E X
EDUCATIONAL
GREGORC
LEARNING
ACHIEVEMENT
IN
P. H. Harasym,
Oflce
of Medical
Be
Education
University
E
R
I
STYLES
ANATOMY
E. J. Leong,
P
AND
G. E. Lucier,
M
E
N
T
S
AND
PHYSIOLOGY
and F. L. Lorscheider
and Department
of Medical
of Calgary,
Calgary,
Alberta
Physiology,
T2N 4N1,
Faculty
Canada
of Medicine,
esults from the Gregorc Style Delineator
(GSD), administered
to 260 undergraduate nursing students, were compared with achievement
scores in a
human anatomy and physiology course. Factor analysis and VARIMAX rotation
demonstrate
that there is no relationship
between any of the four learning styles
allegedly identified by the GSD and achievement in anatomy and physiology. Factor
analysis also shows that the GSD measures only a single bipolar scale of sequential vs.
random ability rather than two bipolar scales comprising
four learning styles, as
suggested by Gregorc. These findings question the validity of the GSD and recommend discontinuing
its use as an indicator of learning styles.
R
AM. J. PHYSIOL.
Key words:
268 (HIV.
PHKSIOL.
educational
EDUC.
13): S56460,
measurement;
psychometric
A decline in resources
for postsecondary
education
has motivated
educational
institutions
to develop
more efficient ways to educate their students.
The
manner
in which
students
learn has come under
scrutiny by researchers.
Kimble (3) defined learning
as a relatively
permanent
change in performance
by
an individual,
arising from experience.
The degree
of performance
change
may be measured
using
some form of assessment
test. On the other hand,
individual
learning
preference
is influenced
by learning style (5). Several psychometric
instruments
have
been developed
to identify an individual’s
preferred
learning
style. One such example
of this type of
instrument
is the Gregorc Style Delineator
(GSD).
- 4046
VOLUME
/ 95 - $3.00
13 : NUMBER
- COPYRIGHT
1 -ADVANCES
analysis
Perceptual
ability
deals with
how
an individual
receives information.
Gregorc (1) claims that perception can be described
on a continuum
as abstract vs.
concrete.
Abstract ability enables information
interpretation
by use of reasoning
and involves intuition
to deal with subjective
concepts
and feelings.
A
person using abstract reasoning
can perceive information that is invisible to the physical senses. On the
other hand, concrete
ability enables an individual
to
interpret
information
that is gathered
by the senses,
such as visual, auditory,
and tactile stimuli,
and to
apply this information
to the physical world.
Ordering
ability deals with how a person arranges
and uses information.
Gregorc (1) claims this ability
is described
on a continuum
as sequential
vs.
random.
Sequencing
is used to organize
data in a
linear fashion, which
enables precise, progressive,
and logical communication.
Random
processing
of
information
occurs in a netlike fashion with a myriad
of interrelated
information.
Random processing
enables an individual
to deal with diverse pieces of
information
simultaneously.
Gregorc
(1) developed
the GSD as a psychometric
instrument
to determine
an individual’s
“mind”
style. This instrument
is based on mediation
abilities
theory, which states that the human mind contains
channels through which information
is received and
expressed
in an efficient
and effective
manner.
There are two types of mediation
abilities,
perception and ordering.
1043
1995.
o 1995
THE AMERICAN
IN PHYSIOLOGY
S56
PHYSIOLOGICAL
EDUCATION
-JUNE
SOCIETY
1995
E
EDUCATIONAL
METHODS
The GSD was administered
to 283 nursing
students
enrolled
in Medical
Science (MDSC) 200, an introductory
first-year
course in human
anatomy
and
physiology
offered
annually
by The University
of
Calgary, Faculty of Medicine
to students enrolled
in
the Faculty of Nursing.
The course is divided
into
two semesters,
with a total of 150 hours of instruction. The course topics within
each semester differ
and are organized
according
to different
body systems. Instruction
is provided
using a multidisciplinary team teaching approach.
Lectures are supple-
: NUMBER
I - ADVANCES
E
R
I
M
E
N
T
S
Results from the GSD were analyzed
to determine
the relationship
between
the learning
styles, identified by the GSD, and performance
on MDSC 200
examinations.
Statistical
analyses were conducted
using the Statistical
Package for the Social Sciences
(SPSS) for Windows
Version
6.0. These results were
first summarized
as descriptive
statistics by dominant learning
style (i.e., means t SE). The dominant learning
style for the group
of students
was
calculated
by averaging
the four GSD scores generated for each student.
Data from the GSD and
examination
scores for each student
were
then
analyzed
by a principal
component
factor analysis
and VARIMAX
rotation
(6). This analysis was used to
reveal the potential
relationship
between
the GSD
and examination
scores and whether
the four distinct learning
styles (CS, AS, AR, and CR) can be
accurately
described
using two unidimensional
bipolar scales of perceptual
and ordering
abilities
(Concrete vs. Abstract
and Sequential
vs. Random).
Finally, the relationship
between
learning
styles and
achievement
in MDSC 200 was determined
using a
one-way analysis of variance (ANOVA).
The GSD ranks a series of matched
word associations in a word matrix and assigns a numerical
value
to each of the four learning
styles. The learning
style
with the largest number
assigned to it is identified
as
the dominant
style employed
by an individual.
Wells
and Higgs (7) employed
the GSD to determine
preferred
learning
styles for nursing students in the
first and fourth
years of their program
and concluded that the majority
of students were either AR
or CS. The main objectives
of the present study were
to determine
the dominant
learning
style in a class
of nursing
students
by use of the GSD and to
determine
whether
certain learning
styles allegedly
identified
by it can be associated
with enhanced
performance
on achievement
tests in a human
anatomy and physiology
course.
13
P
mented
with
structured
laboratory
sessions
that
provide
students
the opportunity
to view prepared
human
specimens
and to relate
the theoretical
aspects of the lecture material
to the human body.
Student
achievement
in MDSC 200 is assessed by
four examinations
(two midterms
and two end-ofterm examinations)
covering
lecture
content
and
two laboratory
examinations,
which are more anatomic in nature. The final grade is based on a single
score, calculated
as a weighted
combination
of the
raw scores from the six examinations.
Student examinations are prepared
by selecting
questions,
developed by the course instructors,
that are stored in a
computerized
test bank. During the academic year a
total of 23 students withdrew
from the course. Only
the 260 students completing
the entire course were
used to analyze the relationship
between
learning
style and achievement
in the course.
The GSD test booklet
developed
by Gregorc
(2) has
combined
the mediation
abilities
into four distinct
mediation
channels
or mind styles: abstract random
(AR), abstract
sequential
(AS), concrete
random
(CR), and concrete
sequential
(CS). These mind
styles are alleged to form two bipolar
scales consisting of CS vs. AR and AS vs. CR. Although
Gregorc
used the GSD to identify an individual’s
mind style,
several users of the instrument
associate mind style
with learning
styles (7). On the basis of Gregorc
categories,
individuals
who are CS learners
tend to
process
information
in a methodical,
instinctive,
deliberate
manner,
whereas
CR learners
tend to
process information
in an intuitive,
impulsive,
and
independent
manner.
However,
AS learners tend to
process information
in an intellectual,
analytic, logical, and correlative
manner,
whereas
AR learners
appear
to process
information
in an emotional,
perceptive,
and critical manner.
VOLUME
X
RESULTS
Table 1 summarizes
the results from the GSD administered to the nursing students.
The dominant
learning styles used by the students were identified
using
the mean values for each learning
style. The results
IN PHYSIOLOGY
s57
EDUCATION
- JUNE
1995
E X
EDUCATIONAL
TABLE
for 260 nursing
GSD class profile
Learning
1
students
Style
26.148
AS
CR
24.028
22.572
27.265
cs
+ 0.303
* 0.271
2 0.348
+ 0.339
Values are means + SE. GSD, Gregorc
Style Delineator;
MDSC,
Medical
Science; AR, abstract random;
AS, abstract
sequential;
CR, concrete
random;
CS, concrete sequential.
indicate
following
that the learning
styles can be ranked
order: CS > AR > AS > CR.
in the
analysis
of nursing
scores
TABLE 2
student
in MDSC
Variable
Lecture
0.862
0.904
0.863
0.902
3
2
0.015
- 0.007
0.103
0.028
Examination
0.854
0.771
L
Learning
AR
AS
CR
cs
Factor
Examination
2
Laboratory
1
0.048
-0.019
Style
- 0.093
0.058
0.045
- -0.012
VOLUME
13
: NUMBER
-0.745
0.802
-0.806
0.848
1 - ADVANCES
I
M
E
N
T
S
The [S] variable was divided
into intervals
on the
basis of Gregorc’s
breakdown
of usage levels of the
learning
styles into dominant,
intermediate,
and
low levels. Gregorc
(1) defined
dominant
styles to
have 27-40
points,
intermediate
styles to have
16-26 points, and low styles to be of lo-15
points.
To break up the [S] variable into similar intervals of
dominant,
intermediate,
and low it was necessary to
double
the values used in the ranges defined
by
Gregorc.
This resulted
in intervals
consisting
of
dominant
(54-80))
intermediate
(32-52))
and low
(20-30).
The formation
of the new [S] variable from
the data obtained
from the GSD scores of the
nursing
students
in MDSC 200 produced
values
ranging
from 29 to 69. The lack of values in the low
range of the [S] variable led to a modification
of the
ranges used to form the three new usage intervals:
dominant
(54~80),
intermediate
(4 l-53))
and low
(O-40)
which transferred
part of the range assigned
GSD and examination
200
Factor
R
The GSD results were modified
accordingly
to obtain numerical
range values for the variables Sequential vs. Random,
which
were divided
into three
levels of usage. The new variable of sequential [S]
was produced
by adding
the values of CS and AS
together.
The new variable
of random
[R] was
produced
by adding
the values
of CR and AR
together.
The resultant
variables,
[S] and [RI, are
linearly related
on a single unidimensional
bipolar
scale. This relationship
can be described
mathematicallyas [S] + [R] = 100. Thus, knowing
the value of
one of these two variables yields the value of the
other. Therefore
information
on one variable automatically provides information
on the other. Hence,
a single ANOVA on the [S] variable (see Table 4) was
used to determine
its relationship
to MDSC 200
achievement.
Table 2 shows the results from the principal
component factor analysis with VARIMAX
rotation.
This
analysis isolated two factors; factor one contains the
examination
scores and factor two the learning
styles identified
by the GSD. No significant
correlation was observed between
examination
scores and
any of the four learning
styles. All six examinations
correlated
highly
and positively
with factor one
(MDSC 200 achievement).
Likewise,
all four learning styles correlated
highly with factor two (learning
styles). However,
factor two indicates
that CS and
AS have positive whereas CR and AR have negative
correlation
coefficients.
This indicates
that one of
the two bipolar
scales reported
by Gregorc
(perceptual ability of Abstract
vs. Concrete)
has collapsed
into the other
bipolar
scale (ordering
ability
of
Sequential
vs. Random).
Factor
E
Gregorc
(1) divided
the numerical
values assigned
to each learning
style into three levels of usage:
dominant,
intermediate,
and low. ANOVA was used
to analyze the relationship
between
levels of usage
for a given learning
style (independent
variable)
and
overall achievement
in MDSC 200 (dependent
variable). This analysis
examined
only the ordering
ability of Sequential
vs. Random,
because
factor
analysis demonstrated
that the existence of perceptual ability (Abstract vs. Concrete)
is not supported
by the data.
MDSC200
GSD Results
AR
P
IN PHYSIOLOGY
S58
EDUCATION
- JUNE
1995
EDUCATIONAL
E X
3
TABLE
Sequential
ability
Sequential
Interval
Ability
and
corresponding
Mean
Low (O-40)
Intermediate
(41-53)
Dominant
(54-80)
31
117
112
Values for mean sequential
are means k SE.
final
course
36.871+0.538
47.78620.335
5922320.338
ability
Final
Course
and mean
Score
63.758k1.923
64.442 20.969
65.192k 1.064
final
course
score
from the lower end of the intermediate interval to
the low interval. Table 3 provides the mean scores
for sequential ability and corresponding mean (+
SE) final course scores for the three intervals of
sequential ability usage. The ANOVA results indicated that the three levels of sequential ability usage
(Table 4) are not related to overall final course
scores.
DISCUSSION
Factor analysis of the GSD and examination scores
produced unexpected results. This analysis demonstrated that there is no relationship between examination scores and learning styles allegedly identified
by the GSD. The most significant finding obtained
from the factor analysis was that the GSD data
collapsed from two bipolar scales into a single
Analysis
of variance
ability
200
to final
Between
groups
Within groups
Total
Degrees
of
Freedom
Sum
Squares
of
Mean
Squares
2
257
259
VOLUME
course
F
E
N
T
S
15 August
1994; accepted
in final
form
21 February
1995.
Probability
References
1.
Gregorc,
Gregorc
: NUMBER
M
Address for reprint
requests:
P. H. Harasym,
Office of Medical
Education,
Faculty of Medicine,
Univ. of Calgary Health Sciences
Center,
3330 Hospital
Dr. NW, Calgary,
Alberta
T2N 4N1,
Canada.
0.768
13
I
The authors thank Dr. R. A. Schulz, coordinator
of The University
of Calgary Teaching
Development
Office, for providing
funding
assistance through
a Royal Bank of Canada teaching development
grant.
Received
Source
R
In conclusion, this study found that the dominant
learning styles for nursing students in MDSC 200,
identified by the GSD, were CS and AR. More
importantly, this study found no relationship between GSD learning styles and achievement in
MDSC 200. In addition, this study demonstrated
that Gregorc’s assumption regarding the existence
of four distinct learning styles is not supported by
the GSD data. These results clearly indicate that the
GSD cannot be used to help predict student achievement on examinations in an introductory anatomy
and physiology course. The main recommendation
from this study is that, because the GSD does not
measure what it purports to measure (and thus lacks
validity), future use of the GSD is not encouraged.
The main objective of this study was to determine
whether the GSD could be used to predict achievement in a human anatomy and physiology course. A
positive association between learning styles and
achievement would assist instructors of anatomy
and physiology in developing more effective lessons
that could enhance student performance. Results in
Table 1 indicate that the dominant learning styles
employed by the nursing students who completed
the GSD are CS and AR, which is in agreement with
Wells and Higgs (7).
TABLE
4
of sequential
scores
in MDSC
E
bipolar scale. For example, factor two in Table 2
shows that perceptual ability (Abstract vs. Concrete)
does not exist as a separate factor but instead is part
of the unidimensional variable of ordering ability.
These results clearly demonstrate that Gregorc’s
assumption regarding the alleged existence of four
distinct learning styles, comprised of two bipolar
scales, is incorrect. In other words, the perceptual
ability (Abstract vs. Concrete) used by Gregorc does
not provide any additional information about the
learner than that already provided by ordering
ability (Sequential vs. Random). Another finding by
ANOVA was that the degree of sequential ability
usage was not related to achievement as measured
by final course scores. Although the results of the
present study bring into question the validity of the
GSD, other reports (4) have also questioned its
reliability as a suitable instrument for identifying
and evaluating learning styles.
scores
Mean
Sequential
Ability
P
I - ADVANCES
IN PHYSIOLOGY
s59
A.
F. An Adult’s
Associates,
EDUCATION
Guide
1982.
- JUNE
1995
to Style.
Columbia,
CT:
EDUCATIONAL
E X
2. Gregorc,
A. F. Inside
Styles: Beyond
the Basics. Columbia,
CT: Gregorc Associates,
1985.
3. Kimble,
G. A. Hilgand
and Marguis
Conditioning
and Learning (2nd ed.) NewYork,
NY: Appleton-Century-Crofts,
1961.
4. Sewall,
T. J. The measurement
of learning
styles: a critique of
four assessment tools. ReportsEvaluative/Feasibility.
Green
Bay, WI: Univ. Wisconsin-Green
Bay Assessment
Center, 1986.
(ERIC ED 267 247).
VOLUME
13
: NUMBER
I - ADVANCES
P
E
R
I
M
E
N
T
S
E. Hooker.
Learning
style preferences
of
practitioners
in a teaching education
pr0gram.J.
12: 49-55,
1983.
6. Weiss,
D. J. Factor
analysis
and counseling
research.
J.
Counsel. Psychol.
17: 477-485,
1970.
7. Wells,
D., and
2. R. Higgs.
Learning
styles and learning
preferences
of first and fourth semester baccalaureate
degree
nursing students.J.
Nursing
Ed. 29: 385-390,
1990.
5. Vittoe,
allied
Allied
IN PHYSIOLOGY
s60
M. C., and
health
Health
EDUCATION
- JUNE
1995