VOLUME

1 4 3. N o ,

4,

1998

PROBLEM-SOLVING STRATEGIES EOR

TEACHING MATHEMATICS TO DEAE STUDENTS

hrcc teaching and learning strategies for problem solving were implemented
with first- and second-year deaf college students enrolled in mathematics
courses at the National Technical Institute for the Deaf (NTID), Rochester
Institute of Technology, These strategies involved the students in ( a ) giving
an explanation to a peer ohserver in sign language, after which they would put
their understanding of a prohlem and its solution in writing; ( b ) visualizing
the prohlem-solving process prior to starting to solve a problem; and ( c )
observing their teacher modeling the analytical process step by step for a
sample problem prior to solving math word prohlems. The students were asked
to solve two types of problems: typical word problems, and a visual/
manipulative puzzle that would provide a problem-solving experience that
would contrast with the experience of solving a problem presented in text
format. The results showed that these kinds of instructional strategies can
enhance the problem-solving performance of deaf and hard of hearing college
students.

KEITH MOUSLEY AND RONALD R.

KEU-V

Mousley is an assistant
professor in the
Department of Physics and
Technlcai Mathematics at
the Nationai Technicai
institute for the Deaf
(NTiD), Rochester institute
of Technoiogy, Rochester,
NY. Keiiy chairs the
Department of Educationai
and Career Research at
NTiD.

VOLUME 143, No, 4.

1998

The need to address the problem-solving

and critical thinking skills of deaf students
throughout the curriculum has long been
emphasized by educators and researchers
in deaf education (Furth, 1973; Luckner,
1992; Luckner & McNeill, 1994; Martin,
1984, 1993; Mousley, 1991; Rohr-Redding,
1985),
Several factors appear to contribute to
the difficulties experienced by deaf students with respect to the problem-solving
process and general reasoning skills, as
well as those specific to mathematical
problems. One such difficulty is in building metacognitive skills. Loera and
Meichenbaum (1993) and Martin (1993)
argue that metacognitive skills are critical
to literacy development in deaf learners
specifically, the cognitive functions emphasized by Feuerstein (1980), such as
taking note of all available data, being systematic in exploring all options, being pre-

cise, focusing on relevant information

only, and overcoming trial-and-error behavior, to name a few.
Second is the tendency of many deaf
students to proceed too quickly to solve a
problem without pausing to think it
through or develop a coherent plan.
While hearing students may exhibit similar
behavior, impulsivity is a behavior consistently observed in deaf students. (Impuisivity is defined as a characteristic of cognition that results in failure to reflect on
the appropriateness or conectness of an
approach to a task or problem, particularly
when the task or problem is perceived to
contain high ambiguity or uncertainty; see
Kagan, 1966; Kagan, Moss, & Siegal, 1963).
Impulsivity, a lack of reflective thinking,
and immaturity have been documented by
several researchers (e.g,, Altshuler,
Deming, Vollenweider, Rainer, & Tendler,
1976; Eabon, 1984; Freeman, 1979;

AMERICAN ANNALS OF THE DEAF

PROBLEM-SOLVING STRATEGIES FOR TEACHING MATHEMATICS TO DEAF SIVDENTS

Meadow & Schlesinger, 1971; Meadow

& Trybus, 1979). In his discussion of
impulsivity and deaf students,
Campbell (1986) described 10 behavioral characteristics that impulsive children, adolescents, and adults would
likely exhibit, several of which are
pertinent to problem solving:
a tendency to work quickly and
make errors;
an inability to attend to and to sort
out relevant features or information,
given a problem situation;
an inability to analyze a problem
carefully, with preference given to a
global or holistic approach.
Such impulsive behavior not only
results in failure to solve the problem
at hand, but also limits one's capacity
to learn from the experience. Impulsive behavior may be related to limited
language abilities that lead to stress
and frustration when one is trying to
deal with or clarify ambiguous situations, as suggested by Furth (1973).
Feuerstein (1980) has also emphasized
the importance of restraining impulsivity when formulating problem solutions.
Third, specific to mathematics and
problem solving, it has been noted
that most of the difficulties experienced by deaf school children are pertinent to solving written math problems (Pau, 1995). Linguistic difficulty
with math word problems has been
identified as a problem area for young
deaf children (Barnham & Bishop,
1991), and reading comprehension
level has been directly related to the
problem-solving abilities of deaf children 8 to 12 years of age (Pau, 1995).
College students enrolled in firstyear and second-year math courses at
the National Technical Institute for the
Deaf (NTID) demonstrate a considerable variability in problem solving that
includes impulsivity, lack of transfer,
breakdowns in reasoning, inability to
organize and properly consider all the
relevant information in the problem
solution, and misunderstanding of the
problem goals. While a distribution of
skills and abilities is to be expected

VoLUME 143, N o . 4. 1998

with any grouping of students, college-level deaf students exhibit this

range of difficulties with a variety of
math problem-solving tasks, particularly when presented these tasks in
text format as typical word problems.
To address such students' needs,
the educational and research literature
offers teachers a number of learning
and teaching strategies that are potentially beneficial to deaflearners' development of critical thinking and problem-solving skills.
For example, research has shown
that the cognitive performance of deaf
learners can be influenced with appropriately structured mediated learning
experiences (Haywood, ToweryWoolsey, Arbitman-Smith, & Aldridge,
1988; Keane & Kretschmer, 1987; Martin & Jonas, 1986; Tzuriel & Caspi,
1992). Mediated learning refers to
those human interactions that generate
the capacity of individuals to change,
to modify themselves in the direction
of greater adaptability and toward the
use of higher mental processes
(Feuerstein, 1980). With mediated
learning, students' direct experiences
with environmental stimuli are transformed through interaction with an
experienced and knowledgeable individual as part of the learning activity.
The mediated interactions are structured to help students develop systematic and generalizable processes of
thought. For example, the Instrumental Enrichment (IE) program developed by Feuerstein (1980) is a
metacognitive approach that gives students repeated opportunities to reflect
on their own thinking processes. In
their study of IE with deaf adolescents,
Martin and Jonas (1986) demonstrated
that the program had efficacy and potential in regard to improving these
adolescents' thinking skills. By the
end of the second year of the study,
the students in the experimental group
showed improvement in problem
solving and logical thinking, and more
frequently demonstrated a variety of
cognitively based behaviors in the
classroom, as shown by teacher obser-

vation checklists. Furthermore, Martin

and Jonas showed that adolescence is
not "too late" to make important and
measurable modifications in the cognitive functioning of deaf and hard of
hearing learners.
It has also been demonstrated that
general strategy instruction can improve the problem-solving ability of
deaf students. Campbell (1989) provided a cognitive behavior modification program consisting of general
strategy instruction, including self-instruction and feedback, to the experimental group of deaf students in his
study. This strategy training was derived from an information-processing
model that stressed elaboration and
reflection on aspects of input, process,
and output stages, and was designed
to enhance reflective problem-solving
ability. The treatment group in
Campbell's study demonstrated improved problem-solving ability, with
highly impulsive students showing the
most gains. Because problem solving
and skill transference require students
to be conscious of their own thinking
processes, mediated learning experiences and strategy training for reflective thinking have excellent potential
to benefit development in these areas.
Other instructional learning strategies with the potential to benefit the
development of reflective thinking and
skill transference have been identified
by Pressley, Snyder, and Cariglia-Bull
(1987), who suggested six alternative
strategies to enhance transfer in problem solving:
1. discovery and guided-discovery
learning of strategies;
2. instruction that is largely observational learning; watching strategy use
and reading about others using
strategiese.g., pairing problem solving and symbolic modeling;
3. guided participation in frequently
encountered normal school tasks;
4. strategy instruction by means of
books and courses
5. direct explanation: strategy instruction that is largely teacher directed;
6. dyadic instruction, in which stu-

AMERICAN ANNALS OF THE DEAF

dents interact with more mature

thinkers who are modeling good
thinking and guiding problem solving, often by providing cues to
students when they cannot manage on
their own.
Woditsch (1991) recommended
two techniques that also have potential to help students develop their own
thinking: vocalization, thinking out
loud; and writing, recording or taking
notes of the process experience.
Although some of the problemsolving difficulties observed with deaf
students are likewise experienced by
hearing students, the diversity of problem-solving experiences, skills, and
related knowledge represented in any
given math class of deaf students
(even at the college level) makes it
difficult for a teacher to take a unified
approach to presenting and teaching
content that involves problem solving.
Furthermore, such a situation complicates a teacher's efforts to apply instructional recommendations from the
research literature, because such recommendations often do not seem applicable to the diverse students in a
given classroom. Thus, in selecting
innovative strategies, a teacher under
these conditions could use an action
research approach, in which data are
collected and analyzed for the specific
group of students at hand.
The present study examined the
efficacy of several problem-solving
strategies with deaf students who were
enrolled full-time at NTID (one of
eight colleges of the Rochester Institute of Technology, Rochester, New
York) and who were taking undergraduate math courses (from Foundations of Algebra up to Concepts of
Calculus) taught by the lead author.
An advantage of this approach is that
research and data collection were integrated into the students' classroom
activities. A limitation was that the
natural enrollment patterns resulted in
relatively small numbers of participating students.
All of the students enrolled in each
of the classes in which a specific prob-

lem-solving strategy was implemented

were included in the study. Cumulatively, a total of 46 NTID students participated. There was no overlap of
student participants between the research on problem-solving strategies 1
(peer observer with signed and written
explanations) and 2 (visualization of
moves prior to solving a manipulative
game puzzle). However, five students
who participated in the research on
problem-solving strategy 2 were enrolled in a course the following year in
which research on problem-solving
strategy 3 (teacher modeling for wordproblem solving) was conducted.
Given the dissimilarity of problem
tasks between strategies 2 and 3, we
considered this overlap of five students acceptable. All of the participating students met the NTID admissions
requirement of at least a 70 dB hearing
loss in their better ear. The procedures for obtaining voluntary consent
of the participating students were approved by the NTID Institutional Review Board for research subjects.
Method

Strategy 1: Peer Observer With

Signed and Written Explanations
The first problem-solving strategy that
was implemented involved the use of
a classmate as an observer. The college student solving the problem was
required to explain in sign language
the goal and rules of the problem to
the observer prior to starting the problem, and then explain his or her strategy for solving the problem at the end.
Subsequently, the students were also
asked to write down the goal, rules,
and strategy they used to solve each
problem. This strategy had a twofold
purpose: ( a ) to show through explanation that students clearly understood
the problem-solving goal and rules,
and ( b ) to assess whether the students' reading levels would affect their
written and signed explanations, as
well as problem-solving performance.
As noted previously, Pau (1995) has
shown that reading comprehension

levels are related to the problem-solving ability of deaf children ages 8 to 12

years. Also, vocalization"thinking
out loud" (for the deaf students in this
study, explaining in sign language)
and writing are two techniques suggested by Woditsch (1991) to help students think through a process and develop their own thinking skills.
This initial study presented two
problems for the students to solve: one
was a visual/manipulative game
puzzle called the Tower of Hanoi that
has only one correct solution in practice, but can vary both in terms of
number of moves and time to solution;
the second was a math word problem
with multiple correct solutions depending on which decision or directional perspective each individual student made in calculating the solution.
Participants
Ten students participated in the strategy 1 exercise. Two were 30 years
old, one was 25 years old, and the remaining seven ranged in age from 19
to 23 years. Six of the students were
female and four were male; all had
hearing parents. While the sexes were
equally represented among the students in the two reading-level groups,
this was not by design of the research.
Regardless of the age range, all of the
students appeared to have the appropriate level of math skills and seemed
typical of the students participating in
undergraduate college math at NTID.
The average assessed functional levels
for math at time of admission to NTID
for the five students in the higher reading group were 10.0 on the Stanford
Achievement Test (SAT) for math computation, 10.7 for math concepts, and
9.5 for math applications. For the five
students in the lower-level reading
group, the respective averages were
11.5 for math computation, 10.8 for
math concepts, and 8.2 for math applications, Reading grade level was assessed by the California Reading Test
at time of entry into NTID. The five
students in the higher-level reading
group had an average grade level of

-fgl 1
VOLUME 143, No. 4,

1998

AMERICAN ANNALS OP THE DEAF

PjtOBlEM-SOLVIIVG STRATEGIES POP TEACHING MATHEMATICS TO DEAP STUDENTS

10.8, with a range of 10.3 to 11.1. The

five students in the lower-level reading
group had an average reading grade
level of 7.9 and ranged from 6.9 to 9.7.
All of the students used sign language
(signed English or American Sign Language [ASL]) for classroom and daily
communication. Five of these 10 students were considered advanced signers based on their functional levels as
assessed by the math teacher, who
was deaf. This assessment was supported by the self-ratings of these five
students at 4 or higher on the 5-point
scale of the NTID Language Background Questionnaire (LBQ) at time of
entry into NTID. The self-ratings on
the LBQ reflect a high degree of congruence with the formal measures of
communication skills done by trained
sign evaluators (Metz, Caccamise, &
Gustafson, 1997).
Procedure
The Tower of Hanoi is a manipulative

game puzzle that has been used in

developmental studies with young
children (Piaget, 1976) and with young
deaf subjects (Luckner & McNeill,
1994). An advantage of the Tower of
Hanoi is that one can increase the
number of manipulative disks to be
moved in order to increase the challenge of solving the problem with the
fewest moves and in the shortest possible time. Briefly, the Tower of Hanoi
consists of three towers (or pegs), on
one of which is a stack of doughnutshaped disks of different diameters.
For the present study, a four-disk configuration was used. The minimum
number of moves to solution for a
four-disk tower is 15 moves. The
problem task is to move all the disks
from the one tower (or peg) to a target
tower and end up with the same
stacked order of disks by moving only
one disk at a time while never placing
a larger disk on top of a smaller disk.
This requires the player to use all three

towers, including the third nontarget

tower, to move the disks one at a time.
The most effective strategy is to move
the top three disks to the third,
nongoal peg (in 7 moves), so that the
largest disk can be moved to the goal
peg (on the eighth move), followed by
movement of the top three disks to the
goal peg (which requires 7 more
moves), for a total of 15 moves. Figure
1 shows both the starting position of
the four disks on the Hanoi Tower and
the successful completion goal of the
four disks stacked in the same descending order on the target tower.
Each participant was tested individually, with another student as observer. The participants and observers
were matched to their most comfortable communication mode. Instructions and testing were conducted by
the lead author, who teaches math, is
deaf, and is fiuent in ASL.
At the time of testing, all participants were asked if they had ever seen

Figure 1
The Tower of Hanoi Puzzle

VOLUME 143, No. 4. 1998

Starting Pag

Goal Pag

Starting Pag

Goal Peg

AMERICAN ANNALS OF THE DEAF

to explain them to the observer.

Then, the participant and observer
were separated and asked to write
their individual summaries of how the
problem was solved.
After completing the tasks for the
Tower of Hanoi, each participant was
immediately presented with a math
word problem. The presentation of
this second problem involved a protocol similar to that used with the Tower
of Hanoi. Again, the lead author, who
teaches math, explained the word
problem to the individual participant
in sign language, while referring to the
typewritten copy of the word problem. The participant was then allowed to read the math word problem
and ask any clarifying questions. At
the start of solving the word problem,
each participant explained the problem and his or her goal for solution to
Briefly, the instructions to each
the peer observer, in sign language.
participant stated the goal of moving
Half the participants were presented
the stack of four disks from the first
with a shirt sale problem, while the repeg (red tower) to the third peg (green
maining half were presented with a
tower) and ending with the same degrass seed sale problem, to prevent
scending order. The middle peg was
any peer observer from solving the
yellow. The four rules were also exsame problem upon becoming the
plained: ( a ) Move only one disk at a
problem solver. Participants were intime. ( b ) Never place a larger disk on
formed that the word problem pretop of a smaller disk. ( c ) Do not hold
sented to them had more than one
a disk while moving another disk. ( d)
correct answer, and that correctness
Picking up a disk is counted as a
depended on their rationale and decimove, even if you put it back on the
sion for selecting one of the purchase
same peg or tower. In addition, the
options. The wording of the two
participants were instructed to indiword problems as presented to the
vidually move each of the four disks in
students is as follows:
the stack in the fewest moves possible
and as quickly as possible.
Sbirt sale problem: Terry was out
The student observer was then
shoppinghe wants to buy one or
brought into the room and joined the
more shirts. He saw a sign in one store
participant at the table. Each particithat stated BUY 2 SHIRTS, THIRD
pant demonstrated his or her underONE IS FREE, The cost for each shirt is
standing of the Tower of Hanoi by
$14.60. He went to another store and
explaining the rules and goal to his or
saw a sign that said 30% DISCOUNT
her respective student observer. Upon
FOR A $14.75 SHIRT. Which situation
completing the explanation, each paris cheaper to buy? State your
ticipant proceeded to solve the puzzle.
reason(s).
The explanation and moves of each
participant were timed and videotaped
Grass seed sale problem: Cindy
for documentation and subsequent
needs to buy grass seed for her large
analysis. At the end of the task, the
backyard. She needs about 200 lbs. of
experimenter asked the participant
grass seed. When she went to the
what strategies he or she had used and

or done the Tower of Hanoi puzzle

before. None responded in the affirmative. The experimenter explained
the puzzle's solution goal and rules to
each participant individually, without
the student observer present. The explanation was conducted in sign language and included showing the participant the actual puzzle tower and
visual diagrams depicting the starting
and ending positions of the stack of
four disks. The participants were also
given typewritten instructions to read.
All participants were then allowed to
ask clarifying questions about the rules
before demonstrating that they understood the rules and the object of the
Tower of Hanoi. This clarifying exchange with the teacher was in whatever mode of communication the student initiated.

VOLUME 143. N o . 4, 1998

store, the sign by the grass seed gave

her a number of choices at both sale
prices and regular prices.
SALE ITEMS
Buy four 60 lb. bags for $18.00 each
Buy three 90 lb, bags for $22,50 each
Buy two 120 lb. bags for $32.50 each
REGULAR PRICE
50 lb. bag each $18.00
20 lb. bag each $8.00
100 lb. bag each $35.75
If you were Cindy, what would be the
best choice for buying grass seed with
minimum waste? Explain your answer.
The goal and rules of the math
word problem were to: ( a ) consider
all options presented, ( b ) compute all
related math information, ( c ) make a
decision on which sale option to select, and ( d ) explain and justify the
decision. After the participant initially
explained the word problem to the
observer and solved the word problem
to his or her satisfaction, the experimenter asked the participant what
strategies he or she had used and to
explain them to the observer. Again,
the participant and observer were
separated and each was asked to write
a summary of how the problem was
solved.
Results
Reading levels did not significantly influence the students' performance in
terms of average time and mean number of moves to solve the Tower of
Hanoi puzzle. For number of moves
to solution, the five students in the
higher reading range averaged 29.4,
compared to 34.2 for the students in
the lower reading levels, t= -.943, df=
8,p> ,05- The students in the higher
reading group averaged 2.9 minutes to
complete the puzzle, compared to 3.2
minutes, t - -.418, df=8,p> .05, for
the lower group.
However, several aspects of the
student performance results in solving
both the Tower of Hanoi puzzle and
the math word problem were indeed
associated with reading levels. For the

AMERICAN ANNALS OF THE DEAF

PROBLEM-SOLVING STRATEGIES FOR TEACHING MATHEMATICS TO DEAF STUDENTS

Tower of Hanoi, when it came to writing the strategies used to move the
disks to solve the puzzle, the five students who had assessed reading
scores in the range between grade levels 10.3 and 11.1 were better at articulating their strategies than the five students with reading scores between
grade levels 6.9 and 9.7. Table 1
shows both the frequency with which

the students in each group articulated

the strategies and the mean performance for each group. Mean performance was calculated on the basis of
the following scoring of their strategies: "no strategies articulated"
equaled 0, "one strategy articulated"
equaled 1, and "both strategies articulated" equaled 2. A / test showed a
statistically significant difference be-

Tabla 1
Performance of Students in Articulating Their Strategies for Solving the Tower
of Hanoi Problem

No strategies Articulated
articulated
alternating
pegstrategy

Articulated
third peg
strategy

Mpei^
foimance

Higher-range
reading ievei
(10.3-11.1)
n=5

1.8

Lower-range
reading ievei
(6.9-9.7)

0.8

Table 2
Students' Average Performance Scores for Explanations in Sign Language and
Written Form (Standard Error in Parentheses)

Tower off Hanoi

Written
Signed
explanation explanation

Math word problem

Written
Signed
explanation explanation

Higher-range
reading level
(10.3-11.1)

49.8
(8.9)

58.2
(6.3)

72.0
(4.9)

78.0
(5.8)

Lower-range
reading ievei
(6.9-9.7)
n=5

33.4
(7.4)

63.0
(5.6)

58.0
(5.8)

52.0
(10.2)

VOLUME 143, No. 4, 1998

tween the mean performance of the

five students with higher reading levels for describing their strategies (Af =
1.8) and the performance of to the five
students in the lower reading level
range iM = 0.8), t = 2.357, df= S,p<
.05.
Furthermore, for both the Tower of
Hanoi and the math word problem,
the students were asked to provide
explanations in both sign language
and written form to demonstrate their
understanding of the problems. For
the Tower of Hanoi, the students were
asked to sign and then write what the
goal and rules were for solving the
puzzle. For the math word problem,
the students were asked to sign and
then write out the goal of the problem,
the information and data that they
considered in solving the problem,
and a statement of their solution. The
students' signed (on videotape) and
written explanations were scored for
the amount of correct information they
included relative to the total information that could and should have been
included. Table 2 shows the average
score and standard error per group for
their explanations in sign language
and written form for both the Tower of
Hanoi puzzle and the math word
problem.
A 2 X 4 ANOVA with repeated
measures was performed (two groups
of participants times four explanations
for each student). The results showed
a significant difference in performance
(F= 6.83, df= 1/8, p = < .05) between
the group of students with higher assessed reading levels (overall average
score 64.5) and the students in the
lower reading level range (overall average score 51.6). Furthermore, analysis showed a statistically significant
difference in performance between
the two groups across the four explanations (F= 4.897, df=5/24, p < .01).
For both the Tower of Hanoi and the
math word problem, the students with
the higher reading levels scored higher
on their written explanations than the
students with the lower reading levels.
However, when it came to the signed

AMERICAN ANIMALS OF THE DEAF

explanation for the visual/manipulative Tower of Hanoi, the students with

the lower reading levels performed
just as well as the students with higher
reading levels, and almost doubled
their own score compared to their
written explanation for the same problem. Interestingly, this pattern did not
occur with the math word problem, on
which the students in the lower reading range received roughly equal
scores for both their signed explanation (M = 58) and their written explanation (Af = 52). In contrast, the students with higher reading levels
scored consistently higher than the
students in the lower range on both of
their explanations for the math word
problem (written M = 72, signed M =
78).
The results seem reasonable in that
the students in the higher reading
group should, logically, better comprehend and be more comfortable
with the language of the math word
problem, and, thus, would be able to
explain more clearly in either sign language or written form their understanding of, and solution to, the problem. By contrast, the students with the
lower reading levels had to start with
the language of the math word problem. Even though it was also introduced to them in sign language by the
teacher, they had to rely on the printed
copy as they solved it, which apparently influenced their understanding
and explanation of the problem in
both sign language and written form.
However, the visual/manipulative
Tower of Hanoi did not depend primarily on language, except for the
rules, which were presented in both
sign language and written form, and
the students could develop a visual
understanding of the disk moves
needed to solve the puzzle. Here
again, the students with lower reading
levels had difficulty with their written
explanation of the problem goal and
rules, as evidenced by their scores (Af
" 33.4), but their mean signed explanation score, 630, was equal to that
the students with higher-assessed

VoLUME 143, N o . 4, 1998

reading abilities (M = 58.2).

Finally, with regard to the students'
primary language issue (i.e., whether
they primarily used ASL or signed English), no differences were observed in
either their patterns of performance for
solving the different problems or in
their explanations.
Strategy 2: Visualization of Moves
Prior to Attempts to Solve the
Puzzle
The second problem-solving strategy
examined involved only the Tower of
Hanoi puzzle. Based on the participants' performance in the first stage of
the present study, we decided to introduce a visualization strategy to see if it
would affect the students' performance with regard to number of
moves to goal solution. This strategy
called for the students in the treatment
group to study the Tower of Hanoi
and visualize their moves without
touching the disks or tower for at least
2 minutes, and up to 3 minutes maximum, before beginning their first
move. The reason for introducing a
visualization strategy was that during
the investigation of the peer observer
with explanation strategy, while a
number of students could clearly recite
the rules and goal solution of the
Tower of Hanoi, such recitation did
not necessarily translate directly into
thoughtful strategies during disk
movement. Often, some participants
seemed to fixate on the rules, and appeared pressured by the time and
speed considerations while moving
the individual disks from tower to
tower. They would appear to lack a
thoughtful plan or strategy to use the
rules to arrive at the goal solution with
the fewest moves possible and as
quickly as possible.
Participants
Sixteen college students took part in
the second study of problem-solving
strategies and techniques. For the
eight students randomly assigned to
the visualization group, the average
age was 21.6 years of age; for the eight

students randomly assigned to the

control group, the average age was
21.1 years. Gender makeup was similar, with one female in the visualization group and two females in the
nonvisualization control group. Only
one student who was assigned to the
visualization group had deaf parents.
All of the students used sign language.
The average reading grade level of
these students as assessed by the California Reading Test was 9.6 for the
eight students assigned to the visualization mode and 9.3 for the eight students assigned to the nonvisualization
control group.
Procedure
The students participating in the second stage of the study were randomly
assigned to one of two conditions: ( a )
The visualization strategy group was
given the instmctions and then asked to
study the Tower of Hanoi and visualize
their moves without touching the disks
or tower for at least 2 minutes, and up
to 3 minutes maximum before beginning their first move; ( b ) the
nonvisualization group was given the
instructions and then allowed to proceed immediately with solving the
Hanoi Tower. As was the case with the
investigation of the peer observer with
explanation strategy (except that no
peer observer was employed), as part
of the instructions each student was
given the opportunity to review the
rules and goals with the instructor, and
demonstrate understanding by restating
them in both sign language and written
form. The intent of the visualization
strategy was to force the participants to
think about their moves in relation to
the rules and to visualize the results of
their moves before actually making
any. With thoughtful visualizing, it was
hoped that students would be able to
perform more efficiently with a goaloriented movement strategy to successfully arrive at the goal, and not be pressured into fixating on either speed or
the rule for placing only smaller disks
on larger disks. Again, consistent with
the debriefing protocol used in the in-

AMERICAN ANNALS OF THE DEAF

PROBLEM SOLVING STRATEGIES FOR TEACHING MATHEMATICS TO DEAF STUDEJVTS

vestigation of the peer observer with

explanation strategy, when each participant completed solution of the
Tower of Hanoi puzzle, the instructor
asked the student to explain what strategies he or she had used to solve the
problem. (Note that no peer observer
was used).
Results

Two participating students were not

included in the analysis of the students
in
the
visualization
and
nonvisualization groups with the
Tower of Hanoi puzzle. One student in
the nonvisualization control group was
unable to complete the puzzle and had
gone past 9 minutes in time and made
224 disk moves. The minimum number of moves to solution for a fourdisk tower is 15, In a small group of
eight participants, this student's performance was so radically different from
the other students' that it would have
drastically altered the means and misrepresented the average performance
of the other students. To balance out
the groups for the analysis, we also
removed the student with the highest
number of moves in the visualization
group. With these two outlying students removed, there were now two
groups of seven students who were
included in the analysis.

In Table 3, we present the average

number of moves and times per visualization and nonvisualization group
for solving the Tower of Hanoi puzzle.
As can be seen from the data in Table
3, the visualization group performed
more efficiently in terms of fewer
movements to solve the problem; this
difference was statistically significant, t
= -2,315, df-U,p<.05.
Although the average times of the
two groups might also suggest that the
visualization group performed more
efficiently timewise, this difference
was not statistically significant, t = 1,381, iif= 12,p> .05, and, thus, cannot be interpreted as a real difference
in terms of time to completion. While
it would have been nice to have the
dimension of time efficiency be influenced by the treatment effect of visualization, in reality the primary goal of
the visualization was to reduce the
number of impulsive and meaningless
moves, which is what it appears to
have accomplished.
One final observation is that even
though deaf college students might be
able to articulate (in either sign or written form) the goal and rules of a problem such as the Tower of Hanoi, that
does not automatically mean they will
clearly apply the rules when solving
the problem. What they articulated

Table 3
Average Number of Moves and Completion Times, Tower of Hanoi Puzzle,
per Group

Average number of
moves

Average oompletlon time

(min.)

Visualization
group
(n=7)

19.9

1.3

Nonvisualization
group

35,1

2,3

VOLUME 143, N o . 4, 1998

and how they behaved were often

very different. It seemed that their
actual problem-solving behavior for
the Tower of Hanoi was often more
impulsive than reflective when the
guiding rules they had been given and
had articulated were foilowed.
Initially, we thought that perhaps
the deaf students' difficulty with problem-solving activities was partly related to their misunderstanding of
some or all of the rules, guidelines,
goals, and information pertinent to
specific problem situations. An erroneous assumption was that if the participants could clearly demonstrate
understanding of the procedural rules
and goals of the Tower of Hanoi problem in either sign language or written
form, then it would be more likely that
they could apply in practice the ruies
or guidelines, consider the relevant
information, and pursue the goal to
successful solution.
While all of the students could
clearly state and write the procedural
rules and goals of this visual/tactile
problem, numerous students struggled
to apply their stated knowledge in
practice. Even though they had just
stated and written the rules that one
can "never place a larger disk on top
of a smaller disk," and "cannot hold a
disk while moving another disk," a
number of participants would then
proceed to attempt to do so in the first
few moves until the experimenter reminded them of the rules. Furthermore, a number of students would fixate on the disk placement rule and on
the amount of time they were taking
while ignoring the problem goal.
Only after they had quickly made 30
or so moves without visible progress
toward the goal solution did they
pause and rethink their approach.
While the student who made 224 rapid
moves was an extreme case, such fixation behavior was exhibited by a number of the students. Also, at the end of
the problem, several participants could
only repeat the rules, and could not
describe any thoughtful strategy that
they had used. (Note: the required

AMERICAN AT^NALS OF THE DEAF

strategy is to move the top three disks

to the third, nongoal peg so that the
largest disk can be moved to the goal
peg followed by the top three disks.)
Similarly, there were several individuals who intuitively solved the Tower of
Hanoi problem in the minimum 15
moves, or as few as l6 moves, yet
were unable to give any explanation
either in sign or written form of how
they did it they could only indicate
that they "just saw how to do it in [my]
head." Obviously, describing an intuitive process is difficult.
Strategy 3: The Teacher Models
the Process for Solving a Sample
Problem
A third problem-solving strategy involved the teacher modeling the
analysis and solution process for a
sample word problem, which would
serve as a procedural model that the
students could then apply when solving similar word problems. The participants were first-year and secondyear students enrolled in one of two
NTID math classes taught by the lead
author. All of the students were given
an initial math word problem, which
served as a screening step for grouping of the students into training and
nontraining conditions. The training
condition was a full class session in
which the teacher modeled, step by
step, the analysis process for a sample
problem. The students were randomly
assigned to either the training or the
nontraining group in such a manner
that there were approximately equal
numbers of students in both groups
who had difficulty in analyzing and
considering all the relevant problem
data, along with those who were more
thorough in analyzing and solving the
initial math word problem.
Participants
There were 20 participating college
students in the implementation of the
third problem-solving strategy. For
the 10 students randomly assigned to
the training group in which the
teacher modeled the problem-solving

VOLUME 143,

No.

4,

1998

process, the average age was 22.3

years; for the 10 students randomly
assigned to the control group, the average age was 24.5 years. The gender
makeup of the treatment group receiving training was two females and eight
males, while there was one female and
nine males in the nontraining control
group. None of the students had deaf
parents. All of the students used sign
language. The average reading grade
level of these students as assessed by
the California Reading Test was 9.9 for
the 10 students assigned to the training
group and 10.4 for the 10 assigned to
the nontraining control group.

Bob and Jean want to visit their college friends in Cleveland, OH, for a
weekend. They live in Philadelphia,
PA. They both work 9 a.m. to 5
p,m. from Monday to Friday. Bob
tends to drive 55 mph and Jean tends
to drive 65 mph. The distance from
Philadelphia to Cleveland is about 425
miles. If you were Bob and Jean,
how would you plan your trip? Your
plan should include time of leaving,
approximate arrival time, who is driving and how long for each, as well as
time for other things that you need to
take care of, like stopping for gas and
food, etc.

Procedure
The training group participated in a
class session for problem solving in
which the teacher modeled the analytical process for a sample problem.
The instructor demonstrated and explained the solution to a math word
problem as an example for the students in the training group by walking
them through the problem step by
step and identifying, calculating, and
considering all relevant information
presented in the problem and explaining the answer in the context of the
problem. These activities were followed by questions and discussion.
The instructor communicated his
"thinking out loud" to the students by
signing and speaking his thoughts as
he analyzed, solved, and explained
the problem. The instructor did not
withhold any information from the students in the training session and took
the time to include all relevant details
of the analytical modeling process.

Wordprobletn 2: Jane and Jean have

only 18 hours to drive 1,000 miles.
Jane drives 70 mph most of the time,
while Jean generally drives only about
55 mph. Howmuch time should
each one drive to complete the trip on
time, including two restaurant stops
and two gas stops? What would also
be the estimated costs? Explain your
answer in detail.

The 10 students in the nontraining

group participated in the regular class
sessions, but received no specific
guidance on problem solving. They
were asked to solve the three math
word problems as part of the course
activities.
The three math word problems as
presented to the students in this third
investigation were:
Wordprobletn 1 for initial screening:

Word problem 3: Frank owns a small

air delivery business in Seattle. He has
to make three deliveries in his small
airplane. The first delivery stop is 50
miles away; the second stop is 75
miles further; and the third stop is
another 100 miles. The return home
from this third stop is 81 miles. His
small airplane engine flies 10 miles to
the gallon and fuel costs $2 per gallon.
Each landing and takeoff uses an extra
2 gallons of fuel. What is the total
round-trip mileage for Frank on this
delivery run? How many gallons of
fuel will Frank use from the initial
takeoff at his home base to his return?
How much will be his fuel costs? Explain your answers in detail.
These word problems were administered to the students in a group classroom situation as part of their mathematics class period, and they
independently solved each problem in
writing. No discussion among the students was allowed; nor was there a

AMERICAN ANNALS OF THE DEAF

PROBLEM-SOLVING STRATEGIES FOR TEACHING MATHEMATICS TO DEAF STUDENTS

Table 4
Observed Frequencies of Students Who Successfully Considered All Available Information and Explained Their
Answers to the Three Word Problems

Wom

WonI

WonI

problem 1
Analyzed
Explained
all
answer
Information

problem 2
Analyzed
Explained
all
answer
infomiation

problem 3
Analyzed
Explained
all
answer
Infomiation

Experimental
group,
training
(n=10)

Control
group, no
training
(n=10)

debriefing session when they completed each problem.

Results
The results of this instructional strategy
demonstrated that teacher modeling in
detail of the complete problem-solving
process (i.e., identifying and considering all available information, calculating, and explaining the answer) influenced the students' performance on
subsequent similar math word problems. Table 4 provides the frequencies of students for both the training
group and nontraining control group
who successfully identified and considered all available information and
then fully explained their answers.
Note that problem 1 was the initial
screening problem used to group the
students, and problems 2 and 3 were
administered to both groups after the
training group experienced the
teacher's modeling of the problemsolving process.
Analyses using 2 X 2 contingency
tables were conducted for each of the
problems. For word problem 1, which
was used to assign students to each
group, there were no statistically significant differences between the group

of students who received training/

modeling and the nontraining group
for identifying and considering all
available data ix2 = .83, p > .05) and
for fully explaining their answers (,x2 =
,39, p > .05). However, for word problems 2 and 3, there was a clear increase in the number of students assigned to the training group (teacher
modeling) who both analyzed "all information" and more fully "explained"
their answer (see Table 4). For word
problem 2, there were significantly
more students in the training group
than in the nontraining group who
identified and considered all available
data ix2 = 5.05, p < .05) and who ftilly
explained their answers (x2 = 5.495, p
< .05). This was also true for the students' explanation of their answers to
problem 3 (x2 = 5.495, p < 05).
As for identifying and considering
all available information, the trend was
in the right direction for the training
group (increased to seven students, as
compared to three on the screening
tejt). However, there was not a significant difference when compared to the
control group because, for whatever
reason, 50% of the control group also
considered all the available information

in problem 3 (.x2 = .83, p > .05).

Overall, the results show that
the teacher's modeling of the problem-solving process influenced the
performance of the students in
identifying and considering all
available information and more
fully explaining their answers with
the subsequent math word problems. In contrast, the control group
of students did not show a similar
improvement pattern.
Discussion and
Recommendations

The findings of our research with

deaf college students in the context
of their undergraduate math
courses (from Foundations of Algebra up to Concepts of Calculus)
demonstrate that the problem-solving performance of deaf college
students can be positively influenced with instructional strategies.
While the results show that students' reading levels can have an
influence on their explanations of a
problem and its solution, the data
also show that deaf college students can improve their problemsolving analyses and explanations

IEJI
VOLUME 143, N o . 4, 1998

AMERICAN ANNALS OF THE DEAF

by applying a procedural model demonstrated by the teacher. Strategies

designed to get the students to think
more carefully before trying to solve a
problem are also beneficial. The
simple technique of asking the students to visualize their movement of
the disks prior to solving the Tower of
Hanoi puzzle was remarkably effective
in reducing extraneous moves, thus
improving the efficiency of their disk
movement to achieve successful
completion of the task.
The real puzzle is why some students who have reached college level
continue to have difficulties analyzing
and solving what appear to be relatively straightforward problems. While
reading ability may partially contribute
to the difficulty some deaf students
have with understanding and dealing
with word problems, it should not prevent them from developing a useful
repertoire of analytical strategies. Like
any other skill, problem-solving abilities can be improved with relevant and
sufficient experiences that contribute
to analytical and critical thinking.
While the problem-solving strategies in the present study were examined with deaf and hard of hearing
college students, such strategies seem
applicable to high school deaf students who are college bound because
such students should be capable of
reflective and analytical thinking. For
younger deaf and hard of hearing students, further research would be
needed to determine the benefits of
such instructional strategies to their
problem-solving performance.
In developing and providing problem-solving experiences to help deaf
students develop and clarify their own
thinking skills, teachers should consider using the strategy of an interactive peer observer with whom the
problem solver can explain and clarify
his or her understanding of the problem in sign language. In some ways,
this serves a purpose similar to that of
the visualization strategy, Not only
does this force the individual student
to think through the problem as he or

VoLUME 143, No, 4, 1998

she attempts to explain it prior to attempting to solve it, but the interactive
nature of having to respond to the
questions of the observer when something is not clearly explained provides
an external check of the student's understanding. A related strategy is to
follow up the peer interaction with a
requirement that the student put in
writing his or her understanding and
explanation of the problem. Both the
interactive peer observer situation and
the written explanation provide opportunities for individual students to
clarify and rethink their understanding
of the problem situation.
When appropriate to the problem
task, a visualization technique, or similar strategies to encourage "thinking
before doing," can also be used. The
goal of such strategies is to inhibit students' tendency to jump into problem
solving without first giving the process
some thought. This strategy was particularly effective with the visual/manipulative puzzle used in the present
study. The focus should be on either
thinking of a solution strategy or considering alternative solution strategies
before actually trying them.
When demonstrating procedural
models for problem solving, teachers
need to provide detailed, step-by-step
explanations to students in sign language, spoken, and written form. In
the process of modeling a procedure,
one cannot assume that any of the
steps are familiar or intuitive to the students. In turn, students should be required to do the same and model their
problem-solving procedures with full
explanation to either the teacher or
their peers.
To provide variety in problemsolving activities, we offer these related suggestions for improving problem-solving skills:
Teachers (or knowledgeable, "advanced" students) can provide a
guided problem-solving experience
for individuals or small groups of students by supplying cues, suggested
steps, and questions at appropriate
times in the analytic process.

Either prior to or as part of solving

a problem, the teacher should involve
students in an interactive/feedback
situation (i.e., pairing with other students, aides, or teachers) in which
they are required to explain sample
problems and answer questions,
The teacher should have the students think about, explain, analyze,
and summarize problem-solving tasks
with sign language, verbalization, written text, and role-playing,
The teacher should have students
"create" or construct different problem
situations similar to the problem
task at hand to demonstrate their understanding of that type,
The teacher should have students
conceptualize (or visualize) several
different approaches to sample
problem-solving tasks prior to solving
an actual problem.
These suggestions are intended to
help students develop greater flexibility through varied practice with problem solving. The internalization and
application of new knowledge and
skills is enhanced by repetitive practice, active participation, interactive
discussion, and evaluative feedback.

Rfr*ncs
Altshuler, K. Z., Deming, W. E., Vollenweider, J.,
Rainer, J. D., & Tendler, R, (1976). Impulsivity and profound early deafness; A cross-cul-

tural inquiry. American Annals of tbe

Deaf, 121, 331-345.
Barnham, J., & Bishop, A. (1991), Mathematics
and the deaf child. In K, Durkin & B. Shire

(Eds.), Language in malbematical education: Researcb and practice. Philadelphia;

Open University Press.
Campbell. D. S, (1986. December). Pause for
reflection: Teaching tbe impulsive student reflective problem-sdving with self-instruction
training. Paper presented at the annual
meeting of the Ontario Educational Research
Association. Toronto, Canada.
Campbell, D. S. (1989). The effects of general
strategy instruction on problem-solving
among impulsive, hearing impaired adolescents. Canadian foumal of Special Education, 5, 159-167.
Eabon, M, F, (1984, May),

On tbe relaHonsbip

between impulsivity andfield-dependence in

hearing-impaired children Paper presented
at the annual meeting of the Midwestern Psychological Association, Chicago,
Feuerstein, R, (1980), Instrumental enrichment.
Baltimore; University Park Press.
Freeman, R. D, (1979)- Psychosocial problems
associated with childhood hearing impairment. In L. Bradford & W. Harding (Eds.),
Hearing and hearing impairment (pp, 405-

AMERICAN AMNALS OF THE DEAF

415). New York: Grune & Stratton,

Furth, H. (1973). Deafness and learning: A psychosocial approach. Belmont. CA:
Wadsworth,
Haywood, H. C, Towery-Woolsey, J., ArbitmanSmith. R., & Aldridge, A. H, (1988). Cognitive education with deaf adolescents: Effects
of instrumental enrichment. Topics in Language Disorders, 8, 2340,
Kagan, J, (1966). Refiection-impulsivity: The
generality and dynamics of conceptual
tempo, pumal ofAbnormal Psychology, 71,
17-24.
Kagan, J., Moss, H,, & Siegal. I, (1963), Psychological significance of style of
conceptualization. Monographs of the Society for Research in Cbild Development, 28
(Serial No. 86).
Keane. K. J,, & Kretschmer, R. E. (1987). Effect
of the mediated learning intervention on cognitive task performance with a deaf population. Journal of Educational Psychology,
79, 49-53.
Loera, P. A.. & Meichenbaum, D. (1993), The
"potential" contributions of cognitive behavior modification to literacy training for deaf
students. American Annals of tbe Deaf,
138, 87-95,
Luckner, J. (1992), Problem solving: A comparison of hearing-impaired and hearing individuals, foumal of tbe American Deafness
and RehabilitaHon Association, 25, 21-27.
Luckner, J. L. & McNeUl. J, H. (1994), Performance of a group of deaf and hard-of-hearing students and a comparison group of hearing students on a series of problem-solving
tasks. American Annals of the Deaf, 139,371377.
Martin. D. S. (1984), Cognitive modification for
the hearing impaired adolescent. Exceptional Children, 51, 235-242.
Martin, D. S. (1993). Reasoning skills; A key to
literacy for deaf learners. American Annals
of tbe Deaf. 138,82-86.
Martin, D. S,, & Jonas, B. (1986). Cognitive modifiahility in the deaf adolescent. Ann Arbor,
MI; ERIC Clearinghouse on Handicapped
Children. (ERIC Document Reproduction Service No. ED 276 159)
Meadow, K, P.. & Schlesinger, H. S. (1971). The
prevalence of behavioral problems in a
population of deaf children. American Annals of the Deaf, 116, 346-348.
Meadow, K. P.. & Trybus, R.J, (1979), Behavioral and emotional problems of deaf children: An overview. In L. Bradford & W.
Harding (Eds.). Hearing and bearing impairment ipp. 395-405X New York: Grune &
Stratton,
Metz, D. E.. Caccamise, F,, & Gustafson. M. S.
(1997), Criterion validity of the language
Background Questionnaire; A self-assessment instrument, foumal of Communication
Disorders, 30, 23-32.
Mousley. K. (1991), Encouraging critical thinking in a mathematics classroom using modified teaching strategies. Profession on parade: Proceedings of the Biennial Meeting,
Convention of American Instructors of tbe
Deaf and Administrators Serving tbe Deaf.
55, 284-286.
Pau. C. S. (1995). The deaf child and solving
problems of arithmetic: The importance of
compreherisive reading. American Annals of
the Deaf, 140, 287-294.
Piaget. J. (1976), The grasp of consciousness:
Action and concept in Ibe young child.
Cambridge, MA: Harvard University Press.
Pressley, M., Snyder, B. L., & Cariglia-BuU. T.
(1987), How can good strategy use be

VOLUME 143. N o . 4,

1998

taught to childrea'; Evaluation of six alternative approaches. In S. M. Cormier & J, D.

Hagman iEiis.'), Thtnsfer of leaming: Contemporary research and applications (pp.
81-121). New York: Academic Press.
Rohr-Redding. C. (1985), Can thinking skills be
incorporated into a curriculum.' In D, S, Martin (Ed), Cognition, education, and deafness: Directionsfor researcb and instruction
(pp. 168-171). Washington. DC: Gallaudet
College Press.
Tzuriel, D.,& Caspi, N. (1992). Cognitive modifiability and cognitive performance of deaf
and hearing preschool children, foumal of
Special Education, 26, 235-252.
Woditsch. G, A. (1991). The thoughtful teacher's
guide to tbinking skills. Hillsdale, Nj;
Eribaum.

AMERICAN ANNALS OF THE DEAF