Journal of Emotional

and Behavioral Disorders

http://ebx.sagepub.com/

Test-Taking Strategy Instruction for Adolescents with Emotional and Behavioral Disorders
CHARLES A. HUGHES, DONALD D. DESHLER, KATHY L. RUHL and JEAN B. SCHUMAKER
Journal of Emotional and Behavioral Disorders 1993 1: 189
DOI: 10.1177/106342669300100307
The online version of this article can be found at:
http://ebx.sagepub.com/content/1/3/189

Published by:
Hammill Institute on Disabilities

and
http://www.sagepublications.com

Additional services and information for Journal of Emotional and Behavioral Disorders can be found at:
Email Alerts: http://ebx.sagepub.com/cgi/alerts
Subscriptions: http://ebx.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav

>> Version of Record - Jul 1, 1993

What is This?

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

Test-Taking Strategy Instruction for Adolescents

with Emotional and Behavioral Disorders
CHARLES A. HUGHES, DONALD D. DESHLER,
KATHY L RUHL, AND JEAN B. SCHUMAKER

ANY STUDENTS WHO qualify

for services for emotional/

behavioral disorders (EBD) exhibit academic deficits (e.g., Coutinho,
1986; Epstein, Kinder, 6k Bursuck, 1989;
Mastropieri, Jenkins, 6k Scruggs, 1985),
and their academic performance often deteriorates as they grow older and proceed
from elementary to secondary school
(Foley, Cullinan, 6k Epstein, 1990). In
fact, among all students receiving services
for educationally handicapping conditions, adolescents with EBD have (a) the
lowest promotion rate, (b) the highest
rate of failing grades in secondary mainstream courses, and (c) the lowest gradepoint average in both special education
and regular education classes (U.S. Department of Education, 1990), The development of instructional procedures that
would enable this population to be more
successful in school is an area of critical
need.
Unfortunately, little research has focused on the validation of academic interventions for this population. Two reviews
of the literature on academic or academically related interventions for students
with EBD (Hughes, Ruhl, 6k Korinek,
1985; Ruhl 6k Berlinghoff, in press) revealed a total of only eight studies examining the effectiveness of an intervention
on the academic functioning of adolescents with EBD. A critical analysis of
these eight studies revealed that the majority dealt with remediating specific basicskills deficits versus teaching students
general strategies to help them acquire or
demonstrate mastery of information presented in secondary mainstream content
classes (e.g., science, social studies). Given
that almost one half of the students classified as EBD are served full- or part-time
JOURNAL

OF

EMOTIONAL

AND

Many secondary students with EBD arc mainstreamed in regular classes and are required to take tests
in those classes to demonstrate their competence with regard to their mastery of the content. Unfortunately, these students often do not use strategies that would help them meet this critical classroom
demand. The purpose of this study was to develop a comprehensive test-taking strategy that includes
many of the test-wiseness strategies cited in the literature and to validate an instructional methodology
for teaching this strategy to secondary students with EBD. Through the use of a multiple-probe acrossstudents design, the instruction was shown to be effective in helping students to master the use of the
steps of the comprehensive test-taking strategy and to apply them in a generative way to a series of
novel tests. Preliminary evidence collected in targeted mainstream classes indicated that the students
were applying the strategy to their regular class tests and that their scores on these tests improved.
The results of this study are similar to those achieved in a companion study with students with learning
disabilities, indicating that EBD and LD students can successfully receive strategic instruction together since
they learned the strategy at comparable rates and achieved comparable levels of mastery and generalization.

in mainstream classrooms (U.S. Department of Education, 1990) and therefore

are expected to meet the demands of
those settings, the paucity of empirical
research on teaching these students learning strategies they can use in mainstream
classrooms is surprising.
This lack of research in the area of
learning strategy instruction for students
with EBD has not gone unnoticed. Many
professionals (Deshler, Putnam, 6k Bulgren, 1985; Epstein ck Cullinan, 1988;
Epstein et al., 1989; Fessler, Rosenberg,
6k Rosenberg, 1991; Foley et al., 1990;
Laycock 6k Korinek, 1989; Luebke, Epstein, 6k Cullinan, 1989) have made explicit calls for such research. They have
suggested that adolescents with EBD
need learning strategies instruction in
order to facilitate both their acquisition and demonstration of knowledge in
secondary content classrooms. For example, because many students with EBD
seem to have acquired few strategic behaviors with regard to taking tests (e.g.,
Forness 6k Dvorak, 1982; Mastropieri
et al., 1985; Scruggs 6k Lifson, 1985) and
have more negative attitudes about takBEHAVIORAL

DISORDERS,

JULY

111),

ing tests than their nondisabled peers

(Scruggs, Mastropieri, Tolfa, 6k Jenkins,
1985), they need instruction in this area.
A few studies have indicated that
teaching test-taking strategies to students
with EBD might be a promising intervention area. For example, Scruggs and Mastropieri (1986) and Scruggs, Mastropieri,
and Tolfa-Veit (1986) determined that
training in test-taking skills or strategies
(e.g., checking work, working quickly,
answer all items, familiarity with subtest
formats) can improve elementary EBD
students' performance on standardized
achievement tests.
More recently, Scruggs and Marsing
(1988) examined the impact of training
test-taking strategies that would be more
applicable to test formats used in classrooms. In their study, adolescents with
EBD were taught to use four guessing
strategies based on the existence of cues
contained within test items (e.g., use
of the same word[s] in the stem of a
multiple-choice item and one of the options). The dependent measure was performance on a test in which students had
to guess the correct response based on
VOL.

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

1,

NO.

i,

PACJES

1 H4

M8

|gQ

one of the four cue-usage strategies (fictitious terms were used on the test so there
was no "right" answer, only a best guess
based on the existence of a cue). Students
who received training in the cue-usage
strategies scored about 50% higher on the
posttest than did a matched control
group.
The findings of this study lend support
to the premise that adolescents with EBD
are deficient in certain test-taking strategies but can learn to apply such strategies
to their advantage. However, when discussing limitations of their study, Scruggs
and Marsing noted that they could not
predict the impact of the students' use of
the guessing strategies on test performance in actual classrooms. Relatedly, no
measures of generalization of strategy use
to other settings (i.e., mainstream classes)
were taken. Additionally, student participants received training only in guessing
strategies and did not receive training
in a comprehensive test-taking strategy
which included other types of test-taking
skills (e.g., time usage strategies, checking strategies, etc.) along with guessing
strategies.
Instruction in such a comprehensive
strategy has been shown to be effective
with adolescents with learning disabilities
(LD) (Hughes 6k Schumaker, 1991) and
might also benefit adolescents with EBD.
Unfortunately, there is a lack of information and consensus regarding the effectiveness of the same instructional procedures for students with LD and students
with EBD (Ruhl 6k Berlinghoff", in press).
Indeed, some authors have cautioned
against assuming that the same instructional procedures will work for both
groups (Scruggs 6k Mastropieri, 1986).
Thus, the present study was conducted to address some of the major shortcomings of previous test-taking research
on students with EBD and to shed some
light on whether instructional procedures
that have been found to be successful
with students with LD can be equally
effective with students with EBD. Specifically, the following questions were addressed: (a) Do adolescents with EBD,
when taught a comprehensive test-taking
strategy, acquire usage of the strategy at
a mastery level, (b) Do these students
generalize use of the test-taking strat190

JOURNAL

OF

EMOTIONAL

AND

egy to mainstream classroom tests, and

(c) Does application of the test-taking
strategy result in improved scores on
classroom tests?

groups, and six students participated in

the third group). Thus, the instruction
took place daily in three class periods at
a time when the students were normally
scheduled to be in the resource room.

The Test-Taking Strategy

METHOD

Instructional sessions took place in a

classroom in a middle school and were
conducted by the first author. The classroom was a typical schoolroom containing desks, chalkboard, and an overhead
projector and screen. Students were
taught in small groups of four to six students. In addition to the students with
EBD, six students with learning disabilities also participated in the instruction
(see Hughes 6k Schumaker, 1991, for a
report of the results of students with
learning disabilities) as well as two other
students who had other disabilities. The
students participated in each small group
depending on their school schedules (four
students participated in each of two

The comprehensive test-taking strategy

taught in this investigation was designed
after (a) identifying the most frequently
cited test-taking skills and strategies in the
literature; (b) reviewing the literature on
the test-taking behaviors of secondary
students with mild disabilities; (c) analyzing the formats as well as the nature and
extent of cued items in secondary content tests for such courses as science,
social studies, and language arts; and
(d) completing a task analysis of testtaking behaviors. When a list of cognitive strategies and principles had been
compiled and categorized, a sequence of
cognitive and behavioral steps was specified, and the first-letter mnemonic device,
"PIRATES," was created to increase the
likelihood that students would remember
the steps in a testing situation. The TestTaking Strategy steps that resulted are
shown in Figure 1.
In the first step, Prepare to Succeed,
students write their name and the mnemonic device "PIRATES" on the test as
a prompt to help them remember the
steps of the strategy. Then they scan the
test to determine the number of sections
and types of questions it contains. Next,
they rank the sections from easiest to
most difficult as a means of indicating the
order in which the sections will be completed. Finally, they begin working on the
section of the test that they specified as
easiest within 3 minutes of the beginning
of the test period.
In the second step of the Test-Taking
Strategy, Inspect the Instructions, students read the first set of instructions and
underline words that suggest how and
where to indicate their answer. For the
third stepRead, Remember, Reduce
students read the first test item carefully
as well as all answer options (if the item
is multiple choice), try to remember or
associate the item with what they studied
in preparation for the test, and eliminate
obviously wrong/absurd response options
(thereby reducing the number of choices

BEHAVIORAL

VOL.

Subjects and Setting

Six students, 4 eighth graders and 2
seventh graders, participated in the study.
All six students had been formally classified as EBD according to state of Florida
guidelines, which are the same as Federal
guidelines. All six students had been
referred for special educational services as
a result of "acting out," noncompliant,
and/or physically or verbally aggressive
behaviors. Four were black males, 1 a
white male, and 1 a white female. Ages
ranged from 13 years 3 months to 16
years 5 months (M= 14.3 years). Full-scale
IQ scores (based on the Wechsler Intelligence Scale for Children-Revised [Wechsler, 1974]) ranged from 78 to 118 (M =
96.5), and reading achievement scores
(derived from the Metropolitan Achievement Test [Prescott, Balow, Hogan, 6k
Farr, 1984]) ranged from 4.0 to 6.9 grade
equivalents (M = 5.6). All the students
were enrolled in an EBD resource classroom two to three periods a day and were
mainstreamed for three or four class periods. All were enrolled in at least one required content class (e.g., science or social
studies) and were earning a grade of D
or F in that mainstream content class.
This class became the target generalization setting for this study.

DISORDERS,

JULY

1^!,

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

I ,

NO.

STEP 1

Prepare to succeed

STEP 2

Inspect the instructions

STEP 3

Read, remember, reduce

STEP 4

Answer or abandon

STEP 5

Turn back

STEP 6

Estimate

STEP 7

Survey

FIGURE 1. Test-taking strategy steps.

they need to consider). In the fourth step,
Answer or Abandon, students supply an
answer for the item/question if they are
sure they know it, or, if they are unsure,
they abandon the item for the moment,
marking it with a symbol to indicate that
it must be answered later.
Students continue recycling through
the third and fourth steps of the strategy
for each of the remaining test items in the
first section of the test until they reach
the end of the section. At this point, they
reapply the Inspect-the-Instructions step
to the instructions for the section of the
test that they identified as next easiest
and then recycle through the third and
fourth steps of the strategy for each item
within the second section of the test. This
process is continued for each successive
test section until students reach the end
of the test.
They now complete the fifth strategy
step, Turn Back, by going back to the beginning of the test to answer abandoned
items. As they encounter the abandoned
items, students can scan the test for the
answer to the item (occasionally, one
item on a test will provide the answer to
another item) or use the sixth strategy
step, Estimate, to help them make a wise
guess for the correct answer for the item.
Students are taught that they should only
guess when they have no idea what the
JOURNAL

OF

answer is. If they must guess, they can

select from three guessing techniques (i.e.,
avoid absolutes, choose the longest or
most detailed option, and eliminate similar choices). These techniques were designed to correspond to the three most
common cues/clues found in teachermade and commercially prepared secondary classroom tests (Hughes, Salvia, &
Bott, 1991).
Finally, in the last strategy step, Survey, students scan the test to ensure that
they have answered all items, and if not,
answer them. Students are taught that
they should not change any of their answers unless they are sure that another
answer is better.

Materials and Measures

Testing Materials and Measures.
Testing materials consisted of 10 tests
which were constructed to be equivalent
and parallel. Hereafter referred to as
"Probe Tests,'* these tests were similar to
a test-wiseness test developed by Slakter,
Koehler, and Hampton (1970). Their purpose was to measure the students* use of
the Test-Taking Strategy.
The Probe Tests each had 29 items,
organized in four sections. Two sections
included a total of 12 true/false items,
and two sections included a total of 17
multiple-choice questions. These two

EMOTIONAL

AND

BEHAVIORAL

types of items were chosen because they

were the most frequently used types in
a sample of 200 secondary tests collected
from eight school districts in three states
(Hughes et al., 1991; Putnam, Deshler, &L
Schumaker, 1993; Putnam, 1992).
Each Probe Test contained five sets of
instructions. One set was provided for
each of the four sections of the test
(specifying how and where to respond).
Additionally, a general set of instructions
at the beginning of the test specified
where to write one's name, the date, and
the class period. Although the items and
instructions on the 10 tests were similar
in format, they were designed to be distinctly different in content to allow formative measurement of the generative use
of the Test-Taking Strategy.
Two types of items appeared on each
test: easy-to-answer items and fictitious
items. The former were constructed so
that students would know the answer.
An example of an easy-to-answer item is
Small dogs are called:
a.
b.
c.
d.

Colts
Puppies
Kittens
Cubs

These items were used to measure

whether the student eliminated (crossed
out) obviously incorrect response options
and answered the test item immediately
instead of abandoning it. Each test included five easy-to-answer items.
Additionally, items containing fictitious terms were constructed for the
Probe Tests so that the answer would not
be known. Thus, guessing was required
on these items because there was no
"real" answer. (These items were similar
to items previously used in the literature
to measure the use of guessing strategies [e.g., by Scruggs & Marsing, 1988;
Slakter et al., 1970]). An example of a fictitious item is

DISORDERS,

The most common use of bason solutions is:

a.
b.
c.
d.

fermentation
pigmentation
cleaning
as an organic compound in oil
solvents

JULY

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

I <) *M ,

VOL.

NO.

191

The 24 fictitious items included 16 that

sampled a student's ability to avoid absolutes (items/choices which contained
words such as never or always), four items
which sampled the student's ability to
eliminate similar options (two or more
choices that were so similar they could
be discounted as answers because only
one answer could be correct), and four
items that sampled the student's ability
to choose the longest, most detailed option (a multiple-choice option which is
obviously longer than the other options,
such as choice "d" in the above example).
Thus, these fictitious items enabled the
measurement of the student's use of the
three guessing techniques included in
the Estimate step of the strategy as well
as the student's choice to abandon the
item because the answer was not known.
A panel of three experts reviewed the
10 Probe Tests. These experts were professors in educational psychology who
taught test and measurement courses at
a major university. They concurred that
the test items would measure the skills
that they were designed to measure and
that the tests were equivalent with regard
to the number and types of items as well
as overall format.

Scoring Materials,
A Test-Taking
Checklist was designed for scoring whether the students were performing the steps
of the Test-Taking Strategy as they took
the Probe Tests. Each item on the checklist corresponded to one of the behaviors
required to perform the strategy steps.
Student strategy use either resulted in a
behavior that could be observed or resulted in a permanent product which could
be evaluated later by a scorer. For example, some of the items on the checklist
were whether the student made a mark
next to an item indicating that the item
was abandoned, whether the student underlined key words in the instructions,
and whether the student answered all the
questions on the test. The checklist contained 62 items. A percentage score for
a given test-taking performance was determined by dividing the number of testtaking responses observed during the
test-taking session or on the completed
test by the total number of items on the
192

JOURNAL

OF

EMOTIONAL

AND

checklist (62). Mastery was set at 90% of

the required behaviors.
Generalisation Measures. In addition to the Probe Tests and the TestTaking Checklist, which were used to
measure the students' acquisition of the
Test-Taking Strategy under controlled
conditions (i.e., in the training setting
using experimenter-constructed Probe
Tests), two generalization measures served
to assess students' use of the strategy outside the training setting. The first generalization measure was test scores from
regularly administered mainstream class
unit tests. One mainstream class (science
or social studies) was targeted for each
student, the teacher was contacted, and
the need for obtaining student test scores
was explained without explaining the
nature of the intervention. Several test
scores were gathered prior to and after
the test-taking instruction for each student from participating teachers. The
number of test scores varied across students depending on the number of tests
given in the respective classes. The mean
number of test scores collected per student was 12 (range = 9 to 14) during the
28 weeks of the study.
The second generalization measure, a
measure of generalized strategy use, was
obtained after students had mastered the
strategy. Two tests for each student from
the targeted mainstream class were gathered and were inspected to determine any
evidence that the student was using the
Test-Taking Strategy (e.g., markings indicating careful inspection of the instructions, abandonment of items, elimination
of absurd choices, all questions answered).
These tests contained multiple-choice and
true-false items only and were either
teacher-constructed tests or tests provided
in the teacher's edition of the texts used
in the class.

to the Probe Tests, Practice Tests were

shorter so that students could practice the
strategy as well as receive corrective feedback within one class period. Each practice test consisted of two sections (one
multiple-choice, the other true-false) containing five items each. The items were
similar to those on the Probe Tests; each
test included two easy-to-answer items
and eight fictitious items.
Three sets of instructions were included on each Practice Test; one for each
section and a general set at the beginning
of the test. Students could make the same
kinds of responses to items on the Practice Tests as on the Probe Tests. The experts who examined the Probe Tests also
studied the Practice Tests and determined
that they were equivalent in content and
format and that they could be used to
measure the skills they were intended
to measure. A shorter version of the TestTaking Checklist was designed to facilitate scoring of the students' practice
attempts. The 28 items on this checklist
corresponded to the steps of the TestTaking Strategy, as described above for
the longer form of the checklist. This version of the checklist was shorter than the
original checklist because of the smaller
number of test sections and test items on
the Practice Tests when compared to the
Probe Tests. A percentage score was derived by dividing the number of responses
observed by the total number of possible
responses. Again, mastery was set at 90%
of the items.

Instructional Materials and Measures. In order to standardize the instructional procedures, an instructional
manual (Hughes, Schumaker, Deshler, ck
Mercer, 1988) was written. It included
step-by-step instructions to be followed in
teaching the Test-Taking Strategy. Additionally, a set of 10 Practice Tests was
produced. Similar in format and content

Reliability.
Interscorer reliability
was determined by having two observers
independently score a random sample of
10 Probe Tests and four Practice Tests,
for a total of approximately 25% of the
tests given during the study. The completed checklists for a test were compared
item-by-item. An agreement was scored
if both observers marked a reponse as being present or if both observers had
marked a response as not present. The
percentage of agreement was calculated
by dividing the number of agreements by
the number of agreements plus disagreements and multiplying by 100. The observers agreed on 696 items out of 732
opportunities to agree for a total percentage of agreement of 95%. O n individual

BEHAVIORAL

VOL.

DISORDERS,

JULY

14 9 3 .

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

I,

NO.

tests, the percentage of agreement ranged

from 89 to 100%.

Procedures
Baseline Procedure. During baseline, a series of Probe Tests was administered to the students, who were given
25 minutes to complete each test. They
were told to answer each question as best
they could and to do what they typically
do to earn the best grade possible on a
test. They were also informed that the
purpose of the test was to determine how
they take tests. They were told that they
might not know the answers to some of
the items on the test, and, if this should
occur, they should make their best guess.
Finally, they were reassured that scores
earned on the test would not result in a
grade in any course in which they were
enrolled.
Instructional Procedure,
The instructional methodology comprised eight
stages similar to those suggested by Ellis,
Deshler, Lenz, Schumaker, and Clark
(1991) for promoting strategy acquisition
and generalization in adolescents with
learning disabilities. In the first instructional stage, the Pretest Stage, the information collected during baseline on student use of test-taking strategies was
shared with each student privately. Then
each student was asked to make a formal
commitment to learn the strategy by writing a goal to that effect. In the second instructional stage, the Describe Stage, the
purpose of the Test-Taking Strategy was
introduced, and the rationale for using
the strategy was discussed. Example situations for strategy use were also described.
Finally, the strategy steps were presented
along with examples of their application.
During this stage (as well as throughout
the instructional process), students were
told that this strategy was not meant to
take the place of studying and that use
of the strategy alone could not guarantee passing grades. They were also informed that the strategy was designed for
use on classroom tests versus standardized
tests.
In the third instructional stage, the
Model Stage, the strategy was applied to
an example test by the instructor. Specifically, he demonstrated how to comJOURNAL

OF

plete each strategy step in sequence while

thinking aloud so that students could see
the steps performed as well as hear the
cognitive processes involved in performing the strategy steps. In later parts of the
demonstration, student involvement was
enlisted to provide students with guided
practice in application of the strategy
steps as well as to check their understanding of the strategy to that point.
In the fourth instructional stage, Verbal Rehearsal, students were led in a
rapid-fire exercise for the purpose of verbally rehearsing and memorizing the steps
of the strategy. After the students had
practiced naming the steps for several
rounds, they were individually checked
for their ability to say the steps in sequence. When a student was able to
name all of the steps of the strategy in
order without prompting, she or he proceeded to the next instructional stage.
In the fifth instructional stage, Initial
Practice, students applied the first four
steps of the strategy (P, I, R, and A) to
Initial Practice Tests. They were told that
they had 10 minutes to complete a test
and instructed to perform the first four
strategy steps only. After each initial
practice attempt, the student's performance was scored using the items on the
short form of the checklist that related
to the P, I, R, and A steps of the strategy. Positive and corrective feedback was
provided to each student privately. Students continued this stage of instruction
until they made 90% of the required responses on one Practice Test.
In the Advanced Practice Stage, the
sixth stage of instruction, students were
administered an Advanced Practice Test
and were instructed to perform all the
strategy steps. They were given 15 minutes to complete the test. After each practice attempt, the student's responses were
scored using the short form of the checklist, and the student was given positive
and corrective feedback.
When a student reached the mastery
criterion by making 90% of the required
responses during an advanced practice attempt, she or he proceeded to the seventh
instructional stage, Posttest. In this stage,
successive Probe Tests were administered
until the student made at least 90% of the
required responses on a given Probe Test.

EMOTIONAL

AND

BEHAVIORAL

In Generalization, the last instructional stage, students were encouraged to use

the Test-Taking Strategy each time they
took a test in their mainstream classes or
elsewhere. Also, students discussed where
and when they would use the strategy.
Finally, they made cue cards to carry with
them, listing the strategy steps. After they
had taken tests in their mainstream classes, students were scheduled for individual discussions to determine (a) whether
they had used the strategy on their last
test, (b) if they found it helpful, and
(c) whether they had modified/adapted
the strategy to help them meet their
needs. Also during this stage, students
took Probe Tests approximately every
2 weeks to check the maintenance of
their use of the strategy steps (these Probe
Tests will hereafter be referred to as
"Maintenance Probe Tests").
The instructional process, excluding baseline Probe Tests and Maintenance Probe Tests, took approximately
12 sessions. Each session lasted about
20 minutes (range = 10 to 30 minutes).
Students were taught in groups of 4 to
6 students. Other students who were classified as learning disabled and having
other disabilities also received instruction
along with the subjects.
An observer made unannounced, random visits to the instructional sessions.
Using a checklist that listed all the components of the instruction, she recorded
whether the instructional procedures
were followed as described above and
whether or not the instructor provided
help to the students while they were taking the Advanced Practice and Probe
Tests.

Experimental Design
A multiple-probe design (Horner 6k Baer,
1978) was employed. Three students participated in each of two applications of
the design. All students received at least
two Probe Tests during baseline. When
the baselines of one student in each design was stable, that student received instruction that was followed by at least
two Probe Tests and then by at least one
Maintenance Probe Test. Once the first
student within each multiple-probe design
met mastery on Probe Tests, the second
and third students in that design were ad-

DISORDERS,

JULY

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

1991,

VOL

I,

NO.

193

ministered another baseline Probe Test.

Another student who demonstrated a
stable baseline at that point was then
given instruction within each design.
After the second student in a design met
mastery on two Probe Tests, the third
student in that design was administered
one more baseline Probe Test and received instruction in the strategy. All students were administered at least one
Maintenance Probe Test. Thus, all students experienced three conditions: baseline, instruction, and maintenance.

RESULTS

the marks the students made on their

tests, some students were using most of
the steps of the strategy while others were
using a few steps. During individual meetings which took place after the first test
was examined, students received feedback
on how they might use more parts of the
Test-Taking Strategy when taking their
next tests, and any modifications they
made in the strategy were discussed. Examination of the second set of tests indicated that all students were using a
majority of the steps of the strategy.
The other generalization measure was
scores/grades on mainstream class tests.
Average grades and scores during base-

ting for the six students was 88% with

scores ranging from 82% to 100%.

Generalization of Strategy Use

to Other Settings
Examination of the tests taken in the targeted mainstream classrooms revealed
some evidence of strategy use by all six
students (e.g., they wrote the word
PIRATES, underlined key words in directions, used an appropriate marking system to indicate "abandoned" items, and
crossed off absurd choices). Nevertheless,
when the first of the two mainstream tests
was examined, variability of strategy use
across students was noted. According to

Acquisition and Maintenance

of Strategy Use
Results of students' acquisition and maintenance of the Test-Taking Strategy are
displayed in Figures 2 and 3. Each figure
contains three graphs, with each graph
showing one student's results. Displayed
on the graphs are the percentage of points
earned by a student on the Test-Taking
Checklist for each test performance. During baseline, students received between
22% and 35% of the points available for
strategy performance on the Probe Tests
(M = 32%). All baselines were stable and
indicated students did not use many of
the test-taking behaviors included in the
strategy. The majority of earned points
came from following directions related to
writing one's name, the date, and the
class period in the correct places on the
test form as well as writing answers in the
correct place.
During instruction, all students, with
the exception of Student 5, scored above
the 90% criterion on Initial and Advanced Practice Tests as well as the Probe
Tests. Student 5 did not reach mastery
on performing required strategic behaviors on the very first Initial Practice Test
but did so on the other tests. Across students, the mean percentage of appropriate test-taking responses was 88% for
both Initial and Advanced Practice Tests.
The mean score for post-training Probe
Tests was 95%. During the maintenance
condition, which lasted between 3 to
11 weeks depending on when instruction
began for each student, the mean score
on probe tests taken in the training setJOURNAL

OF

EMOTIONAL

AND

100-^

Instruction

Hn#

Malntenanca

00-J
00-I

so-JJ
so
70 JJ
70

i J

60
50 -J
J
40
40 JJ

m * Probt Tests
\p initial Practice Tests
X T<Htf
Advanced Practice

-N

30 J
30
JU

20
2 0 --J
20-1

10 _J
J _J
10.
10

I1

1L

100-1
100^

E
co

11

70^

Ijl
HI

80-^

? *7

0 00 --1

LU

J1

6
0*
eoH

Q.

6 00--J

4 00--1

CD
CD
o>
cd
c

30-

*~*

22 00 -- J

0)

1
1I
1

10
IO.

C
0L.-

100-j

**** f

9
900 -- I
8
800 -J

700 -- I
7
6
600 -- I
5
0SO-I
4
400 -- I

xJ

30-

- j > - m

fj

2
020-1
1
100 -J ,

1 I1. I1, L
1 11
,11 , L1 I1 L

2 3 4

1 11

Con
Consecutive
Weeks

FIGURE 2.

Results for Subjects 1, 2, and 3.

BEHAVIORAL

DISORDERS,

JULY

19^!,

VOL.

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

1,

NO.

1 1 1 1

Iratruction

Mtinttnanot

>sJ

IX908070-

eo8

KiX

3020-

10.

med

Tgm

\
1

I I

I l l

100-j

CD

^ - '

90-

LU

/ "

80-

7080-

604030-

'

201

10

a.

1 1 1

10090-

DISCUSSION

80-

Results of this study show that six adolescents with EBD could successfully acquire
and maintain use of a comprehensive testtaking strategy. All students achieved
mastery on Probe Tests, and some students maintained use of the strategy
steps at close-to-criterion levels for up to
11 weeks after instruction. These findings replicate the results of the study by
Scruggs and Marsing (1988) who found
that a group of adolescents with EBD
could master four guessing strategies
which could be used in a test format
similar to classroom tests, and they also
extend these findings by showing that
students with EBD can learn to apply a
complex series of test-taking strategies.
Additionally, this study indicates that
students can generatively use a complex
strategy like the test-taking strategy across
different tests since they took a different
test each time they were asked to apply
the strategy. Also, evidence in the present
study of strategy use on actual tests tentatively demonstrates that the intervention might result in generalization across
settings and across types of tests. Since
five of the six students' test scores showed
improvement, some support is provided
for the notion that test-taking strategy use
may affect classroom test performance.
JOURNAL

* Probt Tsti
O Initial Practlot TMts
X AoVanctO Proctto*

6040-

erce

line and after instruction are shown in

Table 1. During baseline, the average
grade for five of the students was an "F,"
while the grade for the remaining student
was a low "D." The average score for the
group during baseline was 57% (range =
54 to 62). After instruction in the TestTaking Strategy, test scores improved for
five of the students while the test score
for Student 4 stayed about the same. As
noted in the table, this student had an
IQ score of 78, whereas the other students* IQ scores were in the normal
range. The average test score for the
group was 68%, which represents an average group improvement of 11 percentage
points over scores earned in baseline. In
summary, three students* test score averages improved by one letter grade; two
students' grades improved by two or
more letter grades; and one student's
grade stayed the same.

OF

/ / -

70-

80504030-

9)

2010

1 2

1 1
6

1 1
9

10

11

12 13

15 16

17

18

Consecutive Weeks

FIGURE 3.

Results for Subjects 4, 5, and 6.

Interestingly, the results of the present

study are similar to those reported in testtaking studies with adolescents with learning disabilities (LD) (Hughes ck Schumaker, 1991; Lee & Alley, 1981). For example, in the Hughes and Schumaker study,
students with LD gained an average of
14 percentage points on their mainstream
classroom tests after receiving instruction
in the same test-taking strategy as taught
in the present study. Lee and Alley's findings were approximately the same when
a similar test-taking strategy was taught.
Also, the students with LD in the Hughes
and Schumaker study required about the
same number of trials to reach the mastery criterion (an average of two trials in
initial practice and two trials in advanced
practice) as the students with BD in this

EMOTIONAL

AND

BEHAVIORAL

study. Such results indicate that the instructional methodology used in the present study may be equally effective with
adolescents with EBD and LD.

Limitations
The conclusions that can be drawn from
this study are limited by a few factors.
First, grades on tests taken in mainstream
classes were found to be an unstable
measure. Mainstream test scores varied
by as many as 50 percentage points during baseline for many of the students, and
this variability was also apparent, albeit
to a lesser degree, after students began applying the Test-Taking Strategy to their
tests. The sources of the variance in these
mainstream test scores within a given
condition might include study time, ab-

DISORDERS.

JULY

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

1 9 n 3,

VOL.

1,

NO.

J95

TABLE l

Mean Grades Earned by Students on Mainstream Tests

After instruction

Baseline
Student

Grade

1
2
3
4
5
6
Total mean

54%
55%
55%
58%
58%
62%
57%

(F)
(F)
(F)
(F)
(F)
(D)
(F)

(SD)

Grade

16.4
14.1
12.3
13.4
16.6
14.6
8.4

61%
72%
81%
58%
66%
72%
68%

senteeism, type of test, student interest

in a topic, prior knowledge of the content tested, the quality of teacher instruction on a given unit, test item formats,
and presence or absence of cues on tests.
The only way to control for such variables, and thus clearly assess the generalization of a test-taking strategy, would be
to have students in both control and experimental groups enrolled in the same
mainstream class and take the same tests
under the same conditions (Scruggs &
Marsing, 1988). Tests would have to be
standardized across time regarding such
variables as difficulty, type, and format.
This was not possible within the scope
of the current study and with the student
populations available to these researchers.
Another limitation of this study concerns the use of tests containing only
multiple-choice and true-false items.
Whether the Test-Taking Strategy taught
in this study would be effective with other
types of tests such as essay tests or standardized tests is unclear. An additional
limitation of the study is the relatively
small number of participating subjects.
Whether the instructional methodology
would be as effective with other students
with EBD is unknown.
Finally, the magnitude of the gain
made by the students on their mainstream class tests is worth further consideration. Although an 11% gain in
test scores might represent a one grade
change for many students (i.e., those
already passing or those just below the
passing range), students who are performing at very low levels (earning grades of
20-30-40%) might not realize letter-grade
improvement, and, indeed, might still be
1%

JOURNAL

OF

EMOTIONAL

AND

(D)
(C)
(B)
(F)
(D)
(C)
(D)

(SD)

IQ

15.7
8.1
3.9
7.0
11.7
4.2
3.2

118
101
101
78
90
91

failing even though they are using the

strategy. Since the time of this study,
the instructional methodology has been
elaborated (see Ellis et al., 1991), and
conceivably, additional gains might be
realized with this latest version of the
methodology. Nevertheless, the TestTaking Strategy might still not be the
best strategy for students who are performing on tests at very low levels; they
might need to master several strategies
before they become successful at learning new information and expressing their
knowledge on tests. Thus, replication and
extensions of this study are needed.

Instructional Implications
The instructional procedures used in this
study, as well as in studies with adolescents with LD (e.g., Lenz & Hughes,
1990; Schmidt, Deshler, Schumaker, &
Alley, 1989; Schumaker, Deshler, Alley,
Warner, & Denton, 1982) appear to be
effective for teaching learning strategies
to students with different exceptionalities
across several academic tasks (e.g., reading, writing, test-taking). Additionally,
the instructional methodology appears to
promote generalized use of strategies
across similar academic tasks and outside
the training setting (see Deshler & Schumaker, in press, and Schumaker &L Deshler, in press, for reviews of the studies in
this area).
One important aspect of this type of
instruction is its intense nature. For example, instruction must take place daily
over a several week period, and students
must be provided multiple opportunities
to practice the strategy to mastery. The
importance of the intensity of instruction
BEHAVIORAL

DISORDERS,

JULY

I 9g J,

has been shown consistently to be a critical instructional variable in studies of

test-taking training with nonhandicapped
students (e.g., Bangert-Drowns, Kulik, &
Kulik, 1983; Scruggs, White, & Bennion,
1986). Thus, instructors contemplating
implementing this type of instruction
need to be aware of the commitment they
must make to enable students to be successful.
One point worth noting is that all participating students stated they liked learning and using the Test-Taking Strategy.
Relatedly, the first author used no behavior management systems during instruction and observed very little inappropriate
behavior on the part of the students.
While this lack of acting out or off-task
behavior may have been in part due
to the novelty of the trainer, students
appeared to be motivated to learn the
strategy.
Several students made adaptations to
the strategy when using it on their mainstream tests. For example, one student did
not like writing the acronym PIRATES
on the test form, one did not like looking over the test before beginning because
it made him nervous, and one student
said he did not need to underline key
words in the directions because he had
no problems following directions (an examination of his tests demonstrated this
to be true). Thus, the students omitted
these steps. Because student modification
of strategy use is desirable when modifications do not hinder their overall use of
the strategy, they were told their changes
were appropriate.

Cautions
This test-taking intervention should be
considered within the context of a major
trend in the field of special education today. Specifically, because of an increased
emphasis on "inclusionary placements,"
students with disabilities are being assigned to regular classroom settings for
the majority, if not all, of the school
day. Thus, the face validity of teaching
the Test-Taking Strategy to students enrolled in mainstream placements is high,
given the expectation for students to
take a large number of teacher- and/or
publisher-made tests as a part of the demands of the regular class setting. This
VOL.

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

I,

NO.

study demonstrated that students with

EBD were capable of mastering the TestTaking Strategy when they received intensive instruction in a pull-out support
class setting. Furthermore, this study
demonstrated that they applied the strategy to meet demands placed on them in
the regular classroom. Nevertheless, the
characteristics of the instruction that was
offered to these students under the pullout instructional configuration in this
study are important to understand. Specifically, in this setting, students had
ample opportunities to respond as they
practiced using the strategy and they
received immediate and detailed feedback
on the nature of their responses. In short,
the instruction that they received was
strategy-centered, intensive, and consistent.
One of the major instructional challenges facing regular classroom teachers
under the current inclusionary movement
is to replicate the favorable instructional
conditions found in pull-out arrangements that have traditionally enabled
special education teachers to provide intensive instruction in areas where students are skill or strategy deficient. In the
regular secondary classroom, teachers are
under considerable pressure to teach large
amounts of content in relatively short
periods of time. Consequently, the time
available for teaching students specific
learning strategies and other "processes
of learning" are limited. Furthermore,
when such strategies are taught to large
groups of students (e.g., 25 students or
larger) teachers have difficulty giving all
students (especially those at-risk for failure) ample opportunities to practice the
strategy, to give them individualized feedback, and to insure that they have internalized the strategy for automatic, independent use. While the findings of this
research study are very encouraging, a
large number of challenging instructional
questions remain to be answered if such
strategies are to be successfully taught to
students with EBD within regular classrooms. Until these questions are addressed and methods of teaching students
with EBD these strategies within regular
classrooms have been validated, intensive
instructional services for these students
need to be continued such that they can
JOURNAL

OF

strategies. Focus on Exceptional Children, 23,

1-24.
Epstein, M.H., 6k Cullinan, D. (1988). Selected research issues in the education of
adolescents with behavioral disorders.
About the Authors
Monograph in Behavioral Disorders, 11,
CHARLES A. HUGHES is an associate profes106-118.
sor of special education at The Pennsylvania Epstein, M.H., Kinder, D., 6k Bursuck.
State University. He received the PhD at the
(1989). The academic status of adolescents
University of Florida. His professional interwith behavioral disorders. Behavioral Disests include teaching adolescents with learnorders, 14, 157-165.
ing and behavioral problems to self-manage Fessler, M.A., Rosenberg, M.S., &L Rosenacademic and social behaviors. DONALD D.
berg, L.A. (1991). Concomitant learning
DESHLER is a professor in the special educadisabilities and learning problems among
tion department and director of the Center
students with behavioral emotional disfor Research on Learning at the University
orders. Behavioral Disorders, 16, 97-106.
of Kansas. He received the PhD at the Uni- Foley, R.M., Cullinan, D., &. Epstein, M.H.
versity of Arizona. His research interests
(1990). Academic and related functioning
focus on the development and validation of
of mainstreamed and nonmainstreamed
educational interventions for at-risk youth
seriously emotionally disturbed students.
and young adults. KATHY L. RUHL is an asMonograph in Behavioral Disorders, 13,
sistant professor of special education at The
80-89.
Pennsylvania State University. She received
Forness, S.R., &. Dvorak, R. (1982). Effects
the PhD at the University of Florida. Her
of test time-limits on achievement scores of
professional interests include teaching stubehaviorally disordered adolescents. Behavdents with behavior disorders and teacher
ioral Disorders, 7, 207-212.
training. JEAN B. SCHUMAKER is a courtesy
Horner,
R.D., 6k Baer, D.M. (1978). Mulprofessor in the Departments of Human
tiple-probe
technique: A variation of the
Development and Special Education and asJournal of Applied Behavmultiple
baseline.
sociate director of the Center for Research
ior Analysis, 11, 189-196.
on Learning at the University of Kansas. She
received her PhD at the University of Kansas. Hughes, C.A., Ruhl, K., 6k Korinek, L.
(1985). Analysis of predictive research with
Her research interests focus on the developadolescent behaviorally disordered students
ment and validation of educational intervenin public school settings. Programming for
tions for at-risk youth and young adults.
Adolescents with Behavioral Disorders, 2,
Address: Charles A. Hughes, 202 Cedar, Spe105-117.
cial Education Program, The Pennsylvania
State University, University Park, PA 16802. Hughes, C.A., Salvia, J., 6k Bott, D. (1991).
The nature and extent of test-wiseness cues
in seventh- and tenth-grade classroom tests.
References
Diagnostique, 16, 153-163.
Bangert-Drowns, R.L., Kulik, J.A., & Kulik, Hughes, C.A., Schumaker, J.B., Deshler,
D.D., 6k Mercer, C. (1988). The test-taking
C.C. (1983). Effects of coaching programs
strategy: Instructor's manual. Lawrence, KS:
on achievement test performance. Review
Edge Enterprises.
of Educational Research, 53, 571-585.
Coutinho, M.J. (1986). Reading achievement Hughes, C.A., 6k Schumaker, J.S. (1991).
Test-taking strategy instruction for adolesof students identified as behaviorally discents with learning disabilities. Exceptionordered at the secondary level. Behavioral
ality, 2, 205-221.
Disorders, 11, 200-207.
Laycock,
V.K., 6k Korinek, L. (1989). ToDeshler, D.D., Putnam, M.L., &. Bulgren,
ward least restrictive curriculum for behavJ.A. (1985). Academic accommodations for
iorally disordered adolescents. Programming
adolescents with behavior and learning
problems. Programming for Adolescents with for Adolescents with Behavioral Disorders, 4,
11-25.
Behavioral Disorders, 2, 20-30.
Deshler, D.D., 6k Schumaker, J.B. (in press). Lee, P., 6k Alley, G.R. (1981). Training junior
high school LD students to be a test-taking
Strategy mastery by at-risk students: Not
a simple matter. The Elementary School strategy (Research Report No. 38). Lawrence, KS: The University of Kansas InstiJournal.
tute for Research in Learning Disabilities.
Ellis, E.S., Deshler, D.D., Lenz, B.K., Schumaker, J.B., 6k Clark, F.L. (1991). An in- Lenz, B.K., 6k Hughes, C.A. (1990). A word
structional model for teaching learning
identification strategy for adolescents with
truly master the strategies they need to
become successful learners and performers in the educational environment.

EMOTIONAL

AND

BEHAVIORAL

DISORDERS,

JULY

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

MLM,

VOL.

1,

NO.

197

learning disabilities, journal of Learning Disabilities, 23, 137-200.

Luebke, J., Epstein, M.H., & Cullinan, D.
(1989). Comparison of teacher-rated achievement levels of behaviorally disordered,
learning disabled, and nonhandicapped
adolescents. Behavioral Disorders, 15, 1-8.
Mastropieri, M.A., Jenkins, V., & Scruggs,
T.E. (1985). Academic and intellectual
characteristics of behaviorally disordered
children and youth. In R.B. Rutherford, Jr.
(Ed.), Severe behavior disorders of children and
youth (Vol. 8, pp. 86-104). Reston, VA:
Council for Children with Behavioral Disorders.
Prescott, G.A., Balow, I.H., Hogan, T.R., &
Farr, R.C. (1984). Metropolitan Achievement
Test 6: Survey Baiters. San Antonio, TX:
The Psychological Corporation.
Putnam, M.L. (1992). The testing practices
of mainstream secondary classroom teachers. Remedial and Special Education, 13,
11-21.
Putnam, M.L., Deshler, D.D., 6k Schumaker,
J.B. (1993). The investigation of setting
demands: A missing link in learning strategy instruction. In LJ. Meltzer (Ed.), Strategy assessment and instruction for students with
learning disabilities. Austin, TX: PRO-ED.
Ruhl, K.L., 6k Berlinghoff, D.S. (in press).
Research on improving behaviorally dis-

ordered students' academic performance: A

review of the literature. Behavioral Disorders.
Schmidt, J.L., Deshler, D.D., Schumaker,
J.B., & Alley, G.R. (1989). Effects of generalization instruction on the written language performance of adolescents with
learning disabilities in the mainstream classroom. Journal of Reading, Writing and Learning Disabilities, 4, 291-309.
Schumaker, J.B., & Deshler, D.D. (1992).
Validation of learning strategy interventions for students with learning disabilities:
Results of a programmatic research effort.
In B.Y.L. Wong (Ed.), Contemporary intervention research in learning disabilities: An international perspective (pp. 22-46). New
York: Springer-Verlag.
Schumaker, J.B., Deshler, D.D., Alley, G.R.,
Warner, M.M., 6k Denton, P.H. (1982).
Multipass: A learning strategy for improving reading comprehension. Learning Disability Quarterly, 5, 295-304.
Scruggs, T.E., 6k Lifson, S. (1985). Current
conceptions of test-wiseness: Myths and
realities. School Psychology Review, 14,
339-350.
Scruggs, T.E., & Marsing, L. (1988). Teaching test-taking skills to behaviorally disordered students. Behavioral Disorders, 13,
240-244.

Scruggs, T.E., 6k Mastropieri, M.A. (1986).

Improving the test-taking skills of behaviorally disordered and learning disabled students. Exceptional Children, 53, 63-68.
Scruggs, T.E., Mastropieri, M.A., 6k TolfaVeit, D. (1986). The effects of coaching on
the standardized test performance of learning disabled and behaviorally disordered
students. Remedial and Special Education, 7,
37-41.
Scruggs, T.E., Mastropieri, M.A., Tolfa, D.,
6k Jenkins, V. (1985). Attitudes of behaviorally disordered students toward tests.
Perceptual and Motor Skills, 60, 467-470.
Scruggs, T.E., White, K., 6k Bennion, K.
(1986). Training test-taking skills to elementary grade students: A meta-analysis.
Elementary School Journal, 87, 69-82.
Slakter, M.J., Koehler, R.A., & Hampton,
S.H. (1970). Grade level, sex, and selected
aspects of test-wiseness. Journal of Educational Measurement, 7, 119-122.
U.S. Department of Education. (1990).
Twelfth annual report to Congress on the implementation of the Education of the Handicapped Act. Washington, D C : U . S .
Government Printing Office.
Wechsler, D. (1974). Manual for the Wechsler
Intelligence Scale-Revised. Cleveland: The
Psychological Corporation.

{CONTINUED FROM PAGE MS}

Reid, J.B. (1986). Social interactional patterns

Shores, R.E., Gunter, P.L., 6k Jack, S.L.
in families of abused and nonabused chil(1993). Classroom management strategies:
dren. In C. Zahn-Waxler, M. Cummings,
Are they setting events for coercion? Behav&L M. Radke-Yarrow (Eds.), Social and biioral Disorders, 18, 92-102.
ologicalorigins of altruism and aggression (pp.
Shores, R.E., Jack, S.L., Gunter, P.L., Ellis,
238-255). New York: Cambridge UniverD.N., DeBriere, T.J., & Wehby, J.H. (1993).
sity Press.
Analysis of classroom interactions of chilRepp, A C , 6k Deitz, D.E.D. (1979). Reinforcedren with aggressive behavior. Journal
ment-based reductive procedures: Training
of Emotional and Behavioral Disorders, 1,
and monitoring performance of institu27-39.
tional staff. Mental Retardation, 17', 221-226.
Simpson, R.L., 6k Souris, L.A. (1988). ReciRepp, A.C., Deitz, S.M., 6k Deitz, D.E.D.
procity in the pupil-teacher interactions of
(1976). Reducing inappropriate behaviors
autistic and mildly handicapped preschool
in classrooms and individual sessions
children.
Behavioral Disorders, 159-168.
through D R O schedules of reinforcement.
Shores, R.E. (1987). Overview of research on
social interactions: A historical and personal perspective. Behavioral Disorders,
223-241.

Snyder, J., 6k Patterson, G.R. (1986). Effects

of consequences on patterns of social interaction: A quasi-experimental approach
to reinforcement in natural interaction.
Child Development, 59, 1257-1268.

Sprick, R.S. (1981). The solution book: A guide

to classroom discipline. Chicago: Science
Research Associates.
Tremblay, A., Strain, P.S., Hendrickson,
J.M., 6k Shores, R.E. (1981). Social interactions of normal preschool children. Behavior Modification, 5, 237-253.
Wacker, D., Steege, M., Northup, J., Reimers, T., Berg, W., 6k Sasso, G. (1990). Use
of functional analysis and acceptability
measures to assess and treat severe behavior
problems: An outpatient clinic model. In
A . C . Repp & N.N. Singh (Eds.), Perspectives on the use of nonaversive and aversive
interventions for persons with developmental
disabilities (pp. 349-359). Sycamore, IL:
Sycamore.
Weeks, M., & Gaylord-Ross, R. (1981). Task
difficulty and aberrant behavior in severely
handicapped students. Journal of Applied
Behavior Analysis, 14, 463-499.

J9g

BEHAVIORAL

VOL.

Mental Retardation, 14, 11-15.

JOURNAL

OF

EMOTIONAL

AND

DISORDERS,

JULY

1^),

Downloaded from ebx.sagepub.com at MICHIGAN STATE UNIV LIBRARIES on February 24, 2013

1 ,

NO.