Psychology in the Schools, Vol.
35(2), 1998
© 1998 John Wiley & Sons, Inc. CCC 0033-3085/98/020195-03
TEST REVIEWS
Sharon Bradley-Johnson
Central Michigan University
Roid, Gale D. & Miller, Lucy J. (1995, 1997). Leiter International Performance Scale–Revised.
Wood Dale, IL: Stoelting Co.
The original Leiter International Performance Scale, based on studies of children in Hawaii,
was developed by Russell Leiter in 1929. Its purpose was to assess the abilities of children who had
difficulty responding verbally. The current edition is the first revision of the Leiter since 1948.
Description of the Test
The Leiter–Revised (1995, 1997) assesses cognition in individuals ranging in age from 2 years
0 months through 20 years 11 months. It was specifically developed for examinees with communi-
cation disorders, hearing and motor impairments, traumatic brain injuries, attentional problems, and
learning disabilities. Individuals who are not proficient in English are also good candidates for use
of this measure.
Authors of the current edition reported that their work was guided by models of cognitive abil-
ity developed by Gustafsson (1984) and Carroll (1993). They have conceptualized nonverbal cog-
nitive abilities in terms of various mental functions, such as reasoning, visualization, memory, and
attention. The authors emphasized that nonverbal cognitive abilities are not solely dependent on pro-
ficiency in the use of printed materials, words, and numbers. Therefore, administration of the Leit-
er–R is accomplished through the use of pictures, figures, and symbols. Instructions are delivered in
a nonverbal format, using gestures and pantomime.
The Leiter is divided into two groups of ten subtests each. Subtests from the Visualization and
Reasoning Battery (VR) (i.e., Figure Ground, Design Analogies, Form Completion, Matching, Se-
quential Order, Repeated Patterns, Picture Context, Classification, Paper Folding, Figure Rotation)
are used to obtain an IQ score. Subtests from the Attention and Memory Battery (AM) (i.e., Associ-
ated Pairs, Immediate Recognition, Forward Memory, Attention Sustained, Reverse Memory, Visu-
al Coding, Spatial Memory, Delayed Pairs, Delayed Recognition, Attention Divided) are used for
more precise examination of processing in those two areas. Approximate administration time for each
of the batteries is 40–45 minutes. The mental age and ratio IQ scores previously used have been re-
placed by scaled scores, grade/age equivalents, and normal curve equivalents. Percentile scores were
retained. Composite scaled scores have a mean of 100 and a standard deviation of 15; subtests have
a M of 10 and a SD of 3. An attenuated form of each battery can be used as a “screener.”
Examinee age determines the subtests administered. On the VR Battery, 7 subtests are adminis-
tered to each of three age groups (2–5, 6–10, 11–20). For example, children between the ages of 6 and
10 years are administered the Figure Ground, Design Analogies, Form Completion, Matching, Se-
quential Order, Repeated Patterns, and Paper Folding subtests from the VR Battery and all 10 subtests
from the AM Battery. Content of subtests reflects the underlying theoretical orientation of the test au-
thors. For example, the Figure Ground and Form Completion subtests from the VR Battery measure
visual/spatial ability whereas the Attention Sustained, Attention Divided, and Visual Coding subtests
from the AM Battery tap attentional and memory span skills. Subtests begin with at least one teach-
ing item and are untimed. A few subtests offer bonus points for rapid, correct execution of the task.
Assessment of individuals with profound disabilities can be accomplished through the calcula-
tion of a “Growth score.” This is a singular aspect of the Leiter, representing a method of assessing
195
196 Test Reviews
an individual’s progress using successive test administrations. “Growth scores” range from about
380 to 560. The inclusion of the “Growth score” procedure fulfills one of the goals of the current re-
vision, that is, to provide a means of measuring increments of improvement in cognitive ability.
In addition to assessment of cognitive abilities, the Leiter–R offers four social-emotional rating
scales: Examiner, Parent, Teacher, and Self Rating Forms (for children ages 9 through 20). These forms
use a Likert-type scale to obtain information on characteristics that influence functioning at home and
in school. Items related to attention, sociability, activity level, and organizational skills are included.
Normative data for the rating scales was collected separately from the sample used for the VR and AM
Batteries. Small sample sizes were reported. In fact, the authors noted that, because of small sample
size (i.e., 173 total subjects), the Teacher Rating Scale must be viewed as a research measure.
Technical Adequacy
The authors described several goals for this revision. Foremost among these were to obtain a
nationally representative normative sample and to improve the psychometric properties of the test.
Standardization
The Try-Out Edition of the Leiter–R (1994) was comprised of 704 items in 23 subtests. In 1995,
the Standardization Edition revealed that 20 subtests and 691 items had been retained. Results of the
1993 U.S. Census were utilized to develop a stratified sampling plan that would be representative of
geographic and socioeconomic diversity. Consequently, the standardization sample for the VR Bat-
tery, consisting of 1,719 children with approximately equal numbers of boys and girls in each age
group, included similar proportions of examinees based on race, ethnicity, parental education, and
geographic dispersement.
Some problems with standardization were noted. The value of normative data is largely a func-
tion of sample size (Treloar, 1939). Approximately 100 subjects per age group is considered to con-
stitute an adequate sample for a test of this kind. For the VR Battery, this was accomplished for ages
2–5 and 6–11. However, fewer subjects in wider age groupings were utilized for the sample between
the ages of 12 and 17 and fewer than 100 total subjects were reported for ages 18–20. A total of only
763 subjects comprised the entire sample for the AM Battery. Finally, although atypical groups were
included in the norming process, distinct norm groups were not established for special populations.
The inclusion of separate norm groups for atypical subjects, for whom this test was primarily in-
tended, would have been a welcome addition.
Reliability
Results of an internal consistency reliability study, using Cronbach’s a coefficient, showed
mean values ranging from .75 to .90 on subtests of the VR battery and mean values ranging from .67
to .87 on subtests of the AM battery. Reliability estimates for the IQ and composite scores by age
group ranged from .91 to .93 for the Full Scale IQ and from .88 to .90 for the Brief IQ Screener. The
authors noted that lower reliabilities for the screener indicate the advisability of administering the
complete test. Finally, the results of test–retest studies revealed reliability coefficients ranging from
.90 to .96 for the Full Scale IQ across all age groups. However, the authors did not report the length
of elapsed time between test administrations. Further, information is provided for age groupings (i.e.,
2–5, 6–10, 11–20) rather than by age level. In addition, out of the six composite scores obtained on
the AM Battery, only the Attention composite reached .85.
Validity
Content validity of the Leiter–R was established through the use of item–response–theory (IRT)
analysis, reviews of pertinent literature, factor analysis, expert review, and empirical studies of in-
Test Reviews 197
ternal consistency. Criterion-related validity studies revealed an accuracy rate of 96% for the over-
all identification of cognitive delay. Weaker evidence for the identification of giftedness, ADHD and
learning disabilities was reported. The authors stressed the importance of employing multimethod
procedures in assessment of these populations. The results of a concurrent validity study comparing
performance on the Leiter–R and the Wechsler Intelligence Scale for Children–III revealed a corre-
lation coefficient of .86 between the full scale IQ scores on each test. Other studies compared per-
formance on the Leiter–R and subtests of the Stanford–Binet: Fourth Edition, the Wide Range As-
sessment of Memory and Learning and the Test of Memory and Learning with varying results. Finally,
the issue of construct validity was addressed through the description of factor analytic studies. The
authors reported that the following factors were obtained according to age group: Fluid Reasoning,
Visualization, Attention, and Recognition Memory for 4- and 5-year-olds; Reasoning, Visual–Spa-
tial, Attention, Recognition Memory, Associative Memory, and Memory Span for ages 6 to 10; Rea-
soning, Visual–Spatial, Attention, Associative Memory, and Memory Span for ages 11 through 20.
Conclusions
As a nonverbal measure of cognitive abilities, the Leiter–R is a useful tool for evaluating stu-
dents who present with unusual learning problems or obstacles, language disorders, or lack of pro-
ficiency in English. It may also prove to be a viable alternative for assessing individuals who have
had multiple exposures to other intelligence tests. Until additional stability reliability data have been
collected, however, results should be interpreted cautiously.
Examiners who are familiar with the previous version of this instrument will discover that the
test materials have been updated. Some of the original tasks, however, have been preserved. Re-
sponse easels and stimulus cards have replaced the bulkier, wooden materials previously used. The
kit is well organized, colorful, and appealing. However, it is heavy; this is a feature that may prove
challenging to psychologists who must transport test materials from one location to another. The re-
placement of mental age and ratio IQ scores with contemporary score interpretations is a welcome
modernization. The inclusion of explicit teaching items and the general avoidance of untimed tasks
are also advantageous in assessing individuals with special needs.
Familiarity with the test procedures, especially practiced execution of the gestures and pan-
tomimes used in instruction, is necessary for the facile administration of this test. Mastery of Leiter–
R administration will be enhanced by a thorough reading of, and reference to, the detailed and well
organized manual.
Susan A. Sullivan, Ph.D.
Kennedy Krieger Institute
References
Adams, W., & Sheslow, D. (1990). Wide Range Assessment of Memory and Learning (WRAML). Wilmington, DE: Wide
Range Inc.
Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. New York: Cambridge University
Press.
Gustafsson, J. E. (1984). A unifying model for the structure of intellectual abilities. Intelligence, 8, 179–203.
Reynolds, C. R., & Bigler, E. D. (1994). Test of Memory and Learning (TOMAL). Austin, TX: ProEd.
Thorndike, R. L., Hagen, E. P., & Sattler, J. M. (1986). Technical manual for the Stanford–Binet Intelligence Scale (4th
ed.). Chicago, IL: Riverside.
Treloar, A. E. (1939). Elements of statistical reasoning. New York: Wiley.
Wechsler, D. (1991). Manual for the Wechsler Intelligence Scale for Children (3rd ed.). San Antonio, TX: The Psycho-
logical Corporation.