean comprehension ES = .22).
In addition, if the purpose of repeated reading is to
enable students to fluently read and comprehend a particular
passage (i.e., nontransfer), students should be provided with
a cue, and the passage should be repeated three to four times.
A definitive answer as to the type of cue to provide (i.e., flu-
ency, comprehension, or speed and comprehension) could not
be determined because differences in fluency and compre-
hension gains based on the type of cue received were negli-
gible. In two individual studies, O’Shea, Sindelar, and O’Shea
(1985, 1987) directly compared use of a fluency cue versus a
comprehension cue and found that although nondisabled stu-
dents adapted their reading style to the cue provided, students
with LD who were cued for comprehension remembered
more but read as fast as students with LD who were cued for
speed. Until additional research is conducted, the use of a
combined speed and comprehension cue is recommended.
The passage should be read three to four times because when
the passage was read three times (ES = .85) or four times
(ES = .95), mean fluency effect size increases were more than
30% larger than when the passage was read twice (ES = .57).
Reading the passage more than four times does not appear to
be necessary because the difference in comprehension gains
between reading the passage three times (ES = .66) and four
times (ES = .71) was minimal. In addition, four individual
studies (DiStefano, Noe, & Valencia, 1981; O’Shea et al.,
1985, 1987; Stoddard, Valcante, Sindelar, O’Shea, & Algoz-
zine, 1993) investigated the number of readings and found
that gains in comprehension ceased to be significant after the
third reading.
If the purpose of the intervention is to improve overall
reading fluency and comprehension, a corrective feedback
component should be added and passages should be read
until a performance criterion is reached. Corrective feedback
on word errors seems to be essential because all students in-
volved in adult-run interventions were given corrective feed-
back and obtained a large mean fluency effect size (1.37).
The use of a performance criterion (reading until a fixed
number of correct words per minute is reached or reading a
passage within a predetermined time period) is recommended
because interventions that used such a criterion obtained a
mean fluency effect size increase (1.70) that was more than
four times larger than that obtained by interventions that used
a fixed number of readings (.38).
The relative importance of the essential repeated reading
components can be independently validated by reviewing
research that examined the components outside of repeated
reading interventions. For example, DiStefano et al. (1981)
examined cue usage and found that when readers were pro-
vided with a cue, they were able to adapt their reading rate
to fit different purposes. Pany and McCoy (1988) examined
corrective feedback and found that it enhanced both word
recognition and comprehension, although the effectiveness of
feedback may depend on goals and outcome measures (Hoff-
man et al., 1984). The enhanced performance of interventions
that used a performance criterion can be explained by exam-
257
REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004
at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

ining research that compared the use of an equal opportunity
to respond (e.g., such as a fixed number of readings) to the
use of an improvement criterion (e.g., such as correct words
per minute). Underwood (1954) found that equal opportunity
to respond did not result in equal learning. Students who
learned faster received more benefit from each successful
practice trial than did slower learners.
Components deemed by this analysis to be nonessential
were not harmful; their effects simply were not pronounced.
The impact of these components may have been overshad-
owed by more potent components (e.g., peer-run programs
were effective but not as effective as adult-run programs).
The components’ impact may not have been dramatic (e.g.,
peer-run interventions that included modeling did not differ
significantly from peer-run interventions that did not include
modeling), or their importance may remain unknown because
they were implemented in a limited number of studies (e.g.,
only two nontransfer studies used corrective feedback, and
neither of them measured comprehension; none of the inter-
ventions conducted by an adult included a comprehension
component, and only one charted student progress). Regard-
less of the reason, the inclusion of components other than
those deemed essential through this analysis does not appear
necessary to the success of a repeated reading intervention.
Limitations
There are four limitations to the conclusions for this analysis.
First, characteristics of students who may benefit from
repeated reading needs further clarification. Although the
results indicate that repeated reading is effective for nondis-
abled students and students with LD, most studies did not
provide students’ reading levels. As a result, the effectiveness
of repeated reading for students at different reading levels
could not be determined.
Second, most studies did not provide information on the
reading material used in the intervention. Individual repeated
reading studies have investigated passage characteristics,
such as content and word overlap (Faulkner & Levy, 1994;
Rashotte & Torgesen, 1985), and found that using a series of
passages with a high degree of word overlap may play a crit-
ical role in the effectiveness of repeated reading. Other stud-
ies that have not dealt specifically with repeated reading have
shown that student characteristics as they pertain to fluency
may also be important. For example, O’Conner et al. (2002)
found that students made stronger fluency gains when the
material used was at their instructional level rather than
their grade level. Additional research is needed to determine
(a) reading material to use within a repeated reading inter-
vention and (b) the relative importance of the selected read-
ing material on the effectiveness of repeated reading.
Third, this analysis was unable to determine the impor-
tance of including a charting or comprehension component in
a transfer repeated reading intervention. Charting the prog-
ress of students and asking them to retell the story or answer
comprehension questions could be important components;
however, not enough data were available to conduct this type
of evaluation. Preliminary results indicate that including a
charting component may enhance students’ fluency ability,
whereas including a comprehension component may enhance
students’ comprehension ability.
Fourth, effect sizes in this analysis were based on differ-
ences between pretest and posttest scores. Without a compar-
ison control group, the relationships between repeated
reading and gains in fluency and comprehension are open to
other hypotheses. Caution should be used when evaluating
the findings of this analysis because unknown independent
variable(s) may have affected students’ reading achievement.
Additional repeated reading studies that include a control
group should be conducted to clarify the findings of this
analysis.
Implications for Future Research
Although results from this analysis (a) confirmed previous
findings that repeated reading improves students’ reading
fluency and comprehension and (b) delineated essential
repeated reading instructional components, many critical
questions remain unanswered. Most pressing are questions
related to adding instructional components, using peers to
conduct repeated reading interventions, including a modeling
component, and measuring repeated reading’s impact on over-
all reading achievement measures.
Essential repeated reading instructional components
have been defined; however, the addition of components not
investigated in this analysis might significantly improve the
effectiveness of repeated reading. Preliminary results from an
intervention that included a phonics and sight word compo-
nent indicated that these components may prove to be helpful
(Vaughn et al., 2000). The authors obtained a significant
increase in students’ reading fluency.
Caution is advised if peer tutors are used to conduct
repeated reading interventions because the qualities, charac-
teristics, and training they need to be competent tutors have
yet to be determined. One study (Simmons, Fuchs, Fuchs,
Mathes, & Hodge, 1995) in this analysis did use peers and
obtained increases in student achievement commensurate
with that for students in adult-run programs. Research into
the characteristics of an effective tutor or the training a peer
required to implement repeated reading effectively is war-
ranted.
Although preliminary results indicate that modeling is
not an essential repeated reading component, additional
research on including a modeling component is warranted
because no adult-run interventions included this component,
and one peer-run intervention (Rasinski et al., 1994) that
included modeling obtained a significant increase in stu-
dents’ reading fluency.
258 REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004 at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

Last, to evaluate repeated reading’s impact on all aspects
of reading ability, long-term studies need to be conducted.
With the exception of Vaughn and colleagues (2000), whose
interventions lasted 1 year to 3 years, no study in this review
had a duration of more than 6 months. To determine the
impact of repeated reading on students’ reading achievement
in general requires conducting studies of longer duration.
Practical Implications
This review provides two findings that inform practice. First,
it indicates that repeated reading can be used effectively to
improve students’ ability to fluently read and understand a
particular passage and as an intervention to improve students’
overall reading fluency and comprehension ability. Second,
this review delineates essential instructional components to
include within a repeated reading program. Such components
depend on the goal of the intervention. If repeated reading is
intended to improve students’ ability to read and comprehend
a particular passage (i.e., nontransfer), students should be
cued to focus on speed and comprehension and the passage
should be read aloud three to four times. If repeated reading
is intended as an intervention to improve students’ overall
reading fluency and comprehension (i.e., transfer), there are
three essential components: Passages should be read aloud to
an adult, corrective feedback on word errors should be given,
and passages should be read until a performance criterion is
reached. ■
WILLIAM J. THERRIEN, MEd, is currently a special education doctoral
student at The Pennsylvania State University. His research and professional
interests include learning disabilities, mental retardation, reading instruction,
and classroom management. Address: William J. Therrien, The Pennsylvania
State University, 228 Cedar, University Park, PA 16802; e-mail: wjt105@
psu.edu
AUTHOR’S NOTES
1. This article is based on a doctoral investigation carried out under the
supervision of Dr. John Salvia, The Pennsylvania State University, whose
guidance and encouragement are gratefully acknowledged.
2. The author appreciates the feedback provided on an earlier draft of this
article by Dr. Charles Hughes and Dr. Richard Kubina.
NOTES
1. Other components used, but not regularly, were sentence segmentation,
intonation training, and audiotaping of students’ reading. Due to their in-
frequent use, these components were not analyzed.
2. One nontransfer fluency effect size was based on an intervention that
used correct words per minute as the performance criterion (Herman,
1985).
3. In four studies (DiStefano et al., 1981; O’Shea et al., 1985, 1987; Stod-
dard et al., 1993), students read passages seven times. Results indicated
that gains in comprehension ceased to be significant after the third read-
ing.
4. Other components used, but not regularly, were explicit instruction, word
identification skills, phonics instruction, sight word instruction, audio-
tapes of students’ reading, and performance of reading in front of class.
Due to the infrequent use of these components, they were not analyzed.
5. All adult-run interventions that measured fluency provided corrective
feedback.
6. In one study (Homan, Klesius, & Hite, 1993), students read passages four
times.
7. Mean effect sizes are reported only for studies that provided separate data
for nondisabled students and students with LD.
8. Reported mean effect sizes are combined nontransfer and transfer results
and include only studies that provided separate data for nondisabled stu-
dents and students with LD. The combined nontransfer and transfer mean
effect size for all of the evaluated studies was .64 (SE = .04) for fluency
and .42 (SE = .05) for comprehension.
9. Only the experimental condition that had students reread passages with-
out the assistance of an audiotape was included in the meta-analysis.
REFERENCES
*Indicates studies included in the analysis.
Becker, B. J. (1988). Synthesizing standardized mean-change measures.
British Journal of Mathmatical and Statistical Psychology, 41, 257–278.
*Bryant, D. P., Vaughn, S., Linan-Thompson, S., Ugel, N., Hamff, A., &
Hougen, M. (2000). Reading outcomes for students with and without
reading disabilities in general education middle-school content area
classes. Learning Disability Quarterly, 23, 238–252.
Carver, R. P., & Hoffman, J. V. (1981). The effect of practice through
repeated reading on gain in reading ability using a computer-based
instructional system. Reading Research Quarterly, 16, 374–390.
Clay, M. M. (1969). Reading errors and self-correction behaviour. British
Journal of Educational Psychology, 39, 47–56.
Clay, M. M., & Imlach, R. H. (1971). Juncture, pitch, and stress as reading
behavior variables. Journal of Verbal Learning and Verbal Behavior, 10,
133–139.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd
ed.). Hillsdale, NJ: Erlbaum.
Dahl, P. R. (1977). An experimental program for teaching high-speed word
recognition and comprehension skills. In J. Burton, T. Lovitt, & T. Row-
lands (Eds.), Communication research in learning disabilities and men-
tal retardation (pp. 33–65). Baltimore: University Park Press.
Daly, E. J., Martens, B. K., Hamler, K. R., Dool, E. J., & Eckert, T. L. (1999).
A brief experimental analysis for identifying instructional components
needed to improve oral reading fluency. Journal of Applied Behavior
Anaylisis, 32, 83–94.
DiStefano, P., Noe, M., & Valencia, S. (1981). Measurement of the effects of
purpose and passage difficulty on reading flexibility. Journal of Educa-
tional Psychology, 73, 602–606.
*Downhower, S. L. (1987). Effects of repeated reading on second-grade tran-
sitional readers’ fluency and comprehension. Reading Research Quar-
terly, 22, 389–406 (see Note 9).
Eldredge, J. L., Reutzel, D. R., & Hollingsworth, P. M. (1996). Comparing
the effectiveness of two oral reading practices: Round-robin and the
shared book experience. Journal of Literacy Research, 28, 201–225.
Faulkner, H. J., & Levy, B. A. (1994). How text difficulty and reader skill
interact to produce differential reliance on word content overlap in read-
ing transfer. Journal of Experimental Child Psychology, 58, 1–24.
*Faulkner, H. J., & Levy, B. A. (1999). Fluent and nonfluent forms of trans-
fer in reading: Words and their message. Psychonomic Bulletin and
Review, 6, 111–116.
Freeland, J. T., Skinner, C. H., Jackson, B., McDaniel, C. E., & Smith, S.
(2000). Measuring and increasing silent reading comprehension rates:
Empirically validating a repeated reading intervention. Psychology in
the Schools, 37, 415–429.
Gilbert, L. M., Williams, R. L., & McLaughlin, T. F. (1996). Use of assisted
reading to increase correct reading rates and decrease error rates of stu-
259
REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004
at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

dents with learning disabilities. Journal of Applied Behavior Analysis,
29, 255–257.
*Herman, P. A. (1985). The effects of repeated readings on reading rate,
speech pauses, and word recognition accuracy. Reading Research Quar-
terly, 20, 553–565.
Hoffman, J. V., O’Neal, S. F., Kastler, L., Clements, R., Segel, K., & Nash,
M. (1984). Guided oral reading and miscue focused verbal feedback in
second grade classrooms. Reading Research Quarterly, 19, 367–384.
Hollingsworth, P. M. (1970). An experiment with the impress method of
teaching reading. The Reading Teacher, 24, 112–114,187.
Hollingsworth, P. M. (1978). An experimental approach to the impress
method of teaching reading. The Reading Teacher, 31, 624 – 626.
*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.
Kamps, D. M., Barbetta, P. M., Leonard, B. R., & Delquadri, J. (1994).
Classwide peer tutoring: An integration strategy to improve reading
skills and promote peer interactions among students with autism and
general education peers. Journal of Applied Behavior Analysis, 27,
49–61.
LaBerge, D., & Samuels, S. J. (1974). Toward a theory of automatic infor-
mation processing in reading. Cognitive Psychology, 6, 293–323.
Langford, K., Slade, K., & Barnett, A. (1974). An examination of impress
techniques in remedial reading. Academic Therapy, 9, 309–319.
Law, M., & Kratochwill, T. R. (1993). Paired reading:An evaluation of a par-
ent tutorial program. School Psychology International, 14, 119–147.
Layton, C. A., & Koenig, A. J. (1998, May). Increasing reading fluency in
elementary students with low vision through repeated reading. Journal
of Visual Impairment and Blindness.
*Levy, B. A., Abello, B., & Lysynchuk, L. (1997). Transfer from word train-
ing to reading in context: Gains in reading fluency and comprehension.
Learning Disability Quarterly, 20, 173–188.
Levy, B. A., Nicholls, A., & Kohen, D. (1993). Repeated readings: Process
benefits for good and poor readers. Journal of Experimental Child Psy-
chology, 56, 303–327.
Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis (Vol. 49).
Thousand Oaks, CA: Sage.
Logan, G. D. (1997). Automaticity and reading: Perspective from the
instance theory of automatization. Reading and Writing Quarterly, 13,
123–146.
Lorenz, L., & Vockell, E. (1979). Using the neurological impress method
with learning disabled readers. Journal of Learning Disabilities, 12,
420–422.
Lyon, G. R., & Moats, L. C. (1997). Critical conceptual and methodological
considerations in reading intervention research. Journal of Learning
Disabilities, 30, 578–588.
*Mathes, P. G., & Fuchs, L. S. (1993). Peer-mediated reading instruction in
special education resource rooms. Learning Disabilities Research and
Practice, 8, 233–243.
Mefferd, P. E., & Pettegrew, B. S. (1997). Fostering literacy acquisition of
students with developmental disabilities: Assisted reading with pre-
dictable trade books. Reading Research and Instruction, 36, 177–190.
*Mercer, C. D., Campbell, K. U., Miller, M. D., Mercer, K. D., & Lane, H.
B. (2000). Effects of a reading fluency intervention for middle schoolers
with specific learning disabilities. Learning Disabilities Research and
Practice, 15, 179–189.
Meyer, M. S., & Felton, R. H. (1999). Repeated reading to enhance fluency:
Old approaches and new directions. Annals of Dyslexia, 49, 283–306.
Miller, A., Robson, D., & Bushell, R. (1986). Parental participation in paired
reading: A controlled study. Educational Psychology, 6, 277–284.
National Institute of Child Health and Human Development. (2000). Teach-
ing children to read: An evidence-based assessment of the scientific
literature on reading and its implications for reading instruction. Wash-
ington, DC: U.S. Government Printing Office.
Neill, K. (1980, Winter). Turn kids on with repeated reading. Teaching
Exceptional Children, pp. 63–65.
O’Conner, R. E., Bell, K. M., Harty, K. R., Larkin, L. K., Sackor, S. M., &
Zigmond, N. (2002). Teaching reading to poor readers in the intermedi-
ate grades: A comparison of text difficulty. Journal of Educational Psy-
chology, 94, 474 – 485.
*O’Shea, L. J., Sindelar, P. T., & O’Shea, D. J. (1985). The effects of
repeated readings and attentional cues on reading fluency and compre-
hension. Journal of Reading Behavior, 17, 129–141.
*O’Shea, L. J., Sindelar, P. T., & O’Shea, D. J. (1987). The effects of
repeated reading and attentional cues on the reading fluency and com-
prehension of learning disabled readers. Learning Disabilities Research,
2, 103–109.
Pany,D., & McCoy, K. M. (1988). Effects of corrective feedback on word
accuracy and reading comprehension of readers with learning disabili-
ties. Journal of Learning Disabilities, 21, 546–550.
Rashotte, C. A., & Torgesen, J. K. (1985). Repeated reading and reading
fluency in learning disabled children. Reading Research Quarterly, 20,
180–188.
*Rasinski, T. (1990). Effects of repeated reading and listening-while-reading
on reading fluency. Journal of Educational Research, 83, 147–150.
*Rasinski, T., Padak, N., Linek, W., & Sturtevant, E. (1994). Effects of flu-
ency development on urban second-grade readers. Journal of Educa-
tional Research, 87, 158–165.
Reitsma, P. (1988). Reading practice for beginners: Effects of guided read-
ing, reading-while-listening and independent reading with computer-
based speech feedback. Reading Research Quarterly, 23, 219–235.
Samuels, S. J. (1979). The method of repeated reading. The Reading Teacher,
32, 403–408.
Schrieber, P. A. (1980). On the acquisition of reading fluency. Journal of
Reading Behavior, 12, 177–186.
Shany, M. T., & Biemiller, A. (1995). Assisted reading practice: Effects on
performance for poor readers in grades 3 and 4. Reading Research Quar-
terly, 30, 382–395.
*Simmons, D. C., Fuchs, D., Fuchs, L. S., Hodge, J. P., & Mathes, P. G.
(1994). Importance of instructional complexity and role reciprocity to
classwide peer tutoring. Learning Disabilities Research and Practice, 9,
203–212.
*Simmons, D. C., Fuchs, L. S., Fuchs, D., Mathes, P., & Hodge, J. P. (1995).
Effects of explicit teaching and peer tutoring on the reading achievement
of learning-disabled and low-performing students in regular classrooms.
The Elementary School Journal, 95, 387–408.
*Sindelar, P. T., Monda, L. E., & O’Shea, L. J. (1990). Effects of repeated
readings on instructional- and mastery-level readers. Journal of Educa-
tional Research, 83, 220–226.
*Stoddard, K., Valcante, G., Sindelar, P. T., O’Shea, L., & Algozzine, B.
(1993). Increasing reading rate and comprehension: The effects of
repeated readings, sentence segmentation, and intonation training. Read-
ing Research and Instruction, 32, 53–65.
Taylor, N. E., Wade, M. R., & Yekovich, F. R. (1985). The effects of text
manipulation and multiple reading strategies on the reading performance
of good and poor readers. Reading Research Quarterly, 20, 566–574.
Tingstrom, D. H., Edwards, R. P., & Olmi, D. J. (1995). Listening preview-
ing in reading to read: Relative effects on oral reading fluency. Psychol-
ogy in the Schools, 32, 318–327.
Underwood, B. J. (1954). Speed of learning and amount retained: A consid-
eration of methodolgy. Psychological Bulletin, 51, 276–282.
U.S. Department of Education. (2000). Education statistics quarterly: The
national report card: Fourth grade reading 2000. Washington, DC: U.S.
Government Printing Office.
van Bon, W. H. J., Boksebeld, L. M., Font Freide, T. A. M., & van den Hurk,
A. J. M. (1991). A comparison of three methods of reading-while-
listening. Journal of Learning Disabilities, 24, 471–476.
VanWagenen, M. A., Williams, R. L., & McLaughlin, T. F. (1994). Use of
assisted reading to improve reading rate, word accuracy and compre-
hension with ESL Spanish-speaking students. Perc eptual and Motor
Skills, 79, 227–230.
260 REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004 at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from

*Vaughn, S., Chard, D. J., Bryant, D. P., Coleman, M., & Kouzekanani, K.
(2000). Fluency and comprehension interventions for third-grade stu-
dents. Remedial and Special Education, 21, 325–335.
Weinstein, G., & Cooke, N. L. (1992). The effects of two repeated reading
interventions on generalization of fluency. Learning Disability Quar-
terly, 15, 21–28.
Winter, S. (1988). Paired reading: A study of process and outcome. Educa-
tional Psychology, 8, 135–151.
Wolf, M., & Bowers, P. G. (1999). The double-deficit hypothesis for the
developmental dyslexias. Journal of Educational Psychology, 91, 415–
438.
*Young, A., Bowers, P. G., & MacKinnon, G. E. (1996). Effects of prosodic
modeling and repeated reading on poor readers’ fluency and compre-
hension. Applied Psycholinguistics, 17, 59–84.
261
REMEDIAL AND SPECIAL EDUCATION
Volume 25, Number 4, July/August 2004
at UNIV OF VIRGINIA on October 1, 2014rse.sagepub.comDownloaded from