See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/51499048

Response to Intervention: Preventing and Remediating Academic Difficulties

Article  in  Child Development Perspectives · April 2009

DOI: 10.1111/j.1750-8606.2008.00072.x · Source: PubMed

CITATIONS READS
289 1,109

2 authors:

Jack M Fletcher Sharon Vaughn

University of Houston University of Texas at Austin
550 PUBLICATIONS   38,991 CITATIONS    367 PUBLICATIONS   17,355 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Exploring Developmental Changes via Chronnectomics View project

Reading For Understanding View project

All content following this page was uploaded by Jack M Fletcher on 13 December 2018.

The user has requested enhancement of the downloaded file.

 CHILD DEVELOPMENT PERSPECTIVES

Response to Intervention: Preventing and

Remediating Academic Difficulties
Jack M. Fletcher1 and Sharon Vaughn2
1
University of Houston and 2The University of Texas at Austin

ABSTRACT—This article addresses the advantages and 2002). Some children are eventually identified with learning
challenges of service delivery models based on student disabilities (LDs), representing approximately 5% of the school-
response to intervention (RTI) for preventing and reme- age population and approximately 50% of students identified with
diating academic difficulties and as data sources for disabilities in schools (U.S. Department of Education, 2007).
identification for special education services. The primary A variety of state, federal, and district school-based programs
goal of RTI models is improved academic and behavioral attempt to address different obstacles to learning academic skills.
outcomes for all students. The evidence is reviewed for the With the 2002 passage of the No Child Left Behind Act, which
processes underlying RTI, including screening and prog- targets the needs of economically disadvantaged children through
ress monitoring assessments, evidence-based interventions, Title I funding, the federal government placed greater emphasis on
and schoolwide coordination of multitiered instruction. We early intervention, high-quality instruction, and accountability for
also discuss the secondary goal of RTI, which is to provide academic outcomes. In 2004, the Individuals With Disabilities
data for identification of learning disabilities (LDs). Incor- Education Act (IDEA; U.S. Department of Education, 2004),
porating instructional response into identification repre- which governs the provision of special education services in U.S.
sents a controversial shift away from discrepancies in public schools, was also reauthorized.
cognitive skills that have traditionally been a primary basis Noteworthy in the reauthorization was the emphasis on early
for LD identification. RTI processes potentially integrate intervention services and specific provisions allowing districts to
general and special education and suggest new directions adopt service delivery models that focus on the child’s response to
for research and public policy related to LDs, but the intervention (RTI). These models (a) screen all children for
scaling issues in schools are significant and more research academic and behavioral problems, (b) monitor the progress of
is needed on the use of RTI data for identification. children at risk for difficulties in these areas, and (c) provide
KEYWORDS—learning disabilities; response to intervention; increasingly intense interventions based on the response to
reading; IDEA progress monitoring assessments (Vaughn & Fuchs, 2003). Those
children who do not respond adequately may be referred for
a comprehensive evaluation for eligibility for special education
Children struggle to learn reading, mathematics, and writing skills services. Through the comprehensive evaluation, some children
for many reasons, including growing up in economically disad- will be eligible for special education and others may need
vantaged settings, low proficiency in English, emotional difficul- alternative services because their difficulties in learning are not
ties, and even inadequate academic instruction (Donovan & Cross, due to an LD or other type of disability consistent with a need for
special education.
Supported in part by a grant from the National Institute for Child Service delivery models that provide universal screening, prog-
Health and Human Development, 1 P50 HD052117, Texas ress monitoring, and tiered, or layered, interventions have been
Center for Learning Disabilities.
widely adopted in No Child Left Behind and Title I and are a specific
Correspondence concerning this article should be addressed to focus of IDEA 2004. We discuss the research and policy basis for
Jack M. Fletcher, Department of Psychology, University of
Houston, TMC Annex, 2151 West Holcombe Boulevard, Suite these models, focusing only on academic difficulties (primarily
222, Houston, TX 77204-5053; e-mail: jackfletcher@uh.edu. reading) and noting that there is comparable development of RTI
# 2009, Copyright the Author(s) models in the behavioral area (Walker, Stiller, Serverson, Feil, &
Journal Compilation # 2009, Society for Research in Child Development Golly, 1998).

Volume 3, Number 1, Pages 30–37

 Response to Intervention j 31

MULTITIERED INTERVENTION MODELS AND RTI In a common implementation of a standard protocol model
(Figure 1; Vaughn, Wanzek, Woodruff, & Linan-Thompson,
What Are RTI Models? 2006), all students are screened, and those at risk for academic
RTI models are multitiered service delivery systems in which problems are assessed frequently (every 1–4 weeks) on short
schools provide layered interventions that begin in general probes designed to assess progress over time (Stecker, Fuchs, &
education and increase in intensity (e.g., increased time for Fuchs, 2005). Classroom teachers receive professional devel-
instruction to smaller groups of students) depending on the opment in effective instruction and ways to enhance differen-
students’ instructional response. There are many approaches to tiation and intensity through flexible grouping strategies and
the implementation of RTI models, which are best considered as evaluations of progress (Tier 1, primary intervention). Children
a set of processes and not a single model, with variation in how who do not achieve specified levels of progress based on local or
the processes are implemented. These approaches have at least national benchmarks receive additional instruction in small
two historical origins, both representing efforts to implement groups of three to five students for 20–40 min daily (Tier 2,
prevention programs in schools. secondary intervention). If the child does not make adequate
The first source involves schoolwide efforts to prevent progress in secondary intervention, an even more intensive and
behavior problems (Donovan & Cross, 2002; Walker et al., individualized intervention (Tier 3, tertiary intervention) is
1998). These models are associated with a problem-solving provided that may involve smaller groups, increased time in
process in which a shared decision-making team identifies intervention (45–60 min daily), and a more specialized teacher.
a behavior or academic problem, proposes strategies that address Progress is monitored weekly or biweekly. These models link
the problem, evaluates the outcome, and then reconvenes to with special education because inadequate instructional
consider whether the problem has been resolved, leading to response allows for determination of adequate and inadequate
improvements in behavior or learning (Reschly & Tilly, 1999). responders and provides a framework for implementing seam-
The second origin derives from research on preventing less interventions between general and special education.
reading difficulties in children. These approaches typically The implementation of both problem-solving and standard-
use standardized protocols to deliver interventions increasing ized protocol models is a significant effort. First, providing
in intensity and differentiation depending on the child’s effective Tier 1 instruction to all students requires ongoing
instructional response. Both models have been significantly professional development, screening, and progress monitoring
influenced by public health models of disease prevention that of students. Maintaining these practices demands an extensive
differentiate primary, secondary, and tertiary levels of inter- professional development regimen from well-trained and com-
vention that increase in cost and intensity depending on the mitted professionals that are not readily available (National
patient’s response to treatment (Vaughn, Wanzek, & Fletcher, Association of State Directors of Special Education [NASDSE],
2007). 2006). Second, Tier 2 intervention is continuous. Although

Figure 1. A three-tier model for increasingly intense academic and behavioral interventions.
The percentages represent estimates of the number of children who are at grade level (Tier 1) and
who require Tier 2 and Tier 3 services.
Source. From National Association of State Directors of Special Education (2006). Copyright 2006
by the National Association of State Directors of Special Education, Inc. Reprinted with permission.

Child Development Perspectives, Volume 3, Number 1, Pages 30–37

32 j Jack M. Fletcher and Sharon Vaughn

effective Tier 1 reduces the number of students at risk, controversial, and there are no widely accepted criteria for
significant numbers of students (as many as 20%–25% in early identification of inadequate responders. Thus, instructional
reading; Vaughn et al., 2006) require supplemental interven- response is not recommended as the sole determinant of
tions by trained personnel (e.g., classroom teachers, paraprofes- eligibility for special education.
sionals). Finally, many school districts do not perceive that they
have the personnel and resources to effectively implement Evidence-Based Interventions
all the elements of RTI models. Nonetheless, over the past 20 RTI models depend on the implementation of evidence-based
years, many school districts have implemented RTI models interventions designed to prevent or remediate academic difficul-
from kindergarten to high school (Jimerson, Burns, & ties. Numerous syntheses and meta-analyses address the efficacy
VanDerHeyden, 2007). of interventions for students with academic difficulties. Although
a complete analysis is beyond this article, the most comprehen-
Screening and Progress Monitoring sive meta-analysis on interventions for children identified with
A key component of RTI models is universal screening of all LDs was conducted by Swanson, Hoskyn, and Lee (1999), who
children for academic problems. The screening instrument can reviewed and analyzed 180 intervention studies over a 30-year
be norm referenced or criterion referenced, the latter often period. Their findings suggested moderate to high effects across
representing the first assessment of a progress monitoring tool. studies (0.79) and higher effect sizes for interventions conducted
Because screening devices are used with entire grades, the key in resource room settings (0.86) than those in general education
is that the tool can be quickly administered with adequate classes (0.48).
sensitivity and specificity. In general, screenings tend to over- Wanzek and Vaughn (2007) synthesized studies of extensive
identify children as being at risk because the consequence is reading interventions defined as at least 100 sessions (i.e.,
that students’ progress is monitored and/or they are provided approximately 20 weeks of daily intervention). The effects were
a supplemental intervention to enhance their performance in diverse, but most studies reported effect sizes in the moderate to
reading or math (Fletcher, Lyon, Fuchs, & Barnes, 2007). large range. Effect sizes were usually larger if the study (a)
The most common implementation of a progress monitoring involved students in kindergarten and Grade 1 as opposed to
measure involves a technology known as curriculum-based Grades 2–5, (b) used a comprehensive reading program, and (c)
measurement (CBM), which provides brief (1–3 min per child) delivered the intervention one-on-one or in small groups.
assessments that are readily administered and interpreted by In another recent meta-analysis, Scammacca et al. (2007)
classroom teachers and useful for adjusting instruction (Fuchs, examined outcomes from intervention studies conducted with
Deno, & Mirkin, 1984). Typically, a child reads a list of words or older students with reading difficulties. The overall effect size
a short passage appropriate for his or her grade level (or does across all 31 studies in Scammacca et al. (2007) was 0.95, with a
a set of math computations, spells words, etc.). The number of lower overall effect size when only standardized, norm-referenced
words (or math problems or spelling items) correctly read (or measures were used (0.42). For 23 intervention studies that
computed or spelled) is graphed over time and compared against measured reading comprehension, often with experimenter-
grade-level benchmarks. designed measures, the effect size was 1.33; for standardized
There is a substantial research base showing that when used achievement reading measures, the effect size was 0.35. The
by classroom teachers, CBM provides reliable and valid overall findings indicate that for older students with reading
information about how well students are progressing and is difficulties (a) adolescence is not too late to intervene, (b)
associated with improved outcomes (Stecker et al., 2005). students benefit from both word-level and text-level interven-
Controlled studies document that when CBM implementation tions, (c) instruction in reading comprehension strategies is
is compared to classrooms not using CBM, better end-of-year associated with large effects, (d) students are able to learn the
academic outcomes result because teachers modify goals and meanings of words they are taught, and (e) both researcher-
adjust instruction (e.g., Fuchs, Fuchs, & Hamlett, 1989a, implemented and teacher-implemented interventions are
1989b; Fuchs, Fuchs, Hamlett, & Stecker, 1991). Serial assess- effective. However, it may take more intensity and a longer
ments based on CBM have also been used to provide data for period of time to bring older children with reading difficulties
eligibility decisions involving education services (Fuchs & to grade level, which is why prevention efforts are receiving
Fuchs, 1998). greater emphasis (Torgesen et al., 2001).
Despite the value of CBM measures, there are concerns Graham and Perin (2007a, 2007b) provided meta-analyses on
about equivalency of text passages (Francis et al., 2008). In effective writing practices. They identified several instructional
addition, it is unclear how reliably benchmarks from these practices that are associated with improved outcomes for
CBM-type measures can be used to determine movement students, including (a) writing strategies that involve explicitly
across the tiers, and when used, whether the best benchmarks teaching students to plan, revise, brainstorm, and edit (0.82); (b)
are at the district or national level. Finally, the use of CBM summarizing through writing (0.82); (c) collaborating with other
measures as part of the eligibility process is especially students in small groups to provide feedback and write

Child Development Perspectives, Volume 3, Number 1, Pages 30–37

 Response to Intervention j 33

cooperative (0.75); (d) assigning students reasonable goals for insufficient progress to read at grade-appropriate levels even
improving writing (0.70); and (e) other practices such as word after receiving a yearlong intervention followed by additional
processing, sentence combining, and writing process, which all intense interventions (16 weeks) in Grades 1–3 (Denton,
yield small to medium effect sizes. Fletcher, Anthony, & Francis, 2006). However, studies of
There is less intervention research in the academic area of multitiered intervention models yield rates of inadequate
mathematics, although recent research implementing RTI-type responders for early reading as low as 2%–5% (Berninger
frameworks is promising (Fuchs et al., 2005). Baker, Gersten, et al., 2003; Mathes et al., 2005; McMaster, Fuchs, Fuchs, &
and Lee (2002) completed an empirical synthesis that revealed Compton, 2005; Torgesen, 2000). Because the number of
that effective mathematics instruction provides data or recom- students who need intense interventions may be greatly reduced,
mendations to teachers and students (0.57), uses peer pairing to schools may be able to devote the resources needed for effective
support learning (0.62), provides explicit instruction directed remediation of inadequate responders (Burns, Appleton, &
by the teacher including teacher-facilitated approaches (0.58), Stehouwer, 2005; VanDerHeyden, Witt, & Gilbertson, 2007).
and provides practices for communicating student successes to Despite these issues, there are successful district-wide
parents (0.42). implementations of RTI models across the country (Jimerson
et al., 2007; NASDSE, 2006). Many of these implementations
Coordinated Systems of Service Delivery report an increase in overall academic achievement scores and
Despite the research base supporting the assessment and a decrease in special education referrals (e.g., VanDerHeyden
intervention components of RTI, the most daunting aspects et al., 2007). More research focusing on how schools success-
involve schoolwide implementation, where the scaling issues fully implement (and struggle to implement) RTI models will be
are significant. Intervention services in schools are often funded needed. This research must look at outcomes in relation to
by separate entitlement programs, especially Title I and IDEA, historical data so that it will be clear that RTI models improve
that tend to have specific eligibility criteria and historically outcomes for all students, including those who are at risk and not
have made it difficult to blend resources to support schoolwide at risk for academic difficulties. Furthermore, scaling these
intervention models. These programs are often isolated from models nationally will be a significant task.
general education and the classroom, so that instruction can be
fragmented. Because it may take several years to change RTI MODELS AND SPECIAL EDUCATION
practice in implementing RTI models, especially given the
entrenchment of older ways of thinking about instruction, The specific provisions in IDEA 2004 for RTI models have been
schools should move slowly and with care (NASDSE, 2006). controversial in the area of special education. This controversy
Resources will be a concern in many districts unless careful focuses primarily on two issues, the first representing the scaling
appraisals are made of the available resources, which are issues reviewed above. The second issue is the use of RTI
typically redeployed to support RTI models of service delivery. models in the identification of LDs. In contrast to 30 years of
A negative consequence of the fact that RTI is recommended implementation, IDEA 2004 allows school districts to imple-
under the new reauthorized IDEA 2004 is that many educators ment RTI models and move away from identification models that
perceive RTI as simply a special education initiative. In fact, have relied on a discrepancy between IQ and achievement.
implementation of RTI models requires close collaboration and Instead, identification relies on inadequate instructional
implementation with general education, special education, Title response and other criteria, minimizing the role of IQ and other
I, and other entitlement programs. assessments attempting to identify discrepancies in cognitive
Scaling issues are also complicated because of incomplete- ability for identification. Given how entrenched the latter assess-
ness in the intervention evidence base. The question of how to ments are in the everyday practice of evaluating students, this
implement RTI models in secondary schools is daunting, controversy is not surprising. However, the changes in IDEA
especially given weaknesses in research studies on interven- 2004 reflect concerns about (a) the effectiveness of traditional
tions and progress-monitoring tools for older students. It seems implementations of special education in schools and (b) the use of
difficult to conceive of RTI models when the prevention IQ–achievement discrepancies for identification.
component is not strongly implemented. Although there are
data on intensive Tier 3 interventions, they have been infre- Intervention
quently applied as part of a multitiered intervention, reflecting There is a major disconnection between what is known about
in part the sheer cost of research studies based on a multitiered efficacy of instruction for students with academic difficulties
intervention model. Because few studies have been done of and how students are taught in schools, especially for students
children defined as inadequate responders in an RTI model, the most at risk for academic and behavioral difficulties. Studies of
efficacy of a layered Tier 3 intervention in this context is not well outcomes for students placed in special education show flat
established. Preliminary evidence suggests that many of these levels of growth and little evidence that typical interventions
students are difficult to teach, with approximately half showing close the achievement gap (Bentum & Aaron, 2003; Donovan &

Child Development Perspectives, Volume 3, Number 1, Pages 30–37

34 j Jack M. Fletcher and Sharon Vaughn

Cross, 2002; Glass, 1983; Hanushek, Kain, & Rivkin, 1998; REFERRAL SCREENING
(Old Model) NEW
Torgesen et al., 2001; Vaughn, Levy, Coleman, & Bos, 2002). MODEL
The interest in RTI is fueled in part by the focus on instructional
ELIGIBILITY TESTING Monitor TREATMENT 1-2
outcomes and the attempt to reduce the number of students who
need the most intense intervention. As we reviewed above, the Not Eligible Eligible Responders Non-Responders
most consistent evidence about improving outcomes for stu-
ELIGIBILITY TESTING
dents identified with LDs addresses preventing or remediating TREATMENT Monitor

specific academic skills, where the focus on academic domains Not Eligible Eligible
Responders Non-Responders
is especially important (Fletcher et al., 2007). These children TREATMENT 3
are subjected to a potpourri of interventions involving the eyes,
Responders Non-Responders
brain, and perceptual processes that do not involve reading,
writing, and math. The former interventions show little gener- Monitor
alization to academic successes for these students (Mann, 1979;
Vellutino, Fletcher, Scanlon, & Snowling, 2004). Figure 2. A comparison of a traditional eligibility process and a process
incorporating response to instruction (RTI) in a three-tier model.
On the left side, the student is referred for an eligibility evaluation. The
Identification student is either eligible or not eligible; if eligible, the student receives
intervention that is evaluated for 1–3 years. In an RTI model, all students
Although deficits in specific cognitive functions are strongly are screened and those at-risk receive progress monitoring and interven-
associated with different types of LD, a focus on interindividual tion in the general education classroom. If the response to different
differences and discrepancies has not proven to be a reliable interventions is not sufficient to meet progress-monitoring benchmarks,
increasingly intense interventions are provided. If inadequate instruc-
practice for identification (Francis et al., 2005; Shepard, 1980; tional response continues, the student may be referred for an eligibility
Siegel, 1992; Stuebing et al., 2002) and does not lead to evaluation, which would be different because the student was identified as
implementation of appropriate interventions resulting in strong at-risk earlier and the availability of data on instructional response to
influence the type of evaluation that is conducted.
outcomes (Mann, 1979; Vaughn & Linan-Thompson, 2003). Source. From Fletcher et al. (2007). Copyright 2007 by Guilford Press.
These concerns are especially significant for the use of IQ– Reprinted with permission.
achievement discrepancy models of identification.
Two meta-analyses highlight the concerns about the validity
The use of instructional response data, however, is unlikely to
of IQ-discrepancy models (Hoskyn & Swanson, 2000; Stuebing
address all problems related to identifying students with LDs.
et al., 2002). Across studies, both found negligible to small
Within IQ-discrepancy models, one of the persistent problems
overall effect size differences between IQ-discrepant and non-
has been the use of rigid ‘‘cut points’’ for identification of LDs.
discrepant poor readers, with negligible differences on most
The use of rigid cut points for benchmarks and establishing
measures of reading and phonological processing. Other studies
students as high or low responders to instruction could yield the
comparing poor readers with and without significant IQ–
same types of problems with reliability and validity of identi-
achievement discrepancies found no difference in prognosis
fication in RTI models. A problem with cognitive discrepancy
(Francis, Shaywitz, Stuebing, Shaywitz, & Fletcher, 1996;
models is that the attributes (IQ, cognitive processes, and
Share, McGee, & Silva, 1989) or response to instruction
achievement) are usually continuous and normally distributed
(Vellutino, Scanlon, & Lyon, 2000). These validity issues do
when children with brain injury are excluded from the sample
not support the 30-year-old practice instantiated in schools and
(Lewis, Hitch, & Walker, 1994; Shaywitz, Escobar, Shaywitz,
clinics of identifying LDs on the basis of a discrepancy between
Fletcher, & Makuch, 1992; but see Rutter & Yule, 1975, which
IQ and achievement (Donovan & Cross, 2002).
did not exclude brain injury). Deciding reliably where on this
continuum a disability resides is inherently arbitrary and must
RTI Models and Identification rely on criteria other than IQ and achievement scores (Francis
RTI models specifically deemphasize cognitive discrepancies et al., 2005). However, instructional response may also exist on
in the identification process, focusing instead on discrepancies a continuum that has no inherent qualitative breaks. Criteria for
relative to age-based expectations and instruction. Thus, the inadequate response may be as arbitrary as a cut point on an
eligibility process in an RTI model is different from the achievement dimension and simply creating formulae without
traditional model (see Figure 2). In an RTI model, children testing their validity will be no better than IQ-discrepancy
have been screened and monitored early in schooling as models. The use of confidence intervals and an evaluation of the
opposed to traditional eligibility models that depend on referral, consequences of different decisions to intervene or not inter-
usually in the later grades and after failure. In addition, the data vene will help with this issue. Validating decisions against other
on instructional response lead to evaluations that ask how to adaptive criteria not directly tied to academic achievement
best teach the child and that deemphasize the search for would also help determine the adequacy of decisions. Research
cognitive discrepancies. is also needed to evaluate the reliability and validity of

Child Development Perspectives, Volume 3, Number 1, Pages 30–37

 Response to Intervention j 35

decisions made by experts and not decisions based solely on the definition. This subgroup of inadequate responders who
statistical criteria. show low achievement and who do not have other disabilities or
environmental factors that explain low achievement may
LD Identification Requires Multiple Criteria epitomize what is meant by unexpected underachievement.
Children cannot be identified with LDs solely on the basis of It is likely that research on cognitive and neurobiological
instructional response. The consensus group of researchers correlates of LD will begin to focus on children identified
convened for the Learning Disabilities Summit (Bradley, under this model and move away from what historically have
Danielson, & Hallahan, 2002) suggested that three criteria been samples that are a mixture of students with adequate and
were important: (a) response to instruction, assessed through inadequate instructional histories. The result may be new
progress monitoring and evaluations of the integrity of inter- approaches to instruction and new understandings of the
ventions; (b) assessment of low achievement, typically through neurobiological and environmental factors that underlie aca-
norm-referenced achievement tests; and (c) application of demic difficulties and LD.
exclusionary criteria to ensure that low achievement is not
due to another disability (e.g., mental retardation, sensory
disorder) or to environmental and contextual factors (e.g., REFERENCES
limited English proficiency).
IDEA 2004 is consistent with this hybrid model of classifi- Baker, S., Gersten, R., & Lee, D. (2002). A synthesis of empirical
cation. It explicitly indicates that children may be identified for research on teaching mathematics to low-achieving students.
special education only with documentation that low achieve- Elementary School Journal, 103, 336–358.
Bentum, K. E., & Aaron, P. G. (2003). Does reading instruction in
ment is not the result of inadequate instruction. In addition,
learning disability resource rooms really work? A longitudinal
IDEA identifies six domains of low academic achievement in study. Reading Psychology, 24, 361–369.
which LDs may occur. It requires assessment of the traditional Berninger, V. W., Vermeulen, K., Abbott, R. D., McCutchen, D.,
exclusionary criteria, a process that remains inherently vague Cotton, S., Cude, J., et al. (2003). Comparison of three
when the issues are factors such as emotional difficulties, which approaches to supplementary reading instruction for low-achieving
may coexist or result from poor achievement, or economic second-grade readers. Language, Speech, and Hearing Services in
disadvantage, the role of which in relation to low achievement is Schools, 34, 101–116.
Bradley, R., Danielson, L. E., & Hallahan, D. P. (2002). Identification of
difficult to separate from LDs in the absence of adequate
learning disabilities: Research to practice. Mahwah, NJ: Erlbaum.
instruction. Burns, M., Appleton, J. J., & Stehouwer, J. D. (2005). Meta-analytic
review of responsiveness-to-intervention research: Examining
CONCLUSIONS field-based and research-implemented models. Journal of
Psychoeducational Assessment, 23, 381–394.
The primary goal of RTI models is the prevention and remediation Denton, C. A., Fletcher, J. M., Anthony, J. L., & Francis, D. J. (2006).
of academic and behavioral difficulties through effective class- An evaluation of intensive intervention for students with
persistent reading difficulties. Journal of Learning Disabilities,
room instruction and increasingly intense interventions. A
39, 447–466.
secondary goal of RTI models is the provision of useful data that Donovan, M. S., & Cross, C. T. (2002). Minority students in special
contributes to referral and decision making about students with and gifted education. National Research Council Committee on
LDs. If the scaling issues can be addressed, districts that Minority Representation in Special Education. Washington,
successfully implement RTI models may improve achievement DC: National Academies Press.
and behavioral outcomes in all students, especially those most at Fletcher, J. M., Lyon, G. R., Fuchs, L. S., & Barnes, M. A. (2007).
risk for academic difficulties. Learning disabilities: From identification to intervention. New
Regardless of the identification model employed by schools, York: Guilford.
Francis, D. J., Fletcher, J. M., Stuebing, K. K., Lyon, G. R., Shaywitz,
IDEA 2004 requires an assessment of instructional response.
B. A., & Shaywitz, S. E. (2005). Psychometric approaches to the
Definitions of LD have always relied on the elimination of identification of LD: IQ and achievement scores are not
known causes of low achievement, which include inadequate sufficient. Journal of Learning Disabilities, 38, 98–108.
instruction. Those children who do not show evidence of an Francis, D. J., Santi, K. L., Barr, C., Fletcher, J. M., Varisco, A., &
exclusionary condition and who have a cognitive discrepancy Foorman, B. R. (2008). Form effects on the estimation of
have been considered ‘‘unexpected’’ underachievers, which is students’ oral reading fluency using DIBELS. Journal of School
the core construct historically underlying LD (Hammill, Psychology, 46, 315–342.
Francis, D. J., Shaywitz, S. E., Stuebing, K. K., Shaywitz, B. A., &
1993). If inadequate response to quality instruction can be
Fletcher, J. M. (1996). Developmental lag versus deficit models
formally assessed, it represents an inclusionary criterion of reading disability: A longitudinal, individual growth curves
indicating the presence of intractability in instruction. analysis. Journal of Educational Psychology, 88, 3–17.
A different type of child will emerge with LDs if formal Fuchs, L. S., Compton, D. L., Fuchs, D., Paulsen, K., Bryant, J. D., &
assessment of inadequate instructional response is part of Hamlett, C. L. (2005). The prevention, identification, and

Child Development Perspectives, Volume 3, Number 1, Pages 30–37

36 j Jack M. Fletcher and Sharon Vaughn

cognitive determinants of math difficulty. Journal of Educational Reschly, D. J., & Tilly, W. D. (1999). Reform trends and system design
Psychology, 97, 493–513. alternatives. In D. Reschly, W. Tilly, & J. Grimes (Eds.), Special
Fuchs, L. S., Deno, S. L., & Mirkin, P. K. (1984). The effects of education in transition (pp. 19–48). Longmont, CO: Sopris West.
frequent curriculum-based measurement and evaluation on stu- Rutter, M., & Yule, W. (1975). The concept of specific reading
dent achievement, pedagogy, and student awareness of learning. retardation. Journal of Child Psychology & Psychiatry & Allied
American Educational Research Journal, 21, 449–460. Disciplines, 16, 181–197.
Fuchs, L. S., & Fuchs, D. (1998). Treatment validity: A unifying concept Scammacca, N., Roberts, G., Vaughn, S., Edmonds, M., Wexler, J.,
for reconceptualizing the identification of learning disabilities. Reutebuch, C. K., et al. (2007). Reading interventions for
Learning Disabilities Research and Practice, 13, 204–219. adolescent struggling readers: A meta-analysis with implications
Fuchs, L. S., Fuchs, D., & Hamlett, C. L. (1989a). Effects of for practice. Portsmouth, NH: RMC Research Corporation,
alternative goal structures within curriculum-based measure- Center on Instruction.
ment. Exceptional Children, 55, 429–438. Share, D. L., McGee, R., & Silva, P. A. (1989). IQ and reading progress:
Fuchs, L. S., Fuchs, D., & Hamlett, C. L. (1989b). Effects of A test of the capacity notion of IQ. Journal of the American
instrumental use of curriculum-based measurement to enhance Academy of Child & Adolescent Psychiatry, 28, 97–100.
instructional programs. Remedial and Special Education, 102, Shaywitz, S. E., Escobar, M. D., Shaywitz, B. A., Fletcher, J. M., &
43–52. Makuch, R. (1992). Evidence that dyslexia may represent the
Fuchs, L. S., Fuchs, D., Hamlett, C. L., & Stecker, P. M. (1991). lower tail of a normal distribution of reading ability. New
Effects of curriculum-based measurement and consultation on England Journal of Medicine, 326, 145–150.
teacher planning and student achievement in mathematics Shepard, L. (1980). An evaluation of the regression discrepancy
operations. American Educational Research Journal, 28, method for identifying children with learning disabilities.
617–641. Journal of Special Education, 14, 79–91.
Glass, G. (1983). Effectiveness of special education. Policy Studies Siegel, L. S. (1992). An evaluation of the discrepancy definition of
Review, 2, 65–78. dyslexia. Journal of Learning Disabilities, 25, 618–629.
Graham, S., & Perin, D. (2007a). A meta-analysis of writing Stecker, P. M., Fuchs, L. S., & Fuchs, D. (2005). Using curriculum-
instruction for adolescent students. Journal of Educational based measurement to improve student achievement: Review
Psychology, 99, 445–476. of research. Psychology in the Schools, 42, 795–819.
Graham, S., & Perin, D. (2007b). Writing next: Effective strategies to Stuebing, K. K., Fletcher, J. M., LeDoux, J. M., Lyon, G. R., Shaywitz,
improve writing of adolescents in middle and high schools— S. E., & Shaywitz, B. A. (2002). Validity of IQ-discrepancy
A report to Carnegie Corporation of New York. Washington, classifications of reading disabilities: A meta-analysis. American
DC: Alliance for Excellent Education. Educational Research Journal, 39, 469–518.
Hammill, D. D. (1993). A brief look at the learning disabilities Swanson, H. L., Hoskyn, M., & Lee, C. (1999). Interventions for
movement in the United States. Journal of Learning Disabilities, students with learning disabilities: A meta-analysis of treatment
26, 295–310. outcome. New York: Guilford.
Hanushek, E. A., Kain, J. F., & Rivkin, S. G. (1998). Does special Torgesen, J. K. (2000). Individual differences in response to early
education raise academic achievement for students with disabil- interventions in reading: The lingering problem of treatment
ities? (Working Paper No. 6690). Cambridge, MA: National resisters. Learning Disabilities Research and Practice, 15, 55–64.
Bureau of Economic Research. Torgesen, J. K., Alexander, A. W., Wagner, R. K., Rashotte, C. A.,
Hoskyn, M., & Swanson, H. L. (2000). Cognitive processing of low Voeller, K. K. S., Conway, T., et al. (2001). Intensive remedial
achievers and children with reading disabilities: A selective instruction for children with severe reading disabilities: Imme-
meta-analytic review of the published literature. School diate and long-term outcomes from two instructional approaches.
Psychology Review, 29, 102–119. Journal of Learning Disabilities, 34, 33–58.
Jimerson, S. R., Burns, M. K, & VanDerHeyden, A. M. (2007). U.S. Department of Education. (2004). Individuals with Disabilities
Handbook of response to intervention: The science and practice of Improvement Act of 2004, Pub. L. 108-466. Federal Register,
assessment and intervention. Springfield, IL: Charles E. Springer. 70, 35802–35803.
Lewis, C., Hitch, G. J., & Walker, P. (1994). The prevalence of U.S. Department of Education. (2007). 25th annual report to Congress
specific arithmetic difficulties and specific reading difficulties in on the implementation of the Individuals with Disabilities
9- to 10-year-old boys and girls. Journal of Child Psychology Education Act. Washington, DC: Author.
and Psychiatry, 35, 283–292. VanDerHeyden, A. M., Witt, J. C., & Gilbertson, D. (2007). A multi-
Mann, L. (1979). On the trail of process. New York: Grune & Stratton. year evaluation of the effects of a response to intervention (RTI)
Mathes, P. G., Denton, C. A., Fletcher, J. M., Anthony, J. L., Francis, model on identification of children for special education.
D. J., & Schatschneider, C. (2005). The effects of theoretically Journal of School Psychology, 45, 225–256.
different instruction and student characteristics on the skills of Vaughn, S., & Fuchs, L. S. (2003). Redefining learning disabilities as
struggling readers. Reading Research Quarterly, 40, 148–182. inadequate response to instruction: The promise and potential
McMaster, K. L., Fuchs, D., Fuchs, L. S., & Compton, D. L. (2005). problems. Learning Disabilities Research and Practice, 18, 137–146.
Responding to nonresponders: An experimental field trial of Vaughn, S., Levy, S., Coleman, M., & Bos, C. S. (2002). Reading
identification and intervention methods. Exceptional Children, instruction for students with LD and EBD: A synthesis of
71, 445–463. observation studies. Journal of Special Education, 36, 2–13.
National Association of State Directors of Special Education. (2006). Vaughn, S., & Linan-Thompson, S. (2003). What is special about
Response to intervention: Policy considerations and implementa- special education for students with learning disabilities?
tion. Alexandria, VA: Author. Journal of Special Education, 37, 140–147.

Child Development Perspectives, Volume 3, Number 1, Pages 30–37

 Response to Intervention j 37

Vaughn, S., Wanzek, J., & Fletcher, J. M. (2007). Multiple tiers learned in the past four decades? Journal of Child Psychiatry
of intervention: A framework for prevention and identification and Psychology, 45, 2–40.
of students with reading/learning disabilities. In B. M. Taylor Vellutino, F. R., Scanlon, D. M., & Lyon, G. R. (2000). Differenti-
& J. Ysseldyke (Eds.), Educational interventions for ating between difficult-to-remediate and readily remediated poor
struggling readers (pp. 173–196). New York: Teacher’s readers: More evidence against the IQ-achievement discrepancy
College Press. definition for reading disability. Journal of Learning Disabilities,
Vaughn, S. R., Wanzek, J., Woodruff, A. L., & Linan-Thompson, S. 33, 223–238.
(2006). A three-tier model for preventing reading difficulties and Walker, H. M., Stiller, B., Serverson, H. H., Feil, E. G., & Golly, A.
early identification of students with reading disabilities. In D. H. (1998). First step to success: Intervening at the point of school
Haager, S. Vaughn, & J. K. Klingner (Eds.), Validated reading entry to prevent antisocial behavior patterns. Psychology in the
practices for three tiers of intervention (pp. 11–28). Baltimore: Schools, 35, 259–269.
Brookes. Wanzek, J., & Vaughn, S. (2007). Research-based implications from
Vellutino, F. R., Fletcher, J. M., Scanlon, D. M., & Snowling, M. J. extensive early reading interventions. School Psychology Review,
(2004). Specific reading disability (dyslexia): What have we 36, 541–561.

Child Development Perspectives, Volume 3, Number 1, Pages 30–37

View publication stats