Decoding and Spelling Accommodations for Postsecondary Students Demonstrating Dyslexia -- It's More Than Processing Speed
Noel Gregg, Cheri Hoy, Donna Ann Flaherty, Peggy Norris, Christopher Coleman, Mark Davis, and Michael Jordan
The vast majority of students with learning disabilities at the postsecondary
level demonstrate reading decoding, reading fluency, and writing
deficits. Identification of valid and reliable psychometric measures for
documenting decoding and spelling disabilities at the postsecondary level
is critical for determining appropriate accommodations. The purpose of
this study was threefold: (a) to examine the relationship between specific
Woodcock-Johnson III Cognitive and Achievement clusters (WJ III;
Woodcock, McGrew, and Mather, 2001) across populations with and without
dyslexia at the postsecondary level; (b) to identify the strongest WJ III
cognitive predictors for decoding, spelling, and reading fluency across college
students with and without dyslexia; and (c) to discuss the implications
of the findings for assessment and accommodation practices for secondary
and postsecondary students. A total of 101 college students with
documented dyslexia and 100 college students without disabilities participated
in the study. Both word knowledge and processing speed were found
to significantly influence performance in very different ways.
Diagnostic Validity of Wechsler Substest Scatter
Marley W. Watkins
Cognitive subtest scatter has often been considered to be diagnostically significant.
The current study tested the diagnostic validity of four separate
operationalizations of WISC-III subtest scatter: (a) range of verbal, performance,
and full-scale subtests; (b) variance of verbal, performance, and
full-scale subtests; (c) number of subtests deviating by ±3 points from verbal,
performance, and full-scale average; and (d) Mahalanobis distance of
full-scale subtests. The WISC-III normative sample was compared to
1,592 students with learning disabilities (LD). Scatter indices were strongly
correlated with each other (i.e., r = .81 to .95). Receiver operating characteristic
(ROC) curve analyses revealed that using any of the subtest scatter
indices to diagnose LD resulted in correct decisions only 50% to 55%
of the time. Chance would afford similar levels of accuracy.
Applying Cognitive-Behavioral Interventions in Greek Mainstream School Settings: The Case of Learning Difficulties
Maria Zafiropoulou and Chryssoula Karmba-Schina
This study presents a psycho-educational intervention using cognitivebehavioral
techniques to modify and improve thinking strategies as well as
facilitate behavioral adjustment and generalization of strategy use in children
with learning disabilities attending mainstream schools. One hundred
and twenty Greek pupils in 2nd, 3rd, and 4th grade were treated
under four different conditions: psycho-educational intervention, selfinstruction
training, attention control, and no treatment. Results indicated
that the psycho-educational intervention program was clearly superior
to the other conditions, producing meaningful and lasting effects on participants'
cognitive skills and strategies and a significantly heightened
sense of academic self-esteem. Improved task performance was also evident
in a 7-month follow-up assessment. Children's academic achievement
(writing, reading and mathematics performance as rated by their
teachers) also improved satisfactorily compared to that of controls.
Using CRA to Teach Algebra to Students with Math Difficulties in Inclusive Settings
Bradley S. Witzel
The importance of algebra instruction has increased in the United States in
the past few years. Thus, in most states, middle school students are required
to take Algebra 1.Middle school students with math difficulties in inclusion
algebra settings may require a different instructional approach. The purpose
of this research was to compare student achievement in solving linear
algebraic functions across two procedural approaches: a multisensory algebra
model using a concrete-to-representational-to-abstract sequence of
instruction (CRA) and a repeated abstract explicit instruction model. Out
of 231 students who participated, the students who learned through the
CRA model scored significantly higher on the post- and follow-up test. The
success of the CRA model was consistent for students with a history of low,
medium, and high math achievement. Implications of this model and possibilities
for future research are discussed.
Adult Learning Disabilities Screening Using an Internet-Administered Instrument
Sean Lancaster and Daryl Mellard
Identifying individuals with specific learning disabilities (SLD) is a complex
task, particularly for adult populations. Adult agencies such as vocational
rehabilitative services or adult basic education often use different
SLD definitions and criteria, are often understaffed, have limited
resources, and have a shortage of staff trained on SLD screening (Skinner,
Gillespie, and Balkam, 1997). The Adult Learning Disabilities Screening
(ALDS) battery (Mellard, 1999) is one of the few valid screening measures
for adults currently in use. This study validated an Internet-administered
version of ALDS, e-ALDS, by comparing 122 adult education participants
who completed the ALDS in paper-and-pencil vs. Internet formats.
