ArticlePDF Available

Assessing an Instructional Level During Reading Fluency Interventions: A Meta-Analysis of the Effects on Reading

Authors:

Abstract and Figures

The current study meta-analyzed 27 effects from 21 studies to determine the effect assessment of text difficulty had on reading fluency interventions, which resulted in an overall weighted effect size ( ES) = 0.43 (95% CI = [0.25, 0.62], p < .001). Using reading passages that represented an instructional level based on accuracy criteria led to a large weighted effect of ES = 1.03, 95% CI = [0.65, 1.40], p < .01), which was reliably larger ( p < .05) than that for reading fluency interventions that used reading passages with an instructional level based on rate criteria (weighted ES = 0.29, 95% CI = [0.07, 0.50], p < .01). Using reading passages based on leveling systems or those written at the students’ current grade level resulted in small weighted effects. The approach to determining difficulty for reading passages used in reading fluency interventions accounted for 11% of the variance in the effect ( p < .05) beyond student group (no risk, at-risk, disability) and type of fluency intervention. The largest weighted effect was found for students with reading disabilities ( ES = 1.14, 95% CI = [0.64, 1.65], p < .01).
Content may be subject to copyright.
https://doi.org/10.1177/15345084241247064
Assessment for Effective Intervention
1 –11
© Hammill Institute on Disabilities 2024
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/15345084241247064
aei.sagepub.com
Literature Review
Intervention researchers frequently focus on reading fluency
because it is a core domain in reading that is closely linked to
reading comprehension (Basaran, 2013) and general reading
skills (Kim & Wagner, 2015). Reading fluency is the combi-
nation of reading speed, accuracy, and prosody (National
Reading Panel, 2000), and reading interventions that are
designed to increase the speed with which students read also
lead to better reading comprehension and prosody (Keyes
et al., 2016; Lee & Yoon, 2017; Young et al., 2015). Moreover,
assessments of reading fluency are strong indicators of over-
all reading proficiency (Fuchs et al., 2001).
Meta-analytic research has consistently found that
repeated reading of the same text (Samuels, 1979) had mod-
erate effects on reading fluency (d = 0.83) and comprehen-
sion (d = 0.67; Therrien, 2004), even among students with
disabilities (Lee & Yoon, 2017). Maki and Hammerschmidt-
Snidarich (2022) meta-analyzed 33 studies that examined
different reading fluency interventions (i.e., repeated read-
ing, continuous reading of different texts, listening while
reading, or group fluency interventions) and found a small
to moderate effect (g = 0.46), but there was considerable
variability in the effectiveness of the fluency interventions,
which was also found in previous meta-analytic research
(Hudson et al., 2020).
There are several hypotheses as to why there was variabil-
ity in the effects of fluency interventions. First, characteristics
of the students moderated the results (Lee & Yoon, 2017). For
example, reading fluency interventions for students who were
at risk for reading problems or diagnosed with a reading dis-
ability led to larger effects (0.63 for each group) than did
research with students who were not at risk (0.32; Maki &
Hammerschmidt-Snidarich, 2022). Second, characteristics of
the intervention such as number of repetitions (Lee & Yoon,
2017; Therrien, 2004), type of intervention (e.g., repeated or
not; Maki & Hammerschmidt-Snidarich, 2022), and size of
the group receiving the intervention (e.g., individually, dyad,
or small group; Hudson et al., 2020) all moderated the results.
A third potential reason why researchers see variability in
the effects of repeated reading and other fluency interven-
tions could be the relationship between student skill and pas-
sage difficulty. Most reading fluency interventions were
conducted with reading passages rather than other materials
such as books (Maki & Hammerschmidt-Snidarich, 2022),
but little attention has been given to how to best select
1247064AEIXXX10.1177/15345084241247064Assessment for Effective InterventionBurns
review-article2024
1University of Florida, Gainesville, USA
Corresponding Author:
Matthew K. Burns, UF College of Education, University of Florida, 1221
SW 5th Avenue, 1007 Norman Hall, Gainesville, FL 32601, USA.
Email: burnsm1@ufl.edu
Associate Editor: Abigail Allens
Assessing an Instructional Level
During Reading Fluency Interventions:
A Meta-Analysis of the Effects on Reading
Matthew K. Burns, PhD1
Abstract
The current study meta-analyzed 27 effects from 21 studies to determine the effect assessment of text difficulty had on
reading fluency interventions, which resulted in an overall weighted effect size (ES) = 0.43 (95% CI = [0.25, 0.62], p <
.001). Using reading passages that represented an instructional level based on accuracy criteria led to a large weighted
effect of ES = 1.03, 95% CI = [0.65, 1.40], p < .01), which was reliably larger (p < .05) than that for reading fluency
interventions that used reading passages with an instructional level based on rate criteria (weighted ES = 0.29, 95% CI =
[0.07, 0.50], p < .01). Using reading passages based on leveling systems or those written at the students’ current grade
level resulted in small weighted effects. The approach to determining difficulty for reading passages used in reading fluency
interventions accounted for 11% of the variance in the effect (p < .05) beyond student group (no risk, at-risk, disability)
and type of fluency intervention. The largest weighted effect was found for students with reading disabilities (ES = 1.14,
95% CI = [0.64, 1.65], p < .01).
Keywords
curriculum-based assessment, reading/literacy
2 Assessment for Effective Intervention 00(0)
passages for the interventions. In fact, Lee and Yoon (2017)
reported that data about text difficulty “were either absent or
insufficient” (p. 216). Barth et al. (2014) examined the effects
that student and text characteristics in the passages had on
oral reading fluency with approximately 1,700 students in
middle school and found that sight-word reading efficiency
and decoding skills both were significantly related to reading
fluency outcomes. Passage characteristics that were studied
included difficulty, as measured by Lexile (MetaMetrics,
2023), type of passage (narrative or expository), page length,
word concreteness, and cohesion of the words, sentences,
and clauses. Although multiple measures significantly pre-
dicted reading fluency, passage difficulty accounted for more
than 50% of the total variance (Barth et al., 2014),
The difficulty of reading material is often evaluated by
determining if it represents a student’s instructional level or
not. Betts (1946) coined the phrase “instructional level” by
noting that students were more successful readers if they
read 95% of the words correctly, which was later defined as
the level of difficulty for students in which a reader is chal-
lenged but not overwhelmed (Morris et al., 2017). Betts’s
(1946) observation started an industry of assessments and
considerable interest by practitioners in how to best identify
an instruction level for reading intervention and instruction.
Some researchers defined an instructional level for reading
passages used in fluency interventions by percentage of the
words read correctly (Burns, 2007; Carr et al., 2022;
Treptow et al., 2007) or a criterion based on reading rate
(Hawkins et al., 2011; Martens et al., 2007), and some
implemented interventions with passages that matched the
students’ current grade level (Ardoin et al., 2008; Paige,
2011). Although seldomly reported, texts used in reading
fluency interventions that were too difficult for the students
led to low rates of reading growth (Parker & Burns, 2014)
and low comprehension (Treptow et al., 2007) in compari-
son to text that provided a more appropriate level of chal-
lenge. Next, I will discuss methods used to assess passage
difficulty during reading fluency interventions and contex-
tualize the different approaches within the construct of an
instructional level.
Informal Reading Inventories
After recognizing that successful readers should be able to
read 95% of words in a text, Betts later reportedly rethought
the specific numerical criterion and published the first
informal reading inventory (IRI) to help teachers qualita-
tively observe student behavior while reading passages that
represented different grade levels (Johns, 1991). IRIs are
types of assessments that are designed to determine an
instructional level for individual students (e.g., Basic
Reading Inventory [BRI; Johns, 2017], Benchmark
Assessment System [BAS; Fountas & Pinnell, 2007],
Developmental Reading Assessment [DRA; Beaver &
Carter, 2006]). The term instructional level has become
somewhat synonymous with using an IRI to assess students
and to place them into a series of books organized according
to a purported gradient of difficulty (Schwanenflugel &
Knapp, 2017; Shanahan, 2011).
Using an IRI to measure a student’s instructional level to
place that student into a book series was reportedly used by
43% of respondents in a national survey of K-2 reading
teachers, which was the most frequently reported approach
to reading instruction selected in the survey (EdWeek
Research Center, 2020). However, there have been frequent
criticisms about the reliability of IRI data (Bunch, 2017;
Nilsson, 2013), the validity of the resulting decisions (Burns
et al., 2015; Gandy, 2013; Parker et al., 2015), and the effec-
tiveness of teaching reading with leveled books based on
IRIs (Hoffman, 2017). In fact, Betts himself stated that IRIs
are “no more valid than the person who gives them” (Johns,
1991, p. 493).
Accuracy Criteria
Shortly after Betts started writing the first IRI, Hargis
(1987) and Gickling and Havertape (1981) broadened Bett’s
(1946) accuracy criterion of 95% and researched an instruc-
tional level of 93% to 97% correctly read words. An accu-
racy level below 93% represents a frustration level, and
surpassing 97% marks an independent level, which may
result in boredom. Thus, the accuracy definition of an
instructional level represents the interaction between a stu-
dent’s skill and the text, rather than a universally applicable
reading level standard. Moreover, the accuracy criterion for
measuring an instructional level was developed from inter-
vention research because the researchers aimed to identify
reading materials that provided an appropriate level of chal-
lenge to optimize student learning (Coulter & Coulter,
1989; Gickling & Armstrong, 1978).
Practitioners seeking to assess instructional levels with
the accuracy criteria can utilize curriculum-based assess-
ment for instructional design (CBAID; Burns & Parker,
2014; Coulter & Coulter, 1989; Gickling & Havertape,
1981), which involves conducting a 1-minute sampling of
reading material, calculating the percentage of words read
correctly within 2 seconds, multiplying it by 100, and inter-
preting the results according to the instructional level crite-
ria of 93%–97% accurately read words. Data from CBAID
exhibit robust reliability estimates, including interscorer
reliability ranging from .89 to .99, internal consistency from
.87 to .96, alternate form-reliability from .80 to .86, and
test-retest reliability from .82 to .96 (Burns et al., 2000).
Ample evidence also supports the validity of decisions
derived from CBAID data (Burns, 2007; Parker & Burns,
2014).
Providing students with reading material matched to an
instructional level of 93%–97% known has consistently
Burns 3
yielded increased task engagement and reading comprehen-
sion compared to materials that are either too challenging or
too easy (Burns & Parker, 2014; Gickling & Armstrong,
1978; Treptow et al., 2007), even among students with
behavior disorders (Carr et al., 2022). Furthermore, pre-
teaching words to ensure a minimum of 93% accuracy led
to improved reading growth over time (Burns, 2007; Parker
& Burns, 2014). Burns (2007) found a strong correlation (r
= .80) between student growth and the frequency at which
third-grade students with reading disabilities read 93%–
97% of the words correctly.
Fluency Criteria
Around the same time that Gickling and Armstrong (1978)
conducted research on the accuracy definition of an
instructional level, Deno and Mirkin (1977) also published
criteria for assessing an instructional level that was based
on the number of words read correctly in 1 minute. Deno
and Mirkin used the Starlin and Starlin (1973) estimate
from precision teaching of 70–119 words read correctly
per minute (WCPM). In other words, if a student could
read a book or passage with a rate that fell within 70–119
WCPM, then that passage or book would represent an
instructional level. Less than 70 WCPM would be a frus-
tration level, and more than 119 WCPM would be an inde-
pendent level. Shapiro (1996) then discussed the use of
rate criteria to conduct a survey-level assessment in an
assessment-to-intervention model so that students read
passages of varying difficulty until they read one with
70–119 WCPM, which would represent their instructional
level. Although the rate criteria outlined by Deno and
Mirkin (1977) and used by Shapiro (1996) were widely
used in practice and research (e.g., Hawkins et al., 2011;
Malouf et al., 2014; Martens et al., 2007), there were no
published studies addressing the reliability or validity of
the decisions other than the considerable evidence of the
reliability of assessing WCPM with curriculum-based
measurement (Reschly et al., 2009).
Purpose
Given that (a) fluency interventions had positive yet incon-
sistent effects (Hudson et al., 2020; Lee & Yoon, 2017; Maki
& Hammerschmidt-Snidarich, 2022), (b) passage difficulty
moderated the effects of fluency interventions (Barth et al.,
2014; Parker & Burns, 2014), and (c) previous meta-analytic
research did not examine the effect of passage difficulty on
the results (Lee & Yoon, 2017), the current study aimed to
meta-analyze research about reading fluency interventions
to examine the effect that assessing the difficulty level of the
passages had on the results. The following research ques-
tions guided the study: (a) What are the effects of student
reading difficulty level (no risk, at risk, or reading disability)
on reading fluency intervention outcomes? (b) What are the
effects of type of reading fluency intervention on reading
outcomes? And (c) what are the effects of reading fluency
interventions conducted with different approaches to mea-
suring an instructional level as an estimate of passages
difficulty?
Method
The research questions were addressed with meta-analytic
methods from an exhaustive search of electronic databases
and published articles. The processes for both are described
next.
Search Procedure
The search for studies began by searching Academic Search
Premier, PsycInfo, and Education Full Text databases with
the terms “reading fluency,” “instructional level,” “reading
level,” and “repeated reading,” with no restriction on year
of publication, which resulted in 8,051 records for potential
inclusion. The records were compared to the following
inclusion criteria:
1. Reported an original study.
2. Implemented an intervention to enhance reading
fluency by reading text.
3. Reported the difficulty level of reading text used
within the intervention and the method for deter-
mining that level.
4. Conducted the research with school-aged children
and youth.
5. Studied the intervention with a randomized experi-
mental or quasi-experimental group design, or a
single-case experimental design.
6. Measured the effect of the intervention on at least
one reading outcome.
7. Reported enough data to compute an effect size (i.e.,
means and standard deviations for group designs
and graphed individual data points for single-case
designs).
8. Written in English.
As shown in the figure in the Supplemental Materials, a
large number of articles were excluded by only reviewing
the titles and abstracts. The full text of 69 studies was
reviewed and compared to the inclusion criteria, and 49
were excluded.
After conducting the search, 20 unique records were
identified for inclusion. The references of the included arti-
cles were also examined with the inclusion criteria to iden-
tify any additional records. One more study was identified
and included in the meta-analysis, which brought the total
number of included studies to 21.
4 Assessment for Effective Intervention 00(0)
Coding
Three variables were coded from the included studies. Each
variable is described next.
Student Group. The general reading skills of the students
in each study were categorized according to what was
reported by the study authors. A total of eight studies
reported participants with no identified reading difficul-
ties, eight studies were conducted with students identified
as at-risk for reading difficulties, and six studies were con-
ducted with students who reportedly were diagnosed with
a disability. O’Connor et al. (2007) reported data sepa-
rately for students who were at risk for reading problems
and for those diagnosed with a learning disability in read-
ing and were coded separately. Carr et al. (2022) con-
ducted research with students who were reportedly
identified with an emotional or behavioral disorder in
addition to a reading difficulty, and the remaining five
studies with students with a disability implemented a flu-
ency intervention with students with a learning disability
in reading.
Intervention. The studies were coded according to type of
fluency intervention. Nine studies examined repeated read-
ing (n = 9) as outlined by Samuels (1979) in which stu-
dents individually read the same passage multiple times.
Five of the studies implemented repeated reading in a dyad
or small group and was coded as repeated reading in a dyad
or group. Three studies were coded as repeated reading
with support because the intervention included additional
components (e.g., intensive error correction or explicit
instruction in word reading). Finally, four studies had stu-
dents read passages but did not include a repeated reading
component.
Assessing an Instructional Level. Each of the studies was
coded into one of four categories based on how it assessed
passage difficulty. Six studies used percentage of words
read correctly to assess the instructional level, which were
coded as accuracy, and all but one of those used 93%–97%.
Studies that used an IRI to identify an instructional level, or
that used some other approach with levels (e.g., having a
student read a passage that was two grades below their cur-
rent grade), were identified as leveling (n = 7). Seven stud-
ies used the Deno and Mirkin (1977) rate criteria to conduct
a survey-level assessment (Shapiro, 1996) to identify an
instructional level and were categorized as rate criteria.
Finally, three studies had students read passages that were
purportedly written at their current grade level without con-
sidering the instructional level and were categorized as
grade level. Studies that used multiple methods to assess
task difficulty were coded separately.
Reliability
A second trained independent rater also coded the 69
records in the initial pool for inclusion to assess interob-
server agreement (IOA) on initial inclusion decisions. The
total number of records in which both raters judged to be
appropriate for inclusion or inappropriate for inclusion was
divided by the total number of studies and multiplied by
100% to calculate percentage of IOA. The two judges
agreed 97.1% of the time. The two instances of disagree-
ment were discussed, and consensus was reached that both
records should be included in the study.
A second trained and independent coder also coded vari-
ables for the 21 studies to estimate IOA for coding of the
variables (grade group, student group, intervention, instruc-
tional level measure, and dependent variable). The total
number of agreements between the two coders was divided
by the total number of codings and multiplied by 100%,
which resulted in 96.5% IOA between the two raters.
Disagreements were discussed by the two raters, a consen-
sus was reached, and data were recoded as needed before
conducting analyses.
Analyses
Of the 21 studies, 14 were between-group designs, and 7
were single-case designs (SCDs), which necessitated two
different estimates of effect. Data from the 14 between-
group designs were converted to Hedges’s g with an adjust-
ment for small sample sizes, which was identified as the
preferred effect size for continuous data (What Works
Clearinghouse, 2022). SCD data were converted to between-
case standardized mean differences (BC-SMD; Hedges
et al., 2012) using an online web application (Pustejovsky
et al., 2023). BC-SMD is advantageous over other SCD
effect sizes because it averages effects across individuals
within the study and relies less on within-individual vari-
ance (Valentine et al., 2016).
After computing an effect size (g or BC-SMD), each was
then weighted with the inverse of the variance using a ran-
dom-effects model. A random effect was used because it
was assumed that there was true variability in the effects
and in the resulting observed effects (Nikolakopoulou et al.,
2014). Studies that used multiple assessments of the same
construct were combined by computing the mean effect for
that construct. For example, if a study used three measures
of reading fluency, then the effect sizes were computed for
each outcome and averaged, but effects for different mea-
sures (e.g., comprehension and fluency) were each entered
separately into the dataset. Effect sizes were interpreted
using the commonly accepted interpretive scheme of 0.20
for a small effect, 0.50 for a medium effect, and 0.80 for a
large effect (Cohen, 1988). The third research question was
Burns 5
also examined by regressing the effect sizes onto three
study variables to determine the extent to which the student
group, intervention type, and estimate of difficulty pre-
dicted variance in the effects. R2 was used as an estimate of
effect, and an effect of .01 was considered small, .09 as
medium, and .25 as large (Cohen, 1988).
Results
The 21 studies included 829 K-12 students: 17 of the stud-
ies were implemented with students in kindergarten through
fifth grades, 3 were with students in middle school, and 1
was with students in high school. The 21 studies resulted in
29 individual estimates of effect because seven studies
reported the data for separate outcomes and were entered
separately. The 29 effects are displayed in Figure 1, and the
21 studies are described in more detail in the table included
in the Supplemental Materials. The 29 effects were exam-
ined for outliers by converting each to a Z score and identi-
fying any score with an absolute value that exceeded
±1.645 (p < .05) as an outlier. Two studies (Carr et al.,
2022; Fascio-Vereen, 2004) were identified as outliers and
were excluded from the analyses, which resulted in 27 total
effects.
As shown in Table 1, the overall weighted effect was
0.43 (95% CI = [0.25, 0.62]), which suggested an overall
small-to-medium weighted effect that was reliably different
from 0 (p < .001). Given that seven studies used an SCD
approach, the data were first analyzed by study design. The
21 effects from the between-group studies resulted in a
weighted effect of g = 0.40 (95% CI = [0.20, 0.61]), and
the 6 effects from SCD studies resulted in a weighted effect
of BC-SMD = 0.70 (95% CI = [0.22, 1.19]). Although the
weighted effect size from the SCD studies was almost twice
Figure 1. Forest Plot for Included Studies to Show Effect Sizes and the 95% Confidence Interval for Each.
6 Assessment for Effective Intervention 00(0)
as large as the one from between-group design studies, there
was considerable overlap between the two confidence inter-
vals, which suggested that the effects were not significantly
different. Thus, the data were aggregated for subsequent
analyses.
Estimates of publication bias were conducted by creat-
ing a funnel plot and with Egger regression. A visual analy-
sis of the funnel plot shown in Figure 2 suggested a mostly
symmetrically distribution, and the intercept of Egger
regression did not reveal a significant effect, t = 0.63, p =
.53, which did not suggest a publication bias.
The estimates of effect resulted in significantly heteroge-
neous data set Q = 47.35 (df = 26), p = .01 with an I2 of
0.48 and a T2 of .10. Thus, the effects generally exhibited
acceptable heterogeneity to conduct moderator analyses.
The data displayed in Table 1 present the estimates of effect
by potential moderating variables to address the research
questions.
Student Group
The first research question inquired about the extent that
student reading level affected the results. Also shown in
Table 1, the weighted effect size for students with a disabil-
ity was large (ES = 1.14, 95% CI = [0.64, 1.65]), and the
two remaining groups resulted in small weighted effects of
0.27 (95% CI = [0.08, 0.46]) for students without the risk
for reading difficulties and 0.37 (95% CI = [0.11, 0.64]) for
students at-risk for reading difficulties. All three effects
were significantly (p < .01) different than 0.
Intervention
The second research question inquired about the effects of
types of reading fluency interventions. Repeated reading
interventions were essentially equally effective when con-
ducted one on one (ES = 0.48, 95% CI = [0.20, 0.75], p <
.01) or in a dyad or small group (ES = 0.40, 95% CI =
[0.15, 0.65], p < .01), and both represented a small to
medium weighted effect. Although simply reading at an
instruction level was the largest effect (weighted ES = 0.85,
95% CI = [−0.24, 1.94], p = .12) and repeated reading with
additional support led to the smallest weighted effect
(weighted ES = 0.19, 95% CI = [−0.11, 0.50], p = .22),
neither were significantly different from 0, and the confi-
dence intervals for all four intervention types overlapped.
Thus, repeated reading was likely equally effective when
implemented in a small group or individually.
Assessing an Instructional Level
The third research question inquired about how an instruc-
tional level was assessed during reading fluency interventions.
Table 1. Estimates of Effect for Moderator Variables.
Variable kWeighted effect sizea95% CI p
Instructional level determined
Accuracy 6 1.03 0.65 to 1.40 <.01
Leveling system 10 0.31 −0.05 to 0.66 .09
Fluency criteria 7 0.29 0.07 to 0.50 .01
Current grade level 4 0.31 −0.03 to 0.64 .08
Student group
No risk 11 0.27 0.08 to 0.46 <.01
At-risk for reading failure 12 0.37 0.11 to 0.64 <.01
Disability 4 1.14 0.64 to 1.65 <.01
Intervention
Repeated reading 12 0.48 0.20 to 0.75 <.01
Repeated reading in dyad or group 8 0.40 0.15 to 0.65 <.01
Repeated reading with support 4 0.19 −0.11 to 0.50 .22
Reading at instructional level 3 0.85 −0.24 to 1.94 .12
Total 27 0.43 0.25 to 0.62 <.001
aEffect sizes consist of g for 21 effects from between-group designs and between cases standardized mean difference for six effects from single-case
designs.
Figure 2. Funnel Plot to Examine the Potential for Publication
Bias.
Burns 7
As shown in Table 1, using the instructional level criterion of
93%–97% correctly read words led to a large weighted effect g
= 1.03 (95% CI = [0.65, 1.40]) that was significantly different
(p < .01) from 0 and was the largest effect noted among the
four approaches for estimating difficulty. Using the rate criteria
for an instructional level led to a small weighted effect, g =
0.29 (95% CI = [0.07, 0.50]), which was reliably (p = .01)
different from 0 but also reliably smaller than the weighted
effect for the accuracy criteria (p < .05). The weighted effects
for using a leveling system (p = .09) and presenting reading
material that was at the students’ current grade level (p = .08)
also led to small weighted effects that were not reliably differ-
ent from 0. Using grade-level material scored reliably lower (p
< .05) than assessing an instructional level with accuracy cri-
teria but overlapped all other confidence intervals. Using level-
ing systems also overlapped the confidence intervals for the
effects from the other approaches for estimating an instruc-
tional level.
The research question was further examined with the
regression shown in Table 2. The student reading skill
groups significantly predicted the weighted effects and
accounted for a large amount of variance. The type of read-
ing fluency intervention did not significantly predict the
weighted effects and accounted for negligible variance.
Adding in the approach to assess difficulty accounted for an
additional 11% of the variance beyond the other two vari-
ables, which was a significant (p < .05) and medium effect.
Discussion
The current study meta-analyzed research about reading
fluency interventions to examine the effects that the diffi-
culty level of the passages had on the results. The difficulty
level of the reading passage can change the effectiveness of
reading fluency interventions (Barth et al., 2014; Parker &
Burns, 2014) and should be considered when designing
interventions like repeated reading. However, the difficulty
level of the passage and how it was determined is often not
reported in reading fluency intervention research (Lee &
Yoon, 2017). Using an accuracy criterion to determine an
instructional level appeared to be the best approach because
it resulted in the largest effect that was reliably different
from 0 and reliably larger than using rate criteria based on
WCPM to estimate an instructional level. Previous research
found that rate-based data were psychometrically superior
to accuracy data for screening and progress monitoring
decisions (Vanderheyden & Solomon, 2023), but the cur-
rent data suggest that accuracy was superior to rate for
designing intervention, and perhaps the two sources of
information together can fully inform an assessment-to-
intervention model (Burns, 2023; Shapiro, 2010).
The current data were also consistent with previous
research that found positive effects of reading fluency inter-
ventions and repeated reading (Keyes et al., 2016; Young
et al., 2015). Lee and Yoon (2017) found an overall effect of
g = 0.59 (95% CI = [0.56, 0.63]), which compared to the
overall weighted effect found here of 0.43, but Lee and Yoon
(2017) focused on students with reading disabilities. The
effect for students with disabilities in the current study was
ES = 1.14, but that was based on only four studies and should
be interpreted carefully. The Maki and Hammerschmidt-
Snidarich (2022) meta-analysis found an effect of g = 0.46
(95% CI = [0.23, 0.68]), which was almost identical in mag-
nitude to the overall weighted effect found here.
Potential Implications for Research
The method with which researchers assessed the difficulty
of passages used within reading fluency interventions
seemed to affect the results, with accuracy criteria leading
to the largest effect. However, there were only six effects
based on accuracy criteria, which suggested an area for
future research. The current meta-analysis included 21 stud-
ies, which was either consistent with previous meta-analy-
sis (k = 16; Hudson et al., 2020; k = 22; Maki &
Hammerschmidt-Snidarich, 2022; k = 18, Therrien, 2004)
or considerably fewer (k = 34; Lee & Yoon, 2017). One of
the inclusion criteria for the current study was that each
study had to report the purported level at which the reading
materials were written, and 15 located studies did not
Table 2. Regression of Weighted Effects of Fluency Intervention on Type of Intervention and Method to Determine Difficulty Level.
Variable
Model 1 Model 2 Model 3
B SE Beta t B SE Beta t B SE Beta t
Constant −0.16 0.22 −0.73 −0.21 0.27 −0.76 0.36 0.37 0.97
Student group 0.38 0.12 0.55 3.26* 0.38 .012 .55 3.19* 0.32 0.12 .46 2.77*
Reading intervention 0.03 0.08 .05 0.30 −0.01 0.08 <.001 −0.02
Method to determine difficulty −0.18 0.08 −.35 −2.11*
R2 = .30, Δ = .30
F = 10.62*
R2 = .30, Δ < .001
F = 0.09
R2 = .41, Δ = .11
F = 4.44*
*p < .05.
8 Assessment for Effective Intervention 00(0)
include that information. Future researchers should con-
sider including information about the difficulty level of the
text used during intervention.
Potential Implications for Practice
There are multiple implications for practice from the cur-
rent meta-analysis. First, fluency interventions, including
repeated reading, were again found to have a positive effect
on reading. Thus, repeated reading continues to be an evi-
dence-based practice, especially for children with reading
disabilities. However, classwide approaches and repeated
reading that occurred in dyads appeared to be as effective as
repeated-reading interventions that occurred one-on-one
with students without disabilities and may provide a more
efficient option. Moreover, other approaches to enhance
reading fluency (e.g., continuous reading) appeared to be
equally effective as repeated reading, which was also con-
sistent with previous research (Maki & Hammerschmidt-
Snidarich, 2022) and suggested multiple options to
effectively enhance reading outcomes.
Second, practitioners could identify reading passages for
reading fluency interventions by recording the percentage
of words read correctly and comparing it to the accuracy
criterion of 93% to 97%. The current data also suggest that
additional research is needed to examine the rate criteria
suggested by Deno and Mirkin (1977) and reported by
Shapiro (1996) to conduct survey-level assessments.
However, given the small effects noted here, practitioners
could be cautious about using the rate criteria until addi-
tional research validates that approach.
Finally, the small effect for using leveling systems such
as IRIs provides additional evidence about the lack of valid-
ity in using those tools during instruction or intervention
(Burns et al., 2015; Gandy, 2013; Hoffman, 2017). However,
the practice remains prevalent in schools (EdWeek Research
Center, 2020). Practitioners could consider the small effects
found here, and in previous research, and be skeptical about
using IRIs in instruction or intervention.
Limitations
Although the current meta-analysis was consistent with pre-
vious research and the findings added to the literature, the
conclusions should be considered within the context of their
limitations. First, every study included students in second
through sixth grades, and the implications for children in
other grades and for adult learners were not clear. Additional
research is needed to examine the effects of assessing an
instructional level with children younger than second grade
and older than sixth grade. Second, students were identified
as having a reading disability or not by study researchers,
but the criteria to make that designation were often not pre-
sented nor confirmed. Third, the category identified as “no
risk” may have included all students in a particular class or
group and may have included a mix of reading abilities, but
the data were not disaggregated by any particular group,
and no risk was noted. Additional research is needed to
determine the effect of fluency interventions and of reading
at an instructional level for students with average or above-
average reading skills. Fourth, the effects from SCDs and
between-group designs were aggregated, and research
regarding the best process for doing so remains ongoing.
Finally, the study only included 21 studies, which may have
not provided sufficient power to examine some potential
moderating variables.
Conclusion
The instructional level continues to be a confusing construct
that is often conflated with a measurement system that has
been consistently refuted by research (i.e., IRIs). The cur-
rent data suggest that the instructional level is a construct
worth additional research within the context of reading flu-
ency interventions and that perhaps the instructional level is
an interface between individual student skill and text com-
plexity that could be assessed with a sampling of percentage
of words read correctly. Additional research is needed to
better understand the construct and the role of text difficulty
during fluency interventions, but given the number of stu-
dents experiencing reading difficulties in schools, addi-
tional research seems warranted.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
ORCID iD
Matthew K. Burns https://orcid.org/0000-0001-9946-5729
Supplemental Material
Supplemental material is available on the Assessment for Effective
Intervention webpage with the online version of the article
References
*References marked with an asterisk indicate studies included in
the meta-analysis.
*Ardoin, S. P., Eckert, T. L., & Cole, C. A. (2008). Promoting
generalization of reading: A comparison of two fluency-based
interventions for improving general education student’s oral
reading rate. Journal of Behavioral Education, 17, 237–252.
https://doi.org/10.1007/s10864-008-9066-1
Barth, A. E., Tolar, T. D., Fletcher, J. M., & Francis, D. (2014).
The effects of student and text characteristics on the oral
Burns 9
reading fluency of middle-grade students. Journal of
Educational Psychology, 106(1), 162–180. https://doi.org/10
.1037/a0033826
Basaran, M. (2013). Reading fluency as an indicator of reading
comprehension. Educational Sciences: Theory and Practice,
13(4), 2287–2290. https://doi.org/10.12738/estp.2013.4.1922
Beaver, J. M., & Carter, M. A. (2006). The Developmental Reading
Assessment–Second edition (DRA2). Pearson.
Betts, E. (1946). Foundations of reading instruction. American
Book Company.
*Brown, L. T., Mohr, K. A., Wilcox, B. R., & Barrett, T. S.
(2018). The effects of dyad reading and text difficulty on
third-graders’ reading achievement. Journal of Educational
Research, 111(5), 541–553. https://doi.org/10.1080/0022067
1.2017.1310711
Bunch, M. B. (2017). Test review of the basic reading inventory,
eleventh edition. In J. F. Carlson, K. F. Geisinger, & J. L.
Jonson (Eds.), The twentieth mental measurements yearbook
(pp. 112–113). Buros Center for Testing.
*Burns, M. K. (2007). Reading at the instructional level with chil-
dren identified as learning disabled: Potential implications
for response-to-intervention. School Psychology Quarterly,
22(3), 297–313. https://doi.org/10.1037/1045-3830.22.3.297
Burns, M. K. (2023). Examining the learning hierarchy with
accuracy and rate scores for reading fluency among second-
and third-grade students. Journal of Behavioral Education.
Advance online publication. https://doi.org/10.1007/s10864-
023-09536-2
Burns, M. K., & Parker, D. C. (2014). Curriculum-based assess-
ment for instructional design: Using data to individualize
instruction. Guilford Publications.
*Burns, M. K., Pulles, S. M., Maki, K. E., Kanive, R., Hodgson,
J., Helman, L. A., McComas, J. J., & Preast, J. L. (2015).
Accuracy of student performance while reading leveled
books rated at their instructional level by a reading inventory.
Journal of School Psychology, 53(6), 437–445. https://doi.
org/10.1016/j.jsp.2015.09.003
Burns, M. K., Tucker, J. A., Frame, J., Foley, S., & Hauser,
A. (2000). Interscorer, alternate- form, internal consis-
tency, and test-retest reliability of Gickling’s model of
curriculum-based assessment for reading. Journal of
Psychoeducational Assessment, 18(4), 353–360. https://doi.
org/10.1177/073428290001800405
*Carr, C. E., Umbreit, J., & Hartzell, R. (2022). Instructional
level and engagement in students with behavioral disor-
ders. Behavioral Disorders, 47(4), 236–244. https://doi.
org/10.1177/01987429211050014
Cohen, J. (1988). Statistical power analysis for the behavioral sci-
ences (2nd ed.). Lawrence Erlbaum Associates.
Coulter, W. A., & Coulter, E. M. (1989). Curriculum-based
assessment for instructional design (CBAID): The trainers
handbook. Directions & Resources Group.
Deno, S. L., & Mirkin, P. K. (1977). Data-based program modifi-
cation: A manual. Council for Exceptional Children.
EdWeek Research Center. (2020). Early reading instruction:
Results of a national survey. Editorial Projects in Education.
*Fascio-Vereen, S. T. (2004). A comparison of reading interven-
tions using social studies text to improve oral reading fluency
of middle school students [Doctoral dissertation]. Mississippi
State University.
Fountas, I. C., & Pinnell, G. S. (2007). Fountas and Pinnell
Benchmark Assessment System 2. Heinemann.
Fuchs, L. S., Fuchs, D., Hosp, M. K., & Jenkins, J. R. (2001). Oral
reading fluency as an indicator of reading competence: A the-
oretical, empirical, and historical analysis. Scientific Studies
of Reading, 5(3), 239–256.
Gandy, S. E. (2013). Informal reading inventories and ELL stu-
dents. Reading & Writing Quarterly, 29(3), 271–287. https://
doi.org/10.1080/10573569.2013.789782
Gickling, E. E., & Armstrong, D. L. (1978). Levels of instructional
difficulty as related to on-task behavior, task completion, and
comprehension. Journal of Learning Disabilities, 11(9), 559–
566. https://doi.org/10.1177/00222194780110090
Gickling, E. E., & Havertape, S. (1981). Curriculum-based assess-
ment (CBA). School Psychology Inservice Training Network,
University of Minnesota.
Hargis, C. H. (1987). Curriculum-based assessment: A primer.
Charles C. Thomas.
*Hawkins, R. O., Hale, A., Sheeley, W., & Ling, S. (2011).
Repeated reading and vocabulary-previewing interventions
to improve fluency and comprehension for struggling high-
school readers. Psychology in the Schools, 48(1), 59–77.
https://doi.org/10.1002/pits.20545
Hedges, L. V., Pustejovsky, J. E., & Shadish, W. R. (2012). A
standardized mean difference effect size for single case
designs. Research Synthesis Methods, 3, 224–239. https://doi.
org/10.1002/jrsm.1052
Hoffman, J. V. (2017). What if “just right” is just wrong? The
unintended consequences of leveling readers. The Reading
Teacher, 71(3), 265–273. https://doi.org/10.1002/trtr.1611
*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(2),
94–99. https://doi.org/10.1080/00220671.1993.9941172
Hudson, A., Koh, P. W., Moore, K. A., & Binks-Cantrell, E.
(2020). Fluency interventions for elementary students with
reading difficulties: A synthesis of research from 2000–2019.
Education Sciences, 10(3), 52–77. https://doi.org/10.3390/
educsci10030052
Johns, J. L. (1991). Emmett A. Betts on informal reading inven-
tories. Journal of Reading, 34(6), 492–493. http://www.jstor.
org/stable/40032112
Johns, J. L. (2017). Basic reading inventory (11th ed.). Kendall
Hunt.
Keyes, S. E., Cartledge, G., Gibson, Jr., L., & Robinson-Ervin, P.
(2016). Programming for generalization of oral reading flu-
ency using computer-assisted instruction and changing flu-
ency criteria. Education and Treatment of Children, 39(2),
141–172. https://doi.org/10.1353/etc.2016.0011
Kim, Y. S. G., & Wagner, R. K. (2015). Text (oral) reading flu-
ency as a construct in reading development: An investigation
of its mediating role for children from grades 1 to 4. Scientific
Studies of Reading, 19(3), 224–242. https://doi.org/10.1080/1
0888438.2015.1007375
Lee, J., & Yoon, S. Y. (2017). The effects of repeated reading on
reading fluency for students with reading disabilities: A meta-
analysis. Journal of Learning Disabilities, 50(2), 213–224.
https://doi.org/10.1177/0022219415605194
*LeVasseur, V. M., Macaruso, P., & Shankweiler, D. (2008).
Promoting gains in reading fluency: A comparison of three
10 Assessment for Effective Intervention 00(0)
approaches. Reading and Writing, 21, 205–230. https://doi.
org/10.1007/s11145-007-9070-1
Maki, K. E., & Hammerschmidt-Snidarich, S. (2022). Reading
fluency intervention dosage: A novel meta-analysis and
research synthesis. Journal of School Psychology, 92, 148–
165. https://doi.org/10.1016/j.jsp.2022.03.008
*Maki, K. E., Ittner, A., Pulles, S. M., Burns, M. K., Helman,
L., & McComas, J. J. (2023). Effects of an abbreviated
class-wide reading intervention for students in third grade.
Contemporary School Psychology, 26(3), 359–367. https://
doi.org/10.1007/s40688-020-00343-4
*Malouf, R., Reisener, C., Gadke, D., Wimbish, S., & Frankel, A.
(2014). The effect of helping early literacy with practice strate-
gies on reading fluency for children with severe reading impair-
ments. Reading Improvement, 51(2), 269–279.
*Martens, B. K., Eckert, T. L., Begeny, J. C., Lewandowski, L. J.,
DiGennaro, F. D., Montarello, S. A., Arbolino, L. A., Reed,
D. D., & Fiese, B. H. (2007). Effects of a fluency-building
program on the reading performance of low-achieving second
and third grade students. Journal of Behavioral Education,
16, 38–53. https://doi.org/10.1007/s10864-006-9022-x
MetaMetrics. (2023). Lexile grade level charts: What are Lexile
student measures by grade? Lexile and Quantile Hub. https://
hub.lexile.com/lexile-grade-level-charts
Morris, D., Meyer, C., Trathen, W., McGee, J., Vines, N., Stewart,
T., Gill, T., & Schlagal, R. (2017). The simple view, instruc-
tional level, and the plight of struggling fifth-sixth-grade read-
ers. Reading & Writing Quarterly, 33(3), 278–289. https://
doi.org/10.1080/10573569.2016.1203272
*Musti-Rao, S., Hawkins, R. O., & Barkley, E. A. (2009).
Effects of repeated readings on the oral reading fluency
of urban fourth-grade students: Implications for practice.
Preventing School Failure: Alternative Education for
Children and Youth, 54(1), 12–23. https://doi.org/10.3200/
PSFL.54.1.12-23
National Reading Panel. (2000). Teaching children to read: An
evidence-based assessment of the scientific research litera-
ture on reading and its implications for reading instruction:
Reports of the subgroups. National Institute of Child Health
and Human Development, National Institutes of Health.
Nikolakopoulou, A., Mavridis, D., & Salanti, G. (2014).
Demystifying fixed and random effects meta-analysis.
Evidence-based Mental Health, 17(2), 53–57. https://doi.
org/10.1136/eb-2014-101795
Nilsson, N. L. (2013). The reliability of informal reading invento-
ries: What has changed? Reading & Writing Quarterly, 29(3),
208–230. https://doi.org/10.1080/10573569.2013.789779
*O’Connor, R. E., White, A., & Swanson, H. L. (2007). Repeated
reading versus continuous reading: Influences on reading flu-
ency and comprehension. Exceptional Children, 74(1), 31–
46. https://doi.org/10.1177/001440290707400102
*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. https://doi.org/10.1080/10862968509547535
*Paige, D. D. (2011). 16 minutes of “Eyes-on-Text” can make a
difference: Whole-class choral reading as an adolescent flu-
ency strategy. Reading Horizons: A Journal of Literacy and
Language Arts, 51(1), 3–23. https://scholarworks.wmich.edu/
reading_horizons/vol51/iss1/3
*Parker, D. C., & Burns, M. K. (2014). Using the instructional
level as a criterion to target reading interventions. Reading &
Writing Quarterly, 30(1), 79–94. https://doi.org/10.1080/105
73569.2012.702047
Parker, D. C., Zaslofsky, A. F., Burns, M. K., Kanive, R., Hodgson,
J., Scholin, S. E., & Klingbeil, D. A. (2015). A brief report of
the diagnostic accuracy of oral reading fluency and reading
inventory levels for reading failure risk among second- and
third-grade students. Reading & Writing Quarterly, 31(1),
56–67. https://doi.org/10.1080/10573569.2013.857970
Pustejovsky, J. E., Chen, M., Hamilton, B., & Grekov, P. (2023).
scdhlm: A web-based calculator for between-case standard-
ized mean differences (Version 0.7.2) [Web application].
https://jepusto.shinyapps.io/scdhlm
Reschly, A. L., Busch, T. W., Betts, J., Deno, S. L., & Long, J.
D. (2009). Curriculum-based measurement oral reading as
an indicator of reading achievement: A meta-analysis of the
correlational evidence. Journal of School Psychology, 47(6),
427–469. https://doi.org/10.1016/j.jsp.2009.07.001
Samuels, S. J. (1979). The method of repeated read-
ings. The Reading Teacher, 41, 756–760. https://doi.
org/10.1177/002221948301600205
Schwanenflugel, P. J., & Knapp, N. F. (2017, February 28). Three
myths about reading levels: And why you shouldn’t fall for them.
Psychology Today. https://www.psychologytoday.com/us/blog/
reading-minds/201702/three-myths-about-reading-levels
Shanahan, T. (2011). Rejecting instructional level theory. https://www.
shanahanonliteracy.com/blog/rejecting-instructional-level-theory
Shapiro, E. S. (1996). Academic skills problems: Direct assess-
ment and intervention. Guilford.
Shapiro, E. S. (2010). Academic skills problems: Direct assess-
ment and intervention (4th ed.). Guilford.
*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-perform-
ing students in regular classrooms. The Elementary School
Journal, 95(5), 387–408. https://doi.org/10.1086/461851
Starlin, C., & Starlin, A. (1973). Guides for continuous decision
making. Unique Curriculum Unlimited.
Therrien, W. J. (2004). Fluency and comprehension gains as a
result of repeated reading: A meta- analysis. Remedial and
Special Education, 25(4), 252–261. https://doi.org/10.1177/0
7419325040250040801
*Therrien, W. J., & Hughes, C. (2008). Comparison of repeated read-
ing and question generation on students’ reading fluency and
comprehension. Learning Disabilities: A Contemporary Journal,
6(1), 1–16. https://files.eric.ed.gov/fulltext/EJ797653.pdf
*Treptow, M. A., Burns, M. K., & McComas, J. J. (2007). Reading
at the frustration, instructional, and independent levels: The
effects on students’ reading comprehension and time on task.
School Psychology Review, 36(1), 159–166. https://doi.org/1
0.1080/02796015.2007.12087958
*Vadasy, P. F., & Sanders, E. A. (2008). Benefits of repeated
reading intervention for low- achieving fourth-and fifth-grade
students. Remedial and Special Education, 29(4), 235–249.
https://doi.org/10.1177/0741932507312013
Burns 11
Valentine, J. C., Tanner-Smith, E. E., Pustejovsky, J. E., & Lau,
T. S. (2016). Between-case standardized mean difference
effect sizes for single-case designs: A primer and tutorial
using the scdhlm web application. Campbell Systematic
Reviews, 12(1), 1–31. https://doi.org/10.4073/cmdp.2016.1
Vanderheyden, A. M., & Solomon, B. G. (2023). Valid outcomes for
screening and progress monitoring: Fluency is superior to accu-
racy in curriculum-based measurement. School Psychology,
38(3), 160–172. https://doi.org/10.1037/spq0000528
What Works Clearinghouse. (2022). What Works Clearinghouse
procedures and standards handbook (Version 5.0). U.S.
Department of Education, Institute of Education Sciences,
National Center for Education Evaluation and Regional
Assistance (NCEE). https://ies.ed.gov/ncee/wwc/Handbooks
Young, C., Mohr, K. A., & Rasinski, T. (2015). Reading together:
A successful reading fluency intervention. Literacy Research
and Instruction, 54(1), 67–81. https://doi.org/10.1080/19388
071.2014.976678
... The adaptive decoding assessment effectively maintained an optimal challenge level, with 20-25% of items at low accuracy and the remainder split between medium and high accuracy. Having at least half of the items in targeted intervention practice be familiar or (i.e., medium and high accuracy) has been linked with large positive effects on student performance (Burns, 2024). The purposeful design of the Lexercise adaptive decoding assessment also aligns with guidelines about adjusting the level of difficulty for the learner and keeping individual practice sessions relatively short (Burns & Contesse, 2024;Poncy et al., 2015). ...
Research
Full-text available
This study examines the usage and effectiveness of Lexercise, a hybrid reading intervention program combining weekly literacy intervention with structured independent practice, in supporting reading skill development among struggling readers. Analysis of 198 students in grades 2-6 during the 2023-2024 school year revealed strong adherence to the prescribed intervention model of weekly 45-minute structured literacy lessons supplemented by four 15-minute independent practice sessions. Results showed a significant positive correlation between total program time and decoding mastery (r = 0.64, p < .001), with optimal outcomes observed at 24 and 33 weeks of consistent usage. A random forest analysis identified total hours spent, months of program use, and levels completed as the strongest predictors of decoding mastery. Cluster analysis revealed three distinct learner profiles: Early Learners (younger students with moderate usage), Consistent Progressors (older students with high adherence and strong mastery), and Complex Learners (students with high-level completion but limited mastery). While younger students progressed more quickly through program levels, older students required additional time for mastery, particularly at higher levels. These findings suggest that structured, technology-enhanced reading intervention can effectively support reading skill development when implemented with fidelity, though implementation strategies may need adjustment based on learner profiles.
... Teaching new skills so that they are sequenced to build on each other increases student learn-ing. Students who read 93% to 97% of the words correctly had better reading outcomes than those who read at higher or lower percentages (Burns, 2024). Reading 93% to 97% of the words correctly is a research-based definition of an appropriate level of challenge during intervention. ...
... Two meta-analyses have indicated that repeated readings at the instructional level, combined with error correction, performance feedback, and peer-mediated instruction, can improve reading fluency, especially for students with SLDs (Burns, 2024;Strickland et al., 2013). ...
Article
The field of specific learning disabilities (SLDs) has faced three kinds of challenges. Constitutional challenges arise from classification and definitional complexities, identification issues, comorbidity, and ontological debates. Internal challenges include the inherent difficulties of scientific thinking that compete with intuitionism and confirmation bias, the lack of randomized controlled trials to evaluate interventions, the low rate of replication studies, publication bias, and the gap between research on evidence-based practices and implementation. External challenges include philosophical movements in academia, primarily social constructionism and cognitive relativism. They also encompass broader social trends such as neurological reductionism, educational fads, and political conformism. This article specifically focuses on the influence of cognitive relativism in the field of SLDs. Despite these challenges, the field has made incremental progress by committing itself to scientific inquiry. The fundamental purpose of research is truth-seeking, aiming to expand the knowledge base on how best to support students with SLDs. The development of cognitive theories of dyslexia illustrates the refining nature of scientific inquiry as it moves closer to the truth. Upholding rigorous scientific standards is crucial for the future development of the field, providing effective support and consistently enhancing educational outcomes for students with SLDs.
... Two meta-analyses have indicated that repeated readings at the instructional level, combined with error correction, performance feedback, and peer-mediated instruction, can improve reading fluency, especially for students with SLDs (Burns, 2024;Strickland et al., 2013). ...
Article
Full-text available
The field of specific learning disabilities (SLDs) has faced three kinds of challenges. Constitutional challenges arise from classification and definitional complexities, identification issues, comorbidity, and ontological debates. Internal challenges include the inherent difficulties of scientific thinking that compete with intuitionism and confirmation bias, the lack of randomized controlled trials to evaluate interventions, the low rate of replication studies, publication bias, and the gap between research on evidence-based practices and implementation. External challenges include philosophical movements in academia, primarily social constructionism and cognitive relativism. They also encompass broader social trends such as neurological reductionism, educational fads, and political conformism. This article specifically focuses on the influence of cognitive relativism in the field of SLDs. Despite these challenges, the field has made incremental progress by committing itself to scientific inquiry. The fundamental purpose of research is truth-seeking, aiming to expand the knowledge base on how best to support students with SLDs. The development of cognitive theories of dyslexia illustrates the refining nature of scientific inquiry as it moves closer to the truth. Upholding rigorous scientific standards is crucial for the future development of the field, providing effective support and consistently enhancing educational outcomes for students with SLDs..
Article
Full-text available
Previous research used the learning hierarchy (LH) as a heuristic to select reading interventions based on the level of accuracy defined as the percentage of words read correctly. The current study examined the validity of the LH by reporting the prevalence of reading profiles proposed by the framework: Acquisition phase-inaccurate and slow, Proficiency phase-accurate and slow, and Generalization phase-accurate and fast to determine the extent to which the data could be used to drive reading interventions. The design also included a hypothetical phase of inaccurate and fast, which was not included in the LH. Reading fluency data from 223 second-and third-grade students were compared to accuracy (93%) and rate (national grade-level norms) criteria. When data were classified into the LH phases described above, 44.4% (n = 99) of the students were in the Acquisition phase, 23.8% (n = 53) were in the Proficiency phase, and 31.4% (n = 70) were in the Generalization phase. Less than 1% (n = 1) was in the hypothetical phase of inaccurate and fast, and the rarity of this occurrence was predicted by the LH. These data support the LH as a conceptual framework to drive diagnostic assessment, and the importance of examining accuracy data when designing reading fluency interventions.
Article
Full-text available
Curriculum-based measurement (CBM) has conventionally included accuracy criteria with recommended fluency thresholds for instructional decision making. Some scholars have argued for the use of accuracy to directly determine instructional need (e.g., Szadokierski et al., 2017). However, accuracy and fluency have not been directly examined to determine their separate and joint value for decision making in CBM prior to this study. Instead, there was an assumption that instruction that emphasized accurate responding should be monitored with accuracy data, which evolved into the use of complementing CBM fluency scores with accuracy or using timed assessment to compute percent of responses correct and using accuracy criteria to determine instructional need. The purpose of this paper was to examine fluency and accuracy as related but distinct metrics with psychometric properties and associated benefits and limits. Findings suggest that the redundancy between accuracy and fluency causes them to perform comparably overall, but that (a) fluency is superior to accuracy when accuracy is computed on a timed sample of performance, (b) timed accuracy adds no benefit relative to fluency alone, and (c) accuracy when collected under timed assessment conditions has substantial psychometric limitations that make it unsuitable for the formative instructional decisions which are commonly made using CBM data. The conventional inclusion of accuracy criteria in tandem with fluency criteria for instructional decision making in CBM should be reconsidered as there may be no added predictive value, but rather additional opportunity for error due to the problems associated with unfixed trials in timed assessment.
Article
Full-text available
Multi-tiered systems of support (MTSS) provide instruction and intervention to all students in schools. However, when many students within a setting demonstrate reading difficulties, there is likely a tier 1 problem which makes it infeasible to implement tier 2 or tier 3 interventions with large numbers of students due to limited resources. The current study experimentally examined the effects of a brief class-wide intervention with two third-grade classrooms with large numbers of students demonstrating reading difficulties. The treatment group participated in partner reading with paragraph shrinking (i.e., summarizing) every school day for 2 weeks. The intervention resulted in significantly greater fluency growth compared to the control classroom. Moreover, the number of students in the treatment group who scored below the seasonal benchmark criterion fell from 12 out of 21 to 6 of 21 students. The control classroom had 13 out of 24 students who scored below the benchmark before and after the intervention. Results pointed to a promising practice for schools to enact when class-wide reading difficulties are evident.
Article
Full-text available
Executive summary Single‐case research designs are critically important for understanding the effectiveness of interventions that target individuals with low incidence disabilities (e.g., physical disabilities, autism spectrum disorders). These designs comprise an important part of the evidence base in fields such as special education and school psychology, and can provide credible and persuasive evidence for guiding practice and policy decisions. In this paper we discuss the development and use of between‐case standardized mean difference effect sizes for two popular single‐case research designs (the treatment reversal design and the multiple baseline design), and discuss how they might be used in meta‐analyses either with other single‐case research designs or in conjunction with between‐group research designs. Effect size computation is carried out using a user‐friendly web application, scdhlm, powered by the free statistical program R; no knowledge of R programming is needed to use this web application.
Article
Full-text available
Oral reading fluency (ORF) deficits are a hallmark of reading difficulties. The impact of fluency struggles extends beyond word-level difficulties to include deficits in reading comprehension. Sixteen empirical studies conducted in 2000–2019 that examined ORF interventions among elementary students identified as having reading difficulties were reviewed to identify the characteristics (e.g., instructional variables, group size, type of interventionist) of effective ORF interventions and their impact on English oral reading fluency and reading comprehension outcomes. The systematic review revealed that interventions reported centered around repeated reading procedures (86.5%). Across the 16 studies, outcomes for oral reading fluency varied widely and most focused on speed and rate aspects rather than prosody. Effect sizes for rate and accuracy measures ranged from negligible to large (i.e., 0.01 to 1.18) and three studies found large effects for prosody outcomes. Effect sizes for reading comprehension ranged between non-significant and large significant effects. Findings support the use of repeated reading of text to build up ORF of students with reading difficulties. Interventions that were found to be most effective were those that were conducted one-on-one with a trained model of fluent word reading and accuracy. Findings also point to three gaps in our understanding: (1) the efficacy of interventions other than repeated reading, (2) effects of ORF interventions on prosody outcomes, and (3) sustainability of outcomes.
Article
Reading fluency is a fundamental component of reading proficiency, and yet, it is a common area of need for students demonstrating reading difficulties. In response, reading fluency interventions have been researched and implemented relatively extensively, but how much intervention is needed to affect reading fluency growth is not clear. The purpose of this study was to meta-analyze and synthesize the research base regarding targeted reading fluency interventions for kindergarten through 12th grade students and to examine how intervention dosage is conceptualized and measured in these interventions. Thirty-three articles met the study inclusion criteria, and overall effects of reading fluency interventions were moderate (effect size = 0.46, 95% CI [0.23, 0.68], p < .001), with study design (group design ES = 0.41, single case design ES = 0.75) and study rigor moderating study effects (group design met standards ES = 0.57, group design did not meet standards ES = 0.34, single case design met standards ES = 0.79, single case design did not meet standards ES = 0.57). Most studies conceptualized and reported intervention dosage as the amount of time spent in intervention with intervention duration impacting total intervention growth. Implications for practice and research are discussed, particularly the need for additional research examining proximal measurement of intervention dosage in an effort to best understand how opportunities to practice impact reading fluency growth.
Article
This study examined the effects of adjusting the difficulty level of instructional materials on the time on-task and comprehension of four students with emotional and behavioral disorders. All participants previously exhibited low rates of on-task behavior during reading assignments. Students were presented with reading materials at their instructional, frustration, and independent levels to assess the effect on time on-task and comprehension. All four students demonstrated the highest percent of on-task behavior when presented with reading materials at their instructional level. Comprehension scores were highest for all four students at the independent level and lowest at the frustration level.
Article
The author questions the "just right" leveling of student texts as having put limitations on students and teachers in promoting literacy. The historical background for the leveling of text and readers is described. The author identifies a number of unintended consequences associated with the use of leveling and guided reading. Alternatives to leveling and guided reading are presented, with a particular focus on work in informational texts in inquiry.
Article
This study replicated, with modifications, previous research of dyad reading using texts at various levels of difficulty (Morgan, 1997). The current project measured the effects of using above–grade-level texts on reading achievement and sought to determine the influences of dyad reading on both lead and assisted readers. Results indicate that weaker readers, using texts at two, three, and four grade levels above their instructional levels with the assistance of lead readers, outscored both proficient and less proficient students in the control group across multiple measures of reading achievement. However, the gains made by assisted readers were not significantly different relative to the various text levels. When all assessments were considered, assisted readers reading texts two grade levels above their instructional levels showed the most robust gains in oral reading fluency and comprehension. Lead readers also benefited from dyad reading and continued their respective reading developmental trajectories across measures.