1'1?"

 

 

 

 

t‘wveu-AHH

  

 

 

~1qu

 

 

 

 

 

 

 

 

 

:9

ODS

  

 

fPh

 
 

 

 

eéso

     

1. . E Tum.

 

:N

 

e31)";

 

 

..0
VA
Dr”

ENJTssi

 

 
 

 
 

 

f

 

 
 

 
 

at

 

(NNNG

 

<e

 

 

_IVCQNI
MAT ‘

 

 

 
   
 

 

 

 

155,... E .. L: x:

 

 
 

 
 

 

 

E

 

 

 

 

 

NAHU

D0

{PAULf

1

 

 

 

 

(if;

 

‘ <

 

 

 
 

 

 

 

 
 

v.2? ..§o.._...::....

 

    

  

_. a: 7
{Lang .5

. v!
4 Q.

 

 

 

  

g; :14:—
EE. ..
2..., l .

          

 

   

1.4.,

1V3 I1:

 

      

Jar , 5w ,
. .27.: .
i

5;

  

 

 

 

: metmrriuxau,

my?” W $3“

Michirmz 3' €358
a Um. smity

l-h‘ifi'l ‘ ’1? 1* my wag-m

  

:5

35f '1‘- “I

This is to certify that the
thesis entitled

A COMPARISON OF THREE METHODS OF MATCHING
STUDENTS IN LEARNING CELLS

presented by

Larry Paul Donahue

has been accepted towards fulfillment
of the requirements for

Ph.D. degree 1n Secondary Education
and Curriculum

MV/M

MajU rpr essor

0-7639

 

 

ABSTRACT

A COMPARISON OF THREE METHODS OF MATCHING
STUDENTS IN LEARNING CELLS

BY

Larry Paul Donahue

The primary purpose of this study was to compare
three methods of matching partners in a learning cell.
The three methods were to pair students whose cognitive
style maps were highly similar, to let students pair
themselves and to pair students randomly. It was
intended that the achievement of students in the three
conditions be compared and that ratings of the learning
cell experience by students in the three kinds of learn—
ing cells be compared.

Secondary purposes of the study included explor-
ing the influence upon achievement and ratings of the
learning cell experience of matching partners who were
of the same or different sex and of matching partners
who were relatively the same or different in terms of
chronological age.

The three basic research questions were:

1. Do student pairs who are matched by cogni-

tive style similarity attain higher achievement and

Larry Paul Donahue

rate the learning cell experience more favorably than
those who are randomly matched or those who self select?

2. Do student pairs who are similar in chrono—
logical age differ in achievement and the rating of the
learning cell experience from students who have larger
age differences?

3. Do student pairs of the same sex differ in
achievement and the rating of the learning cell experi—
ence from students of different sexes?

Students in the introductory psychology course
at Oakland Community College comprised the population of
the study. The sample consisted of those students in
four sections of the course taught either by Dr. Mueller
or by Dr. Svagr during fall semester of 1974. A cogni—
tive style similarity group, a self select group and a
random group were formed in each of the four sections.
Fifty-two of the original sixty-one learning cells par—
ticipated in the study to its conclusion.

Each instructor had one day section that met
twice per week and one evening section that met once a
week. During the first meeting of each day section
there was a lecture for the first forty minutes followed
by a thirty-five minute learning cell session. The
second meeting began with a twenty-five minute class
discussion and review. Next there was a thirty minute

learning cell session and then there was a twenty minute

Larry Paul Donahue

period to respond to a weekly achievement instrument.
Each evening section had the same format as two consecu-
tive daily meetings combined. The study lasted four
weeks.

There was no required pattern of events that a
learning cell had to adopt. Three examples of learning
cell patterns were presented during the first learning
cell meeting, and the subjects were free to choose one
of them or to define a new pattern.

Each week each instructor distributed a study
guide for the following week's work. It included items
that were representative of the key points in a chapter.
The study guides were used throughout the term and they
became the core of the learning cell sessions.

The dependent variables of the study were

(1) weekly achievement, (2) achievement on a final

 

 

examination and (3) rating of the learning cell experi-

 

 

ence. The weekly achievement dependent variable was

 

analyzed via ANCOVA at the .05 alpha level with the mid-

term as the covariable. The final achievement and the

 

learning cell rating dependent variables were analyzed

 

via ANOVA at the .05 alpha level.

The main conclusions of the study follow.

1. The use of cognitive style similarity based
upon cognitive style mapping as a criterion for matching

students in learning cells was no better than the use of

 

Larry Paul Donahue

self selection or randomization in terms of achievement
and rating of the learning cell experience.

2. The closeness of the chronological ages of
cell mates did not affect achievement or rating of the
learning cell experience by the subjects.

3. The sameness or difference of the sex of
cell mates did not affect achievement or rating of the
learning cell experience by the subjects.

4. No combination of the three independent vari-
ables in the study (Matching Method, Age Difference, Sex
Factor) affected achievement or the rating of the learn—
ing cell experience by the subjects.

Four implications for the potential user of
learning cells resulting from this study are listed
below.

1. The findings support the idea that none
of the different ways of matching people influenced
achievement or learning cell ratings. This implies
that if one were going to use learning cells he could
randomly assign partners or let them self select with-
out regard to their ages or sex. The use of cognitive
style similarity to match people would not be recom-
mended because the random or self select methods require
much less time and effort.

2. Both instructors reported that the subjects

responded very favorably to having study guides

Larry Paul Donahue

available. Perhaps other learning cell participants
would also benefit from getting study guides from their
instructors.

3. Certain learning cells contained incom—
patible partners but the design of the study required
that the partners remain together during the study. It
is suggested that partners ought to be permitted to
change periodically if desired.

4. It may be desirable to use the learning
cell as an alternative rather than requiring every stu-

dent to participate.

 

A COMPARISON OF THREE METHODS OF MATCHING

STUDENTS IN LEARNING CELLS

BY

Larry Paul Donahue

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Secondary Education and Curriculum

1975

GD COpyright by

LARRY PAUL DONAHU E
l 9 7 5

 

 

DEDICATION

This thesis is dedicated
to my parents who

encouraged me to dream.

ii

 

ACKNOWLEDGMENTS

Dr. Allan J. Abedor, my advisor and guidance
committee chairman, has been a constant source of
encouragement and thoughtful direction from the initial
idea to this completed thesis. His willingness to
always do a little bit extra and his ready smile have
helped me many times during the past three years. Sin-
cere thanks are also expressed to the other members of
my guidance committee: to Dr. Paul W. F. Witt who will
always occupy a special place in my thoughts and who has
touched my life with an indelible hand, to Dr. Cas Gentry
who was so instrumental in my coming to Michigan State,
and to Dr. Norm Bell.

I wish to thank Dr. Joe Hill for the generous
way he shared his time and knowledge. Dr. Virginia Svagr
and Dr. Joe Mueller made this study possible through
their cooperation. Their contribution was considerable.

Dr. Paul Woodworth, Dr. Barry Bratton and
Dr. Bob Stone were fellow graduate students when this
thesis was but an idea. They helped immeasurably as I
began this study.

Sam Burnett was my principal during the three
years I spent as a seventh grade teacher in Toledo. His

iii

practical, forthright approach continues to influence

me. He inspired me to grow professionally and personally

during a most formative time in my career. I want him to

know how much I value the experiences I had at McGregor.
My warmest appreciation is offered to my wife

Pat and to my daughter Stephanie for the loving home

they helped to maintain often in my absence. I cannot

imagine being happier than I have been these past three

years. Thank you both for being you.

iv

 

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . .

LIST OF FIGURES . . . . . . . . . . . . . . . . .

Chapter

I.

II.

III.

BACKGROUND . . . . . . . . . . . . . . . .
Problem Statement . . . . . . . . . . .
Purpose of this Study . . . . . . . .
Assumptions . . . . . . . . . . . . . .
Delimitations of the Study . . . . . .
Research Questions . . . . . . . . . .
Research Hypotheses . . . . . . . . . .
Definitions of Important Terms . . . .
Overview of Other Chapters in the Study

REVIEW OF RELATED LITERATURE . . . . . .

What is the Learning Cell? . . . .

What is the Theoretical Support for
the Learning Cell? . . ... . . . .

What Advantages Does the Learning
Cell Offer? . . . . . . . . . . . .

How Have Researchers Matched Students
in Learning Cells? . . . . . . . . .

How Have Researchers Determined
Cognitive Style Similarity?

What is the Support for Using Cognitive
Style Similarity as a Criterion for
Matching Students in a Learning Cell?

Implications for this Study . . . .

METHODS AND PROCEDURES . . . . . . . . .

Design Over Time . . . . . . . . . . .
Population and Sample . . . . . . . .
Procedures . . . . . . . . . . . . .
Treatment . . . . . . . . . . . . . .
Instruments . . . . . . . . . . . . .

Design Over Variables . . . . . . . . .

Hypotheses . . . . . . . . . . . . . .

Page
vii

viii

f—J

OQWMwaN

ll
12
16
24
31

34
41

44

44
49
49
52
53
54
56

 

 

 

 

 

 

Chapter

IV. ANALYSIS OF DATA . . . . . . . . . . . . .

Hypotheses and Data Related to

Weekly Achievement . . . . . . .
Hypotheses and Data Related to

Final Achievement . . . . . . . . . .

Hypotheses and Data Related to Students'

Rating of the Learning Cell

Experience . . . . . . . . . . . . .
Interpretation of the Data Analysis . .
Summary . . . . . . . . . . . . . . . .

V. SUMMARY, CONCLUSIONS, IMPLICATIONS AND

RECOMMENDATIONS . . . . . . . . . . . .

Summary . . . . . . . . . . . . . . . .
Conclusions . . . . . . . . . . . . .
Conclusion
Conclusion
Conclusion
Conclusion .
Implications . . . . . . . . . . . . .
Recommendations . . . . . . . . . . . .
Recommendation
Recommendation
Recommendation
Recommendation

ewNN-N
o
o
o
o

ALONE-4

BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . .

APPENDICES

vi

Page

61

61

64

67
70
74

76

76
78
78
78
79
79
80
82
82
83
83
83

85

9O

 

 

LIST OF TABLES

Table Page
1. Relationship of Covariable to Achievement

Measures . . . . . . . . . . . . . . . . 46

2. ANCOVA for Weekly Achievement . . . . . . . 64

3. ANOVA for Final Achievement . . . . . . . . . 67

4. ANOVA for Rating of the Learning Cell 70

Experience . . . . . . . . . . . . . . . .

vii

   

LI ST OF FIGURES

Figure Page

1. Variable Matrix . . . . . . . . . . . . . 55

viii

 

 

CHAPTER I

BACKGROUND

Rising costs coupled with reduced income due to
shrinking enrollments and proportionally lower financial
support have thrust many institutions of higher education
into an economic dilemma. This has caused several univer-
sities and colleges to reduce or eliminate selected pro-
grams, and there have been instances of certain institu-
tions being forced to close their doors due to insolvency.
One outcome of this trend of fiscal difficulties has been
a decreased teacher/student ratio. The college teacher
is therefore challenged to provide effective instruction
at the same time that class size increases without corres-
ponding gains in resources.

A promising alternative to the large lecture classes
was introduced at McGill University by Kingsbury in 1966.
This innovation called the learning cell or learning dyad
consists of two people working together to master coopera-
tively a given learning task. Since no additional costs
need be incurred to implement the learning cell method in

a class except perhaps for resource or supervisory peOple

l

 

and given the financial problems that plague higher educa—
tion, the further development and refinement of the learn—
ing cell may be desirable.

Recent work done at the Institute for Educational
Sciences at Oakland Community College under the direction
of Hill and Nunney has produced a most powerful way of
identifying individual learner characteristics. Batteries
of tests and inventories are administered to a learner and
the results are then analyzed to describe the various ways
that thatlearner might use to seek meaning. The unique
method one uses to seek meaning from one's environment is
known as educational cognitive style. The process of de-
termining a learner's cognitive style is called mapping
and the resultant profile is referred to as a cognitive

style map.

Problem Statement

 

Although the learning cell has been shown to be a
viable instructional alternative, there is no formalized
procedure to match potential learning cell mates. The
criteria for the determination of partners seem to be sub—

jective ones.

Purpose Of The Study

 

The primary purpose of this study was to test three
methods of matching partners in a learning cell. The three

methods were to pair students whose cognitive style maps

are highly similar, to let students pair themselves and

to pair students randomly.

Specifically, this study was intended to do these

two things:

1.

Compare achievement of students in the three
conditions.

Compare ratings of the learning cell exper-
ience made by students in the three kinds of

learning cells.

This study also explored the influence upon achievement

and ratings of the learning cell experience of matching

partners who were of the same or different sex and of

matching partners who were relatively the same or dif-

ferent in terms of chronological age.

Assumptions

 

There are seven assumptions that were made regard—

ing this study.

1.

The students who were the subjects in the study
were representative of all the students who
take the introductory psychology course at
Oakland Community College.

The responses of the subjects to the various
instruments used in the study were normal.

The instruments used to measure cognitive

style were valid. Thousands of subjects have

   

responded to the instruments and the resulting
empirical data have been used in an ongoing
revision process.

4. The instruments used to measure achievement
which were instructor produced were valid
because of the content expertise of the
instructors.

5. The instrument used for the learning cell
experience rating was valid.

6. There was independence of groups.

7. The groups were equivalent. Analysis of Co-
variance was used to assure group equiva—

lence in terms of achievement.

Delimitations Of The Study

Although a study of this kind could have been
conducted with several age groups and in various educa—
tional settings, this study occurred at Oakland Community
College with subjects in an introductory psychology
course. To have conducted the study elsewhere would
have required the collection of considerable data about
cognitive style which were readily available at Oakland
Community College. Generalization of the findings of the
study to settings different from that of the study or to
content areas other than psychology should be done with
caution. A final delimitation is that no treatment repre—
senting regular instruction, a control group, was in the

study.

   

Research Questions

 

The three questions below represent the primary
emphasis of this study.

1. Do student pairs who are matched by cognitive
style similarity attain higher achievement and rate the
learning cell experience more favorably than those who
are randomly matched or those who self select?

2. Do student pairs who are similar in chrono-
logical age differ in achievement and the rating of the
learning cell experience from students who have larger
age differences?

3. Do student pairs of the same sex differ in
achievement and the rating of the learning cell exper-

ience from students of different sexes?

Research Hypotheses

 

The following hypotheses were derived from the
reSearch questions. Achievement was measured in two ways.
The first was the existing instructors' evaluation of how
well a student could either apply the course material to
his own environment in essay form or demonstrate mastery
of the content by responding to a multiple choice test.
The second achievement measure was a final examination.
Those items on the final that tested content that was
covered prior to the beginning of the study were not used

to calculate the achievement scores for the study. A ten

 

item Likert type instrument was used to assess student

rating of the learning cell experience.

H

l:

N

The mean score for the four weekly
measures of achievement of students

who are matched by cognitive style
similarity will exceed the mean sources
of the randomly paired students and the
self selected students. ___

 

 

 

The mean score for the four weekly
measures of achievement of students
whose age difference is narrow will
differ from the mean score of students
whose age difference is wide.

 

 

The mean score for the four weekly measures
of achievement of students whose sex factor
is the same will differ from the mean score
of students whose sex factor is different.

 

 

There will be a matching method by age
difference interaction for the dependent
variable, weekly achievement.

 

 

There will be a matching method by sex
factor interaction for the dependent vari-
able, weekly achievement.

 

 

The mean score for the final examination
of students who are matched by cognitive
style similarity will exceed the mean
scores of the randomly paired students
and the self selected students.

 

 

 

The mean score for the final examination of
students whose age difference is narrow will
differ from the mean score of students whose
age difference is wide.

 

 

The mean score for the final examination

of students whose sex factor is the same
will differ from the mean score of students
whose sex factor is different.

 

 

There will be a matching method by age dif—
ference interaction for the dependent variable,
achievement on the final examination.

 

 

   

H10: There will be a matching method by sex
factor interaction for the dependent
variable, achievement on the final exam—
ination.

 

 

H11: The mean rating of the learning cell ex-
perience by students who are matched by
cognitive style similarity will exceed
the mean ratings by students who are
randomly paired and by students who are
self selected.

 

 

 

 

 

 

The mean rating of the learning cell exper-
ience by students whose age difference is
narrow will differ from the mean rating

by students whose age difference is wide.

12‘

 

 

H : The mean rating of the learning cell ex-
13 . .
perience by students whose sex factor is
the same will differ from the mean rating
by students whose sex factor is different.

 

 

There will be a matching method by age
difference interaction for the dependent
variable, rating of the learning cell
experience.

 

14‘

 

H15: There will be a matching method by sex factor
interaction for the dependent variable,
rating of the learning cell experience.

 

Definitions of Important Terms

 

The listing of key terms and their meanings is
provided to communicate to the reader the use of those
terms in the restricted context of this study.

1. Student: A subject in the study who must be enrolled
in one of four sections of introductory psychology
taught by Dr. Mueller or Dr. Svagr at Oakland Commu-
nity College during fall semester of 1974.

2. Instructor: Dr. Mueller or Dr. Svagr.

 

 

 

Learning Cell: Two students working together to

 

master cooperatively a given learning task.

Cell Mates: The two students who comprise a learning

 

cell.

Cognitive Style: Educational Cognitive Style. The

 

unique method one uses to seek meaning from one's

environment.

Orientation: The relative strength of a given ele-

 

ment within one's cognitive style. Orientation is

expressed as one of three conditions.

a.

Major: A given element appears in the 50th-
99th percentile range of a distribution for
that element.
Minor: A given element appears in the 26th-
49th percentile range of a distribution for
that element.

Negligible: A given element appears at or below

 

the 25th percentile in a distribution for that

element.

Matching Method: The technique used to determine

 

cell mates. The three methods in this study are:

a.

Cognitive Style Similarity: Learning cell

 

whose cell mates' cognitive styles are at least
86% alike.
Random: Learning cell whose cell mates were

paired by chance.

c. Self Select: Learning cell whose cell mates

 

mutually agreed to work together without
instructor intervention other than directions
to choose partners.

8. Age Difference: The closeness of the chronological

 

ages of cell mates expressed in one of two conditions.
a. Narrow: Learning cell in which the chronological
age difference of the cell mates is in the lower
half of the distribution of the chronological
age differences of the cell mates of all the
learning cells of the study.

b. Wide: Learning cell in which the chronological

 

age difference of the cell mates is in the upper
half of the distribution of the chronological
age differences of the cell mates of all the
learning cells of the study.

9. Sex Factor: The sex composition of a learning cell.

 

The two groups are:

a. Same: Both cell mates are female or both are

 

male.

b. Different: One cell mate is female and one is

 

male.

Overview of Other Chapters in the Study
Chapter II provides a review of the literature

related to learning cells and cognitive style similarity.

 

10

Chapter III explains the methods and procedures used in
the study while Chapter IV is a presentation of the analy-
sis of the data. Chapter V, the final chapter, includes

a summary of the study, conclusions and recommendations

for further research.

   

CHAPTER II

REVIEW OF RELATED LITERATURE

The review of the literature presented in this

chapter is structured by six questions that were used to

analyze the research studies and other scholarly works.

The questions are:

l.

2.

What is the learning cell?

What is the theoretical support for the
learning cell?

What advantages does the learning cell offer?
How have researchers matched students in
learning cells?

How have researchers determined cognitive
style similarity?

What is the support for using cognitive style
similarity as a criterion for matching stu-

dents in a learning cell?

The first four questions deal directly with the learning

cell and the last two questions draw upon studies from the

educational sciences that involved educational cognitive

style similarity as a matching variable. This review is

organized to present each of the six questions and the

11

12

responses to those questions that are derived from the
literature. Implications for this study resulting from

the literature review will then be cited.

What Is The Learning Cell?

 

The process of defining the learning cell will
occur at two levels. The first will deal with the char—
acteristics that are common to all learning cells with
particular attention given to how learning cells and tu-
torial systems differ. The second level assumes the first
and will proceed to list six factors that may be useful
in describing the variations of the learning cell.

Goldschmid (1971) regards the learning cell as
two people working together in the classroom, the same
definition used by Schermerhorn (1972). A more precise
statement is offered by Alexander, Gur, Gur and Patterson
(1973): "A learning cell is a dyadic unit in which part-
ners mutually teach and learn from each other [p.JJ." The
partners in a learning cell are both students and it is
assumed that both have yet to master the given learning
task. Teachers and tutors are thus excluded from a learn-
ing cell. Any tutorial arrangement provides for the ver-
tical flow of information from the tutor down to the
tutee. The tutor may be a teacher, a student who has
previously mastered the learning task (Hapkiewicz, 1972;
Sheppard and MacDermot, 1970), or a student who has not

necessarily mastered the learning task but who has been

l3

trained to be the tutor (Rosenbaum, 1973). A tutorial
mode cannot be a learning cell because the learning cell
requires that both learners be novices.

Nearly all of the learning cells that have been
reported were initiated by a teacher or a researcher.
However, Alexander, Gur, Gur and Patterson (1973) con—
ducted one study to investigate students who had decided
to work in learning cells independent of any suggestion
by a third party. This study is mentioned to confirm that
two person learning systems that qualify as learning cells
do occur outside the formal classroom or laboratory.

Thus a basic way that learning cells may differ is the
source of the idea to initiate the learning cell.

The scarcity of literature about informal learn-
ing cells might be partially explained by the fact that
those learning cells function outside the formal structure
of an educational institution and are ignored by most
researchers. It is suggested that informal learning
cells are probably quite common and further research like
that by Alexander, Gur, Gur and Patterson could be use—
ful to those who want to try the learning cell idea in a
learning system.

Cell mates in a student-initiated learning cell
define the interaction pattern of the learning cell for
themselves. Students in a teacher- or researcher-ini-

tiated learning cell might define the pattern or the

 

 

 

 

 

 

l4

pattern might be imposed by the teacher or researcher.
This suggests another way that learning cells can differ
which is who defines the interaction pattern of the learn-
ing cell. The majority of studies about learning cells
that were not student-initiated showed the teacher or
researcher deciding the pattern (Dick, 1965; Goldschmid,
1970; Hartley and Hogarth, 1971; Kingsbury, 1968; Myers,
Travers and Sanford, 1965; Schermerhorn, 1972). Amaria,
Biran and Leith (1969) let the students define the pat-
tern. One of the studies by Alexander, Gur, Gur and Pat-
terson (1973) treated the source of pattern as a variable.
Another factor that can be used to distinguish
between different learning cells concerns advanced prep-
aration by the cell mates. In some cases the cell mates
were urged to read and study a given assignment before
the next learning cell session. This was true in the
studies reported by Goldschmid (1970) and Schermerhorn
(1972). The subjects who participated in experiments
done by Alexander, Gur, Gur and Patterson (1973), Amaria,
Biran and Leith (1969), Amaria and Leith (1969), Hartley
and Hogarth (1971), Kingsbury (1968) and Myers, Travers
and Sanford (1965) worked on the learning task only while
in the learning cell setting. The research into student-
initiated learning cells by Alexander, Gur, Gur and Pat-
terson (1973) showed that most cells chose to prepare in

advance but that a few did all of the studying together.

15

Goldschmid (1970) who required his students to
prepare in advance offered them two ways that the prep-
aration might occur. In one option both partners were to
be held responsible for the same assignment and in the
other option each student would have a different assign—
ment. Learning cells in which partners prepare in advance
can thus be broken into two smaller groups using the vari-
able, assignment congruity.

Two other factors that might be used to describe
differences between learning cells both deal with the use
of written study questions. The first issue is whether
the learning cell session ought to center around study
questions that have been written before the session be-
gins or whether the format should be more conversational
with questions arising spontaneously. Specific questions
written in advance appeared in the work of Goldschmid
(1970), Hartley and Hogarth (1971), Kingsbury (1968),
Myers, Travers and Sanford (1965) and Schermerhorn (1972).

The final factor that relates to written study
questions is the origin of the questions. They could be
authored by the students or they could be externally gen-
erated. Stone (1974) has suggested that more effective
learning in a learning cell might result if the instructor
were to write the questions and provide them for the stu-
dents. Externally generated questions were used by Hart-

ley and Hogarth (1971), by Kingsbury (1968) and by Myers,

l6

Travers and Sanford (1965). Goldschmid (1970) and
Schermerhorn (1972) defined the learning cell partners
as the source of the written questions.

The writer has defined the learning cell as two
peeple working together to master cooperatively a given
learning task. It was assumed that neither person should
have prior mastery of the task and so tutorials were ex-
cluded. Six factors were identified that could be used
to describe how learning cells may differ. Expressed as
questions the six factors are:

1. Who initiated the learning cell?

2. Who defined the interaction pattern?

3. Was advance preparation done by the cell

mates?

4. If advance preparation was done, were the

assignments of the cell mates congruent?

5. Were written study questions used to guide

the session?

6. If written study questions were used, who

wrote them?

What Is The Theoretical Support
For The Learning Cell?

 

 

An effort will be made to build a theoretical
base for the learning cell by identifying selected fac—
tors that are believed to facilitate learning and linking

each factor to the learning cell. The purpose is to

17

provide a theoretical perspective for the reader as he
approaches this study.

Four principles were presented by Moore and Ander-
son (1969) for the design of clarifying learning environ-
ments, those that enable the learner to see more clearly
what is happening in the instructional environment and
what he is doing in that environment. The productive
principle which is the third one offered by Moore and
Anderson will be deleted from this discussion because it
concerns criteria for content selection based upon the
logical structure of the content. The learning cell is
an instructional arrangement that treats decisions about
content selection as givens.

The perspectives principle states that increasing
the number of perspectives that a learner can take will
improve learning. The popular notion that a good way to
learn something is to teach it (Bruner, 1965; Bugelski,
1964; Gagne and Roher, 1969; Johnson, 1972; Schermerhorn,
1973) seems to demonstrate this principle. The responsi-
bility for teaching tends to add other dimensions to one's
approach to a learning task. The dynamics of the learn-
ing cell typically cause the participants to alternate
between the learner and the teacher roles. This Oppor-
tunity to be both learner and teacher should provide a
wider range of perspectives than could be experienced if

only the learner role were available.

18

The second or autotelic principle says that
learning is enhanced if the environment affords protec-
tion for the learner. The consequences of the learner's
interaction with the environment should minimize physi-
cal risk as well as psychological and social risk. It
is in the latter area that the learning cell has shown
evidence of the autotelic principle. Torrance (1969)
found that young children were more likely to try a dif-
ficult task if they were in pairs than if they were alone
or before an entire class. Later studies also by Torrance
(1970, 1971) showed that higher levels of creativity were
attained on creative tests by both members of a dyad than
by students working alone and that students in dyads had
stronger feelings of stimulation, enjoyment and origie
nality. The dyad would appear to be a useful means of
helping students to feel less anxious about attempting a
task perceived to be difficult by those students and of
building a more supportive environment.

The personalization principle stresses the impor—
tance of an environment that is responsive and reflexive.
It is responsive if it encourages the learner to generate
questions and if it then gives relevant responses to that
learner. The learning cell does encourage questioning
behavior from the students. Concerning the responses in a
learning cell Stone (1974) says, "Feedback is relevant,
timely, frequent, and likely to be presented on a cognitive

level more closely related to that of the student [p. 19]."

19

A reflexive environment facilitates the learner's
gaining a better understanding of himself as a learner.
The learning cell is a social unit that is often charac-
terized by alternating questions and responses. In such
a setting one might think about how his partner approached
a problem and how this compared with his own method. This
could lead to an open discussion of study habits and styles
of learning. Such events are not guaranteed in a learning
cell but the learning cell does seem to make them more
probable than certain other methods of instruction. The
learning cell provides another readily available learner
who is working toward mastery of the same learning task.
Using such a person as a basis of comparison could be a
step in the process of self assessment.

Reflexive behavior might also result from playing
the teacher role in the learning cell. As previously
noted, preparing as a teacher and preparing as a student
can be quite different. The teaching side of the learn-
ing cell offers two more referents that the learner might
use to assess himself. He could compare his learning
style to his own teaching style and he could compare his
learning style to his partner's teaching style.

If an instructional environment can engage a
learner in active participation, then it is generally
agreed that effective learning is more likely than if the

learner were more passive. The key to the previous

20

statement is the distinction between active and passive
participation. According to Burton (1967): "Active par—
ticipation is preferrable to the kind of passive reception
usually involved in listening to a lecture or watching

a motion picture [p. 105]." Burton appeared to equate
active participation with a measure of the learner's ver-
bal behavior or physical movement since a learner attend-
ing to a film or a lecture would probably be quiet and
immobile. This way of judging participation seems inade-
quate since they are superficial measures. Reading a book,
attending a lecture or watching a film can engage the
learner in a very intensive experience.

Rogers (1967) preferred to use the degree of
responsible involvement as the critical criterion for
judging the merits of participation. One participates
more responsibly when he has a larger share of deciding
his own course of action. After noting that the typical
learning environment places the control and responsibility
for information, rewards, participation and organization
outside the learner, Argyris (1962) concurred with Rogers
by saying that a learning environment should move the
learner toward increased self-responsibility and internal
commitment which result in a more independent learner.
Further support for this interpretation of active partici-
pation comes from Mager (1961) who contended that increas—
ing the learner's control of the learning experience would

result in greater learner motivation.

21

Bruner (1968) recommended that an individual who
can see his own contributions as beneficial to a group to
which he belongs would be reinforced for his behavior
and thus more highly motivated. If Bruner's idea can be
added to those of Rogers, Argyris and Mager expressed
above, then additional support for the learning cell can
be suggested. This combination would describe a learning
cell in which a learner could benefit from having a major
decision making role and also from making significant in-
put toward accomplishing the goals of the group. Such an
environment would afford enough autonomy to avoid external
control of all decisions and would foster an atmosphere of
cooperation. A learning cell does not necessarily dupli-
cate the hypothetical ideal just described because there is
no assurance that teacher domination will not be replaced
by partner domination or that one will be able to contri-
bute to cell effectiveness. However, the learning cell
can provide the opportunity for some degree of responsible
participation for each individual and it may enable a
student to further a group's cause. In the latter respect
the learning cell may be better able to meet this qualifi—
cation than larger units since the ratio of available time
to number of participants is greater in a two person system
than in any system with more than two members.

This discussion of theoretical support for the

learning cell will end with a consideration of the learning

22

cell as a message sent by the instructor to the learner.
According to Harrison (1973):

The way man arranges himself in time and space can
have sign value for the astute observer. The way
man creates, selects, arranges, and presents his
artifacts may reveal much about his intentions,
feelings, and aspirations. Often, this use of time,
space and object is done without an intent to com-
municate. It is a symptom, a signal, an index.

But the aware communicator can use time, space and
artifacts in the purposeful creation of signs. He
can use these code systems to structure communication
situations, to facilitate or inhibit interaction, to
maintain and perpetuate effective communication
systems [p. IX-2].

The designer of a learning environment is in a particular-
ly important position because his decisions about time,
space and artifacts affect and even dictate the behavior
of all who enter that environment. Even the very young
respond with precision to the nonverbal cues that come
from the way that time, space and artifacts present them-
selves.

Davis (1970) alerted educators to be aware of the
nonverbal messages inherent in learning environments when
he wrote:

The point of these final comments is simply to make
clear the fact that the selection and design of in-
structional models is not merely a question of im-
proved learning of course content, that other forms
of learning must also be kept in mind. In the final
analysis, a student learns only those messages he
processes, and one ought never to forget, that the
model also communicates a message. As a matter of
fact, in many courses, messages about the model seem

to be the only messages that some students ever
really learn [p. 38}.

23

After comparing the effects of co-operation and competi-
tion upon learning, Thompson (1972) agreed with Davis when
he advised that there were other considerations beyond
pupil performance when evaluating co-operation and compe—
tition. One should also assess the "side effects" of
using co-operation or competition.

At a time when accountability, based on student
performance, is being emphasized the words of Davis and
Thompson suggest that the narrow cognitive outcomes view
of accountability is rather like the tip of the iceberg.
Dewey (1938) addresses this issue when he wrote about
collateral learning.

Perhaps the greatest of all pedagogical fallacies
is the notion that a person learns only the partic-
ular thing he is studying at the time. Collateral
learning in the way of formation of enduring
attitudes, of likes and dislikes, may be and often
is much more important than the spelling lesson

or lesson in geography or history that is learned.
For these attitudes are fundamentally what count

in the future. The most important attitude that
can be formed is that of desire to go on learning
[1)- 48.] .

What messages might a teacher be communicating to
his students via the learning cell? The following are
suggested examples of those messages.

1. You are capable of accepting more responsi-

bility for your own learning.

2. Your ideas are worthwhile.

3. Working together with another person is a

valuable experience.

24

4. You should have practice sharing your ideas

with another person.

5. Learning can and should be enjoyable.

6. I would prefer to support rather than direct

your learning.

7. Learning addresses more than cognitive know-

ledge.
It does seem that students who work in learning cells
often like the experience. Perhaps one reason that the
learning cell is well accepted by students is that some
of the messages listed above are being sent and received
and the students appreciate those messages.

An attempt has been made to show theoretical sup-
port for the learning cell by presenting factors that may
improve learning and then relating those factors to the
learning cell. It has been suggested that the perspectives
principle, the autotelic principle and the personalization
principle of Moore and Anderson (1969) may be satisfied by
the learning cell and that the learning cell may encourage
active participation in the form of responsible behavior
by the student. Finally, the learning cell has been
described as a nonverbal message that may communicate
beyond the mere transmission of course content.

What Advantages Does The
Learning Cell Offer?

 

 

Central to the evaluation of an innovation is the

gathering and interpretation of evidence as that new idea

25

is applied. Adecision to adopt, amend or reject the inno-
vation should stem from such an analysis. The purpose
of this section is to present and interpret evaluative
information regarding the effects upon the learner of
using the learning cell and other two person interaction
systems with elements common to the learning cell. Al—
though a prime concern is the effect upon cognitive learn-
ing, the influence upon the learner's attitudes and upon
the learning environment will also be considered.
-Goldschmid (1970) designed four instructional
options and let students select the option they preferred.
The four options were to participate in a discussion
group of six to twelve students, to be in a seminar group
of ten to twelve, to work in a learning cell or to pur-
sue an essay option. Although there were no differences
between groups on a final examination, the learning cell
group did better on an unannounced essay examination. The
learning cell group also showed higher rankings on mea—
sures of morale throughout the term and they had the
highest peer and self evaluations of achievement.

Myers, Travers and Sanford (1965) conducted a
study to compare the effectiveness of four learning con-
ditions in learning a rote-memory task. Elementary stu-
dents were assigned to play the teacher role in a dyad,
the pupil role, to be in a dyad where they would reverse

roles or to use self-instructional materials alone.

 

26

Students in the pupil condition showed superior learning
and students in the teacher condition had the least learn-
ing. Students in the reverse roles condition, the typical
learning cell format, learned about equally and it was
reported that these students maintained a higher degree

of interest and attention throughout the three days of

the task. This reference to morale seems noteworthy since
the likelihood for suCcess by a learner is usually en-.
hanced if his attention and interest can be maintained.
The investigators suggested that the instructional activ-
ity represented by the reverse roles or learning cell
condition seemed to be most likely to be useful in the
classroom and that it should be developed further.

A study reported by Alexander, Gur, Gur and Pat-
terson (1973) showed that the learning cell was a more
effective means of helping students learn to solve mathe-
matical problems dealing with linear motion of physical
bodies than was individual study. More enthusiasm was
observed in students in the learning cell condition which
is in agreement with the findings of Myers, Travers and
Sanford. Further evidence for this point is furnished by
Torrance (1970) who found that both kindergarten and
college students working in pairs showed greater perse-
verance in a task requiring the creation of original re-
sponses than students working alone. Torrance stated,

"At the end of the fourth task, Ss in dyads seemed to be

27

going stronger than ever and to be having fun while those
working alone seemed to be fatigued and ready to stOp
[p. 393]."

Alexander, Gur, Gur and Patterson (1973) reported
a further study that investigated spontaneous or naturally
occurring learning cells. Thirty-six students were inter-
viewed who were studying together in dyads for some
course. Of the twenty-five students for whom data were
available, the mean grade in the course for which students
had studied in a dyad was higher than the previous grade
point average and the grade point average for the other
courses that term. Generalizations must be somewhat lim-
ited since data for eleven of the thirty-six students
could not be obtained.

Sheppard and MacDermot (1970) compared an experi-
mental group who participated in dyadic interaction to a
control group in an introductory psychology class. The experi-
mental group scored significantly higher on a hundred-
question final objective examination and on a five-
question essay examination. The experimental group also
ratedtflmacourse more favorablijNterms of being well orga-
nized, stimulatinganuienjoyable. The internal validity of
this study is questioned because the experimental subjects
were told in advance that the hundred-question final exami-
nation would not count in determining their course grades

whiletflmasubjectsixithe control group learned that fifty

 

28

per cent of their grades depended on the test. The sub-
jects in the experimental group went intotflmafinal exami-
nation already knowing what their grades were for the
course. The investigators pointed out that these condi-
tions should have favored control group performance. An
alternative explanation is that the experimental group was
relaxed while the control group had high test anxiety and
that the results reflect this difference.

In a similar study Hapkiewicz (1972) found that
students who participated in dyadic interviews demonstra-
ted higher achievement on in-class tests in an educational
psychology course and they rated the course more positive—
ly than students in a control group. No differences on
the final examination were detected between groups.

Comparing the effectiveness of paired versus
individual work on programmed instruction materials was
the topic of two studies reviewed. Amaria, Biran and
Leith (1969) found that children in the dyadic situation
demonstrated superior learning to children working alone
for both heterogeneous and homogeneous pairs. They sug-
gested that the social relationship in dyads may lead
students to focus specifically on the development of
the teaching steps and that this attention results in a
higher degree of success in the learning task.

A similar point was made by Gartner, Kohler and

Riessman (1970) who made a comprehensive review of the

29

literature related to tutorial programs. They found that
it was the tutor who benefited more in a tutorial setting
and this was attributed to the insight into the teaching—
learning process gained by the tutor and more specifi-
cally an understanding of how the tutor himself learned.
Thelen (1969) who also reviewed tutorial programs noted
that a major benefit of tutorials was that the tutors
learned how to learn.

College students worked alone or in pairs in a
programmed math course in an experiment by Dick (1965).
No differences were found between the groups on daily
tests, the mid-term, the final examination, a transfer
test or on course ratings. The final examination was re-
administered a year later and the students who had worked
in pairs did significantly better than those who had
worked alone. Dick indicated that the social interaction
may have accounted for the superior results. The pairs
took longer to finish each of the twenty—eight units
which may have increased each learner's actual interaction
with the course content. Verbal interaction acting as a
reinforcer of learning was also identified as a potential
factor to explain the retention differences.

The studies that have been reported suggest that
dyadic interaction systems can be effective instructional
modes to facilitate cognitive learning. It was noted that

studentsiJNdyads often displayed attending behavior longer

30

and that they were better able to understand the struc-
ture of the learning task. Alexander, Gur, Gur and
Patterson (1973) reported that students in learning
cells covered more material and also went intormnxadepth
with the material. Student interaction with academic con—
tent and subsequent cognitive performance relative to
that content are important concerns, but it is contended
that a discussion of these concerns is necessary but not
sufficient as one evaluates the learning cell.

In assessing the influence of tutorial programs
Thelen (1969) stated that dyadic systems can be a means
of causing a more positive environment of concern for
others rather than one of competitiveness. An article

in Nation's Schools (1968) that described a buddy system

 

for learning said this about students in the program:
"They don't feel that they are on their own. They share
their goals [p. 113]." Alexander, Gur, Gur and Patterson
(1973) and Goldschmid (1971) agreed that feelings of
alienation in large classes can be reduced with the learn-
ing cell. Students reported feeling better about going
to class and about taking tests. An important feature

of peer-mediated instruction is that the social value of
working together is encouraged according to Rosenbaum
(1973). Frazier (1970) wrote that pair learning teaches
students to live and work together as they seek a common

goal, to better understand oneself, to respect the

31

uniqueness of each person and to care about another's
well-being. Frazier concluded by saying, "After all,
building bonds between people in twos forms the base of
most of the truly satisfactory human relations-~in the
family, in friendship and in the business world [p. 100]."

A review of the literature about dyadic learning
systems with a special emphasis on the learning cell has
shown certain benefits that this kind of instruction may
offer. They include improved cognitive learning, active
participation, a more personalized learning environment,
feedback that is frequent and_timely, more confidence
within the learner and a keener insight into self and
others. It appears that the learning cell is an instruc-
tional medium with sufficient potential to merit further
development and refinement.

How Have Researchers Matched Students
In Learning Cells?

 

 

An investigation of the influence that the method

of matching students may have upon the learning cell is

a primary emphasis of this study. It is therefore of
interest to note the methods used in previous studies

to match learning cell partners and the rationale for
using those methods. Opinions about matching have been
offered by people who have observed but not necessarily
researched learning cells and those opinions will also

be examined.

32

The random pairing of students was the most common
method appearing in the literature. Random pairing was
used in one of the studies by Alexander, Gur, Gur and
Patterson (1973), in the study by Dick (1965), in the
study by Myers, Travers and Sanford (1965) and in the
studies by Torrance (1969, 1970, 1971). Although no
reason for the use of random assignment was offered in
any of these studies, the researchers were concerned
with independent variables other than method of matching
students and therefore they probably used random assign—
ment to exclude method of matching as a variable.

Amaria, Biran and Leith (1969) used IQ scores
and Hartley and Hogarth (1971) used prior grades to match
students into homogeneous and heterogeneous pairs. There
had been an indication from previous research on co-oper—
ative learning that mixed ability grouping might be bene—
ficial to learning. These two studies were conducted
to test that idea. The variables, introversion/extro-
version and high anxiety/low anxiety, were added to the
variable, homogeneity/heterogeneity, in a study by Amaria
and Leith (1969). The two personality factors were se-
lected because they were thought to influence performance
in a co-operative learning situation.

One study by Alexander, Gur, Gur and Patterson
(1973) which was not completed paired students according

to sex, academic competency and previous experience in

33

the subject area. These three characteristics were
chosen because all of the information would be readily
available to an instructor wishingtx>use learning cells.

The matching methods usedkanoldschmid (1970) and
Schermerhorn (1972) could not be determined. In summary,
there were six studies that used random pairing and there
were four that used a method other than random pairing.

Westmeyer (1965) studied the learning pairs that
existed in chemistry laboratories. He stated, "Studies
have shown that groups tend to be more compatible if they
are set up as a result of the students' choices rather
than the arbitrary judgment of the teacher [p. 355]."

The studies that Westmeyer used as a basis for his posi—
tion are not cited in his article nor did he test his
idea in a research study.

Frazier (1970) recommended that the following
factors be considered in pairing: sex, ability, disposi-
tion, background experiences, maturity, skills. Frazier
did not advise how to translate the consideration of
these factors into decisions about pairing. Mauer (1968)
described pair learning where the pairs were homogeneous
but he failed to list the criteria used to determine
homogeneity. Neither of these two people offered a ra—
tionale to support his matching method nor did either

conduct any research to test the methods.

34

The learning cell is an instructional mode that
has only recently been employed at the direction of the
teacher and examined by the researcher. Most research
about learning cells has treated the learning cell as an
independent variable. Students working in learning cells
were compared to students working alone on the same task.
Not much attention has been given to independent vari-
ables that might influence the learning cell. It is sug-
gested that advances in the use of the learning cell will
depend upon research that is directed toward such vari-
ables that are within the learning cell itself.

How Have Researchers Determined
Cognitive Style Similarity?

 

 

What is the Support for Using Cognitive
Style Similarity as a Criterion
for Matching Students in
a Learning Cell?

 

 

 

The central concern of this study is a compari-
son of three methods of matching students in learning
cells with particular attention given to the use of cog—
nitive style similarity as a matching variable. Seven
studies have been reviewed that treated cognitive style
similarity as an independent variable and that used
achievement and/or an attitude or effectiveness ratings
as dependent variables. Other variables were also con-
sidered by the researchers but this review will focus
directly on the variables mentioned above because only

they are germane to this study. The reader is advised

35

that none of the studies deal directly with matching
one student with another and that this limits somewhat
their applicability to this study.

Wasser (1969) mapped the teacher and ten stu—
dents in each of three sixth grade classrooms. The
map of each student was compared to his teacher's map
and the number of cognitive style elements they shared
was noted. Each of the ten students in a class was ranked
according to the number of elements that that student
shared with his teacher. It was found that the teacher
tended to give higher letter grades in math, language,
health, social studies, science, reading and spelling to
the students who were in the upper half of the distribu—
tion, those most like the teacher, than to the students
who were less similar.

DeLoach (1969) investigated whether correlation
on a high to low similarity continuum of the cognitive
style of an administrator and an instructor would act as
a significant variable in the evaluation of the in—
structor by the administrator. Five instructional di-
vision chairmen and two instructors from each of the
same five divisions in a community college were the sub-
jects. DeLoach compared all of the instructor cognitive
style maps to each administrator map and then put each
instructor into a high or low similarity group relative

to each administrator. The mean of the evaluation scores

36

assigned by the administrators to the instructors in the
low similarity groups was significantly lower than the
mean of the evaluation scores of the highly similar
groups.

Thirty—two students and two teachers in a draft—
ing course at a community college were the subjects in a
study by Fragale (1969). The total number of cognitive
style elements for a given student was calculated as
well as the number of elements that the student shared
with his teacher. The number of shared elements was
divided by the total elements of the student and the
quotient was converted to a percentage. Those students
whose percentages exceeded fifty were the high match
group and the others were the low match group. Fragale
described two conditions of agreement and two conditions
of disagreement regarding the relationship between de-
gree of match and final grades in the course. The match
and grade factors were in agreement if a student were
in the high match group and got a high grade in the course
or if a student were in the low match group and got a low
grade. The two disagreement conditions were either
high match-low grade or low match-high grade. One
teacher had 66% of his students in an agreement situation
and the other had 84% in an agreement situation. Fragale
concluded that the matching of the cognitive styles of

‘teachers and students did affect the education processes.

 

37

Schroeder (1969) studied the cognitive styles of
one hundred eighteen high school English students and
their teacher to determine if students similar to the
teacher in terms of cognitive style would rate the teacher
higher on effectiveness and get higher grades than less
similar students. A 70% criterion was used to distinguish
similar and disjunct student groups. The particular
method used to determine the percentages was not clear.
Schroeder found that the similar group did rate the
teacher as more effective and did receive higher grades
to a significant degree.

Fifty students and two instructors in a math
course at a community college were studied by Blanzy
(1970) to see if the similarity of student and teacher
cognitive style were related to the achievement of per-
formance goals or to the evaluation of the teacher by a
student. The number of profiles for each subject was
determined by multiplying together the number of elements
in each of the three sets of a cognitive style map. The '
number of profiles that a student shared with a teacher
was divided by the total number of profiles in that stu-
dent's map. The resulting decimal numbers were ranked and
divided into an upper half and a lower half to make two
groups. Blanzy found no difference between the groups in
terms of performance goals finished or the evaluation of

the teacher by the students.

 

38

A study of two hundred fifty-five students and
thirteen instructors in a nursing course was conducted
by Lange (1972) to determine if the matching of a stu-
dent's cognitive style map and preferred teaching style
to the instructor's cognitive style and teaching style
would affect the grades of the students or their percep-
tions of their instructor. This study differs from the
others in that teaching style was added to cognitive
style as a matching criterion. Each element of an in-
structor's cognitive and teaching style was assigned a
value of two. If a student matched an element exactly,
then the student also was assigned a two for that element.
If the instructor had a major and the student a minor or
if the instructor had a minor and the student a major,*
then one was assigned to the student for that element.
After all of the points had been tabulated for a student
in this manner, the number was divided by the total points
of the teacher and the quotient converted to a percentage.
To be in the matched group a student had to have at least
an 80% match with the instructor. Lange found no overall
difference in final grades between the matched group and
the non-matched group but the matched group did perceive

their instructor more positively.

 

*A description of major and minor orientations
appears in the Definitions of Important Terms section
of the first chapter.

 

 

39

Oen (1974a) compared the cognitive styles of
four hundred eight technical institute studentstx>their
teachers' cognitive styles using a system similar to
Lange's. Oen (1974b) used the number of common elements
shared by a teacher and a student to rank the students
and then used the median to divide the students into a
match group and a mismatched group. The findings of this
research showed that the matched students were more likely
to get higher grades than the mismatched students.

A review of the seven studies just cited shows
that no two researchers used precisely the same method
to determine cognitive style similarity. There seem to
be three questions that describe the important methodolog-
ical variables regarding how cognitive style similarity
was established in these studies. Is the number of cogni-
tive style elements common to both people used directly
to define two conditions of cognitive style similarity
or is the number used as the numerator of a ratio? If
a ratio is used to compute a decimal or percentage that
represents the degree of match, is the denominator of the
ratio the total number of cognitive style elements of the
student or the teacher's total number? Is the median
used to distinguish between two conditions of similarity
or is a predetermined percentage used? Since the answers
to these questions vary across the studies, the actual

meaning of cognitive style similarity is also a variable.

 

40

An analysis of the various results of the studies and a
synthesis of the findings into singular conclusions can-
not be done because technically each of the seven
researchers had a different definition of terms.

There seems to be value in noting the general
direction of the findings of the studies because to treat
the studies as seven unrelated entities is to negate some
of the potential of the research. Each of the research-
ers has in a sense engaged in a pioneering venture. The
methods and definitions have fluctuated because of the
probing nature of the research. It is probable that the
chief contribution of these studies has been to lay the
foundation for further work rather than to test ideas
conclusively.

Of the six studies that addressed the relationship
between teacher-student cognitive style similarity and
academic achievement or grades, four supported the idea
that a student tends to have more success when his cogni-
tive style is relatively similar to his teacher's cogni-
tive style than if it is less similar. Two studies did
not support the idea.

In three of the four studies that considered
whether the rating of the teacher by the students was
higher among students whose cognitive styles are relative-
ly similar to that of the teacher or in the case of

DeLoach's study for administrator's ratings of teachers,

 

 

41

it was found that the ratings were higher When cognitive
style similarity was higher.

It appears that no research has been done that
involves the use of cognitive style similarity as an
independent variable to match people in a dyadic con—
dition. In an effort to build upon the knowledge gen—
erated by others, seven studies involving cognitive
style similarity were investigated and reported in this
review even though they did not deal with learning pairs.
These studies are interpreted as being suggestive of the
potential value that may lie in the further exploration
of how cognitive style consonance or dissonance between
people may influence the purposeful relationships of those
people. It is further suggested that the general findings
of the studies reported earlier support the notion that
cognitive style similarity should at least be entertained

as a criterion to match partners in a learning cell.

Implications for this Study

The review of the literature was organized to
address six questions. The responses to those questions
influence the study from the initial statement of hypoth-
eses and definitions to the statement of procedures.
Listed below are seven important implications provided
by the literature review.

1. The experimental treatments should conform

to the learning cell definition.

 

 

 

 

 

 

42

2. The procedures of the study should state
specifically the parameters of the learning cells in
terms of the six variables that describe differences
between learning cells.

3. Each learning cell should have a degree of
autonomy to encourage responsible participation.

4. The theoretical and practical support for the
learning cell reinforce the assumption that the learning
cell is superior to regular instruction so this study
will not use a control group.

5. An effort will be made to go beyond the learn-
ing cell literature to identify variables that may influ-
ence the learning cell since most research has treated
the learning cell as an independent variable. Age dif—
ference is offered as a researchable variable because of
the nature of the community college population. Students
may differ by as much as two generations, and a community
college instructor might profit from knowing whether the
age difference between cell mates should be considered
in forming pairs. Sex is a variable which was ignored
in the studies reviewed. Any instructor who uses learn—
ing cells must decide to stratify by sex or to ignore sex
as a variable. There is no evidence from the learning
cell literature to support either position and so deci—
sions about sex are arbitrary. The use of sex as a vari-

able in this study seems justified because there is a

43

need for research evidence upon which to base decisions
about the attention given to sex when matching students.
6. The procedures of the study should include a
precise account of how cognitive style similarity is
determined.
7. There is justification for the use of cogni-

tive style similarity as a matching variable.

CHAPTER III

METHODS AND PROCEDURES

This chapter has three major sections. The first
is Design Over Time which explains the relationship be—
tween the experimental treatments and the instrumentation.
It also includes: (1) a description of the subjects of
the study, (2) the procedures followed to assign those
subjects to the experimental treatment groups, (3) the
nature of the experimental treatment and (4) a discussion
of the instruments used in the study. The second section
of the chapter is Design Over Variables. It includes the
variable matrix of the study. The final section, Hypoth—
eses, lists the research and statistical hypotheses that
were tested and the data analysis process for each hypoth-

esis.

Design Over Time
Two designs were used in this study. The first
which is shown below accommodated one dependent variable
for achievement, performance on weekly measures of

achievement.

44

45

gel: R 01 X1 0
R - A form of random Class 1 R 01 X2
assignment R 01 X3
01 - Mid-term ______
R X 02
X1 - Self Select group Class 2 R 01 X2 02
X2 - Cognitive Style R 01 X3 02
Similarity group ______
X3 - Random group R 01 X1 0
O2 - Weekly achievement Class 3 R 01 x2 0
measures R 01 X3 0
--- Intact classes ——————
used R 01 X1 0
Class 4 R 01 x2
R 01 X3

The original plan was to use this design for both
dependent variables of achievement, weekly achievement and
final achievement. The mid-term examination was proposed
as a covariable for both of these dependent variables.
Table 1 contains data relevant to the decision to use the
covariable with each achievement measure. Since the co-
variable had a relatively high, positive correlation with
weekly achievement (R = .58) and since it accounted for
33.55% of the variance of weekly achievement, it was de-
cided to use the mid-term as a covariable with the vari-
able, weekly achievement. ANCOVA was used to analyze
those hypotheses related to weekly achievement.

The relationship between the covariable and the

final was relatively weak (R = .27). Since only 7.23%

46

of the variance of the final could be accounted for by
the covariable, it was decided not to use the mid-term
as a covariable with the variable, final achievement.
ANOVA was used to analyze those hypotheses related to

final achievement.

TABLE l.—-Relationship of Covariable to Achievement

 

 

Measures.
fi 0 I :3
-a m o H.Q :
304-) '50“! O Q)
m o $.3'3 m'H a I; 3 c
m 0 0'3 m 0'0 8 m m o
H a c-a-o a m > .4 -a -H
a 2 2:. sea as: .2
a a as: 3%. as: 8
2 > >0> >0..Q 030 Q
Weekly 68.49 46.67 33.55% .58 Use co-
variable
Final 15.09 14.36 7.23% .27 Do not
use co-
variable

 

In the design already presented subjects were not
randomly assigned to each class as indicated by the broken
lines between classes, but rather intact classes were used.
To control for the threat to internal validity due to the
selection factor a form of random assignment was used to
put the subjects into the three groups, Self Select, Cogni-
tive Style Similarity and Random which are X1, x2 and X3
respectively in the design. Analysis of Covariance was
also used to control for the selection factor with perfor-

mance on a mid—term examination being the covariable.

47

The mid-term examination is shown as 01 in the design

and occurred two weeks prior to the first learning cell
sessions. Each X stands for the four weeks of learning
cell interaction and the mean of the four weekly achieve—
ment measures is represented by 02.

One threat to internal validity was the testing
factor. The experience of taking one test might have
affected the ability of a subject to take a later test.
The mid-term examination, the covariable for ANCOVA,
could have influenced performance on the following test.
This concern for testing appeared to be slight since all
subjects responded to the same testing instruments and
so if the testing factor did exist it should have applied
to everyone. Also, the weekly measures were used during
each of the ten weeks before the study and so all subjects
were familiar with how to respond to the instruments.

A final concern for internal validity was instru-
ment decay which in this case could refer to a change in
the scoring of the essay response used by one instructor
due to fatigue or boredom or a change in the difficulty
of the multiple choice tests used by the second instruc-
tor. Since all three groups appeared in each class, it
seems likely that the effects of instrument decay would
have been evenly distributed across groups.

It appears that the chief threat to external

validity was the reactive effects of experimental

 

48

arrangements. Subjects were told that they were going

to be participants in a research study and so they could
have faked behavior to conform to imagined expectations,
but the likelihood of such behavior would seem to have
been the same for all groups. The use of regular instruc-

tors rather than strangers should have helped make the

 

setting more natural and therefore helped to control this
factor.

Another feature of the study that could have
diminished the external validity is that all three groups
occurred in each class. It was possible that the subjects'

knowing that three groups were in the class could have in-

 

fluenced their behavior apart from the actual learning

cell experiences. An attempt was made to control for this
factor by having all of the pairs formed prior to the study
so that subjects were unable to see any differences in how
subjects were matched. No distinctions were made between
the groups by the instructors because the instructors did
not know which pairs were in each group.

Rating of the learning cell experience and achieve-
ment on the final examination were the two dependent vari-
ables that the second design addressed. This design dup—
licated the first except that no observations were made
prior to the experimental treatments. The only observa-
tion for the variable, rating of the learning cell exper-

ience, was in the form of an attitude rating form that was

 

49

administered after the final learning cell session. The
final examination was the observation for the other vari-
able. The concerns for internal and external validity
that the first design suggested were the same for this
design except that testing and instrument decay were elim-

inated.

Population and Sample

 

Students in the introductory psychology course
at Oakland Community College comprised the population of
this study. The sample consisted of those students in
four sections of the course taught either by Dr. Mueller
or by Dr. Svagr during fall semester of 1974. It was
assumed that students in the four sections wererepresen-
tative of the population. There were one hundred four

subjects who participated in the study to its conclusion.

Procedures

 

During the first week of the term each instructor
informed the students in each of the four sections that
they would be participating in a study during the last
few weeks of the term. A copy of that statement appears
in Appendix A.

The following procedure was used to determine who
might become the self select pairs. After five weeks of
the term had passed, each instructor announced that the

class was to form pairs in order to work on some task

 

 

 

50

related to the course. Each instructor left the room
for five minutes and then returned. The names of each
pair were recorded and then the pairs performed the task.
During the next class meeting each subject responded to
a one item instrument which asked whether that subject
would want to work with his partner again during a later
session or if he would rather not have the same partner.
That instrument is presented in Appendix B.

Before the subjects were assigned to the three
learning cell groups in each class, the number of possible
pairs per class was determined. A triad was formed in
one class because there was an odd number of students.

The triad did not participate in the study. The number of
learning cells in each of the three matching groups for
each class was then made. An attempt was made to have
each of the matching groups equally represented in each
class and for the study as a whole. When the study began,
there were twenty-one Self Select cells, twenty Cognitive
Style Similarity pairs and twenty Random cells.

After the number of Cognitive Style Similarity
pairs in a class had been established, that many subjects
were randomly chosen from the class list. One of those
selected individuals was randomly chosen and his cognitive
style map compared to the cognitive style map of another
subject randomly drawn from the remaining pool of subjects.
The procedure for comparing the cognitive style maps of

individuals is presented in Appendix C.

 

51

If the degree of match between the two subjects
was 86% or higher, the two became learning cell mates.

If not, the subject drawn from the class pool was returned
and another randomly drawn until a partner was found for
the original subject. This process was repeated until

all of the subjects that had been identified to be in the
Cognitive Style Similarity group had partners.

Eighty-six percent had been established as a
criterion for membership in the Cognitive Style Similarity
group during a pilot study in the summer term preceding
the study. In the pilot study the cognitive style map of
each of the twenty—seven subjects was compared to the cog-
nitive style maps of all of the other subjects. A distri-
bution of the degrees of match was made and it was found
that the 86% point cut off the upper third of the distri-
bution.

After fall term had begun but before the study
started, one of the four classes in the study was randomly
chosen. The procedure that had been used during the pilot
study was repeated to make a distribution of the degrees
of match for all of the possible pairs of the thirty—one
subjects in the class. It was found that neither the
means nor the variances of the two samples (the pilot
study group and the class randomly selected from the four
classes participating in the study) were significantly

different at the .05 level. Therefore, 86% was used as

52

the criterion for determining whether a pair would be in
the Cognitive Style Similarity group in all four classes
in the study.

Those who were not placed in the Cognitive Style
Similarity group formed the other two groups. The Self
Select group was made by selecting pairs who had both
agreed to work together again from the list that had been
compiled earlier in the term. The rest of the subjects

were randomly paired and became the Random group.

Treatment

 

Each instructor had one day section that met twice
per week and one evening section that met once a week.
During the first meeting of each day section there was
a lecture for the first forty minutes followed by a thirty-
five minute learning cell session. The second meeting
began with a twenty-five minute class discussion and re-
view. Next there was a thirty minute learning cell session
and then there was a twenty minute period to take a ten
item multiple choice test covering the week's work for
Dr. Mueller's students or to write an essay applying the
week's content from the course to their own environment
for Dr. Svagr's students. Each evening section had the
same format as two consecutive daily meetings combined.

There was no required pattern of events that a
learning cell had to adopt. Three examples of learning

cell patterns were presented during the first learning

53

cell meeting, and the subjects were free to choose one
of them or to define a new pattern. The suggested learning
cell patterns appear in Appendix D.

There was a degree of content structure. Each

week each instructor distributed a study guide for the

following week's work. There were four parts: (1) Can
you define these? - (2) Can you explain these? - (3) Can
you compare these? - (4) Do you know the significance of

these? A few items that were representative of the key
points in a chapter were listed under each question.
Subjects were encouraged to add their own items to the
study guides. The study guides were used throughout the
term and they became the core of the learning cell sessions.

A sample study guide is included in Appendix E.

Instruments

 

Most of the instruments used in the study were
those already in use by the instructors. The measures of
achievement came from weekly tests and a final examination.
The pre-treatment measure of achievement came from the mid-
term examination.

One instrument was created to measure subjects'
reactions to their learning cell experiences. Prior to
the pilot study a pool of Likert-type items was generated
cooperatively with another researcher also studying the

learning cell. Some of the items were revised after

54

obtaining feedback from selected faculty members at
Michigan State University and from the pilot study sub-
jects. Ten items that dealt with general attitudes
toward the learning cell were selected for this study.
A final item was added that asked each subject to indi-
cate the predominant communication pattern of his learn-
ing cell. The instrument for rating of the learning
cell experience is presented in Appendix F.

The data for calculating cognitive style simi-
larity came from the instruments used at Oakland
Community College to measure cognitive style which had

been administered before the study began.

Design Over Variables

 

One variable matrix was used for all three
dependent variables. The three independent variables
are presented and all the groups described by these
three variables can be visualized. The distribution
of the subjects across the cells is shown. The three
dependent variables are given along with the covari-
able, the mid-term. Note that the covariable was used

only with the weekly achievement variable.

55

 

 

CO-
VARI-
INDEPENDENT VARIABLES ABLE DEPENDENT VARIABLES
LEARNING
MATCHING AGE SEX N MID- WEEKLY . FINAL CELL

METHOD DIFFERENCE FACTOR TERM ACHIEVEMENT ACHIEVEMENT RATING

 

 

 

Same 1
Wide 1
Cognitive
Different 5
Style .1
Similar- Same l0

ity Narrow

 

Different 3

 

 

 

 

 

 

 

Same 5
Wide A
Self Different 2
Select Same 8
Narrow
Different 2
Same 5
Wide
Different 1
Random
Same 7
Narrow

 

Different 3

 

 

 

 

 

 

 

 

 

Figure l.——Variable Matrix.

56

Hypotheses

The following research and statistical hypotheses

were tested.

A brief description of the statistical treat—

ment for each hypothesis is included.

H1:

The mean score for the four weekly measures
of achievement of students who are matched
by cognitive style similarity (CSS) will
exceed the mean scores of the randomly (R)
paired students and the self selected (SS)
students.

 

 

H1: xcss> XR & Xss Ho‘ XcssI XR& Xss

Analysis of H involves the use of ANCOVA

l

with the mid—term the covariable to determine

whether the three matching groups differed on the

weekly achievement measure. If ANCOVA were signif-

icant , then a post hoc analysis would be employed

to investigate the null hypothesis.

The mean score for the four weekly measures
of achievement of students whose age dif-
ference is narrow (N) will differ from the
mean score of students whose age difference
is wide (W).

Hz‘ in E 2w H0' in = 2w

involves the use of ANCOVA

 

 

 

Analysis of H2

with the mid-term the covariable to determine

whether the two age difference groups differed

on the weekly achievement measure.

The mean score for the four weekly measures
of achievement of students whose sex factor
is the same (S) will differ from the mean
score of students whose sex factor is dif-
ferent (D).

 

 

57

H3:XS#SED H:X=X
Analysis of H3 involves the use of ANCOVA with
the mid-term the covariable to determine whether
the two sex factor groups differed on the weekly
achievement measures.
H4: There will be a matching method (MM) by age

difference (AD) interaction for the depenr
dent variable, weekly achievement.

 

 

 

H : MM interacts with AD H : MM does interact
4 0 .
With AD

Analysis of H4 involves assessing the potential
interactive effects of the two variables, matching
method and age difference, regarding the weekly

achievement measure.

 

H5: There will be a matching method by sex factor
(SF) interaction for the dependent variable,
weekly achievement.

 

MM does not in-
teract with SF

H5: MM interacts with SF HO:

Analysis of H involves assessing the po-

5

tential interactive effects of the two variables,
matching method and sex factor, regarding the

weekly achievement measure.

H : The mean score for the final examination of
students who are matched by cognitive style
similarity will exceed the mean scores of
the randomly paired students and the self
selected students.

 

 

 

 

H6: xcss> XR & Xss Ho‘ xcssI XR & Xss

58

Analysis of H6 involves the use of ANOVA to

determine whether the three matching groups dif-

fered on the final achievement measure. If ANOVA

were significant, then a post hoc analysis would

be employed to investigate the null hypothesis.

The mean score for the final examination of
students whose age difference is narrow
will differ from the mean score of students
whose age difference is wide.

as”... “New

involves the use of ANOVA

 

Analysis of H7

to determine whether the two age difference groups

differed on the final achievement measure.

The mean score for the final examination of
students whose sex factor is the same will

differ from the mean score of students whose
sex factor is different.

 

 

H8: XS # XD H : X = X

Analysis of H8 involves the use of ANOVA to

determine whether the two sex factor groups dif-

fered on the final achievement measure.

There will be a matching method by age dif-
ference interaction for the dependent vari-
able, achievement on the final examination.

 

H9: MM interacts with AD H0: MM does not in-
teract with AD

Analysis of H involves assessing the potential

9

interactive effects of the two variables, matching

method and age difference, regarding the final

achievement measure.

59

H10: There will be a matching method by sex
factor interaction for the dependent vari-
able, achievement on the final examination.

 

 

H : MM interacts with SF H : MM does not

10 O n I
interact With
SF

Analysis of H involves assessing the po-

10

tential interactive effects of the two variables,
matching method and sex factor, regarding the final

achievement measure.

H11: The mean rating of the learning cell exper-
ience by students who are matched by cogni—
tive style similarity will exceed the mean
ratings by students who are randomly paired
and_by students who are self selected.

 

 

 

 

H11: Xcss> XR & xss Ho‘ XcssI XR & xss

Analysis of H11 involves the use of ANOVA to
determine whether the three matching groups dif—
fered on the rating measure. If ANOVA were signi-

ficant, then a post hoc analysis would be employed

to investigate the null hypothesis.

The mean rating of the learning cell exper—
ience by students whose age difference is
narrow will differ from the mean rating by
students whose age difference is wide.

H12: XN F Xw Ho‘ XN = Xw
Analysis of H12 involves the use of ANOVA to

determine whether the two age difference groups

12'

 

 

differed on the rating measure.

60

The mean rating of the learning cell exper-
ience by students whose sex factor is the
same will differ from the mean rating by
students whose sex factor is different.

13‘

 

x=x

H : XS # X H S D

13 D 0‘

Analysis of H13 involves the use of ANOVA

to determine whether the two sex factor groups
differed on the rating measure.
There will be a matching method by age dif-

ference interaction for the dependent vari—
able, rating of the learning cell experience.

14‘

 

 

H : MM interacts with AD H : MM does not
14 0 . .
interact With
AD

Analysis of H involves assessing the po—

14
tential interactive effects of the two variables,

matching method and age difference, regarding the

rating measure.

H There will be a matching method by sex factor
interaction for the dependent variable,

rating of the learning cell experience.

15‘

 

MM interacts with SF HO: MM does not
interact with
SF

H15:

Analysis of H involves assessing the po—

15
tential interactive effects of the two variables,
matching method and sex factor, regarding the rating

measure .

CHAPTER IV

ANALYSIS OF DATA

The purpose of this chapter is to present an an-
alysis of the data collected in this study. A total of
fifteen hypotheses were tested, all of which are presented
in the order they appeared in Chapter III. Data relevant
to each hypothesis are presented following the hypothesis.

This chapter is divided into five sections. The
first includes hypotheses and data related to weekly
achievement. Section two contains hypotheses and data
related to achieVement on the final examination. The data
and hypotheses regarding the students' evaluation of the
learning cell experience are presented in the third sec-
tion. Section four is an interpretation of the data
analysis and the final section is a summary of the chapter.

Hypotheses and Data Related
to Weekly Achievement

 

 

H ° The mean score for the four weekly measures
of achievement of students who are matched
by cognitive style similarity will exceed
the mean scores of the randomly paired
students and the self selected students.

61

 

62

Data relevant to this hypothesis are presented in
Table 2. The test of main effects for the independent
variable, matching method, produced an F-ratio of .1100
(P = .8961). Since P was not .05 or less, the null hypoth-
esis stating that the mean for the cognitive style similar—
ity group did not exceed the means of the other two match-
ing groups could not be rejected.

H : The mean score for the four weekly measures
of achievement of students whose age dif—
ference is narrow will differ from the mean
score of students whose age difference is
wide.

Data relevant to this hypothesis are presented in
Table 2. The test of main effects for the independent
variable, age difference, produced an F-ratio of .4500
(P = .5063). Since P was not .05 or less, the null hy—

pothesis stating that the means of the two age difference

groups are equal could not be rejected.

H The mean score for the four weekly measures
of achievement of students whose sex factor
is the same will differ from the mean score

of students whose sex factor is different.

3:

Data relevant to this hypothesis are presented in
Table 2. The test of main effects for the independent
variable, sex factor, produced an F-ratio of .0089
(P = .9256). Since P was not .05 or less, the null hy—
pothesis stating that the means of the two sex factor

groups are equal could not be rejected.

 

63

H4: There will be a matching method by age dif-
ference interaction for the dependent vari-
able, weekly achievement.

Data relevant to this hypothesis are presented in

Table 2. The test of the interaction of the two independent
variables, matching method and age difference, produced

an F-ratio of .1209 (P = .8865). Since P was not .05 or
less, the null hypothesis stating that there is no inter-

action between matching method and age difference could

not be rejected.

H5: There will be a matching method by sex factor
interaction for the dependent variable, weekly
achievement.

Data relevant to this hypothesis are presented in
Table 2. The test of the interaction of the two indepen—
dent variables, matching method and sex factor, produced
an F-ratio of .0310 (P = .9695). Since P was not .05 or
less, the null hypothesis stating that there is no inter-
action between matching method and sex factor could not be
rejected. I

Although no hypotheses regarding the interaction
between age difference and sex factor or between matching
method, age difference and sex factor were included in this
study, data relevant to these interactions are included
in Table 2. The test of the interaction of the two inde-
pendent variables, age difference and sex factor, produced
an F—ratio of 3.2251 (P = .0803). The test of the inter-

action of the three independent variables, matching method,

64

TABLE 2.--ANCOVA for Weekly Achievement.

 

 

Source MS DF F P Decision

End 5.1363 2 .1100 .8961 Do not reject null
Imnxnheshsl

.AD 21.0057 1 .4500 .5063 Do not reject null
IbnxthesflsZ ‘

SF .4132 l .0089 .9256 Do not reject null
innxtheaus3

MM x AD 5.6408 2 .1209 .8865 Do not reject null
hmxfihesks4

MM:x SF 1.4480 2 .0310 .9695 Do not reject null
hmxnheSESS

.AD x SF 150.5339 1 3.2251 .0803 --

MM x AD x SF 29.3440 2 .6287 .5387 --

Error 46.6157 39 - —- —-

 

(MM = Matching Method, AD = Age Difference, SF = Sex Factor)

age difference and sex factor, produced an F—ratio of
.6287 (P = .5387). Neither interaction was significant
at the .05 level.

Hypotheses and Data Related
to Final Achievement

 

H : The mean score for the final examination of
students who are matched by cognitive style
similarity will exceed the mean scores of
the randomly paired students and the self
selected students.

Data relevant to this hypotheses are presented in

Table 3. The test of main effects for the independent

65

variable, matching method, produced an F—ratio of .6358

(P = .5348). Since P was not .05 or less, the null hypoth-
esis stating that the mean for the cognitive style similar-
ity group did not exceed the means of the other two match-
ing groups could not be rejected.

H7: The mean score for the final examination of
students whose age difference is narrow Will
differ from the mean score of students whose
age difference is wide.

Data relevant to this hypothesis are presented in

Table 3. The test of main effects for the independent
variable, age difference, produced an F-ratio of .0189

(P = .8913). Since P was not .05 or less, the null hy-
pothesis stating that the means of the two age difference
groups are equal could not be rejected.

H : The mean score for the final examination of
students whose sex factor is the same will
differ from the mean score of students whose
sex factor is different.

Data relevant to this hypothesis are presented in
Table 3. The test of main effects for the independent
variable, sex factor, produced an F-ratio of .0671
(P = .7970). Since P was not .05 or less, the null hypoth—
esis stating that the means of the two sex factor groups
are equal could not be rejected.

H : There will be a matching method by age dif-

ference interaction for the dependent vari-
able, achievement on the final examination.

66

Data relevant to this hypothesis are presented in
Table 3. The test of the interaction of the two indepen-
dent variables, matching method and age difference, pro-
duced an F-ratio of .9247 (P = .4050). Since P was not
.05 or less, the null hypothesis stating that there is no
interaction between matching method and age difference
could not be rejected.

H10: There will be a matching method by sex
factor interaction for the dependent
variable, achievement on the final
examination.

Data relevant to this hypothesis are presented

in Table 3. The test of the interaction of the two in-
dependent variables, matching method and sex factor,
produced an F-ratio of 1.0358 (P = .3643). Since P was
not .05 or less, the null hypothesis stating that there
is no interaction between matching method and sex factor
could not be rejected.

Although no hypothesis regarding the interaction
between age difference and sex factor or between matching
method, age difference and sex factor were included in
this study, data relevant to these interactions are in-
cluded in Table 3. The test of the interaction of the
two independent variables, age difference and sex factor,
produced an F—ratio of 1.8687 (P = .1793). The test of
the interaction of the three independent variables, match-

ing method, age difference and sex factor, produced an

67

TABLE 3.--ANOVA for Final Achievement.

 

 

Source MS DF F p Decision

PEI 9.5967 2 .6358 .5348 Do not reject null
hmxkheaks6

.AD .2857 1 .0189 .8913 Do not reject null
hmxnmcsh37

SF 1.0131 1 .0671 .7970 Do not reject null
hwxfiheaksB

MM.x AD 13.9567 2 .9247 .4050 Do not reject null
hwxfihesh59

MMLx SF 15.6345 2 1.0358 .3643 Do not reject null
hypothesis 10

AD x SF 28.2067 1 1.8687 .1793 --

MM,x AD x SF 10.5534 2 .6992 .5030 --

Error 15.0939 40 - -- --

 

F-ratio of .6992 (P = .5030). Neither interaction was
significant at the .05 level.
Hypotheses and Data Related to

Students' Rating of the
Learning Cell Experience

 

 

 

H11: The mean rating of the learning cell exper-
ience by students who are matched by cogni-
tive style similarity Will exceed the mean
ratings by students who are randomly paired
and by students who are self selected.

Data relevant to this hypothesis are presented in

Table 4. The test of main effects for the independent
variable, matching method, produced an F-ratio of .4073

(P = .6682). Since P was not .05 or less, the null

68

hypothesis stating that the mean rating for the cognitive
style similarity group did not exceed the mean ratings
of the other two matching groups could not be rejected.

H12: The mean rating of the learning cell exper-
ience by students whose age difference is
narrow will differ from the mean rating
by students whose age difference is wide.

Data relevant to this hypothesis are presented in

Table 4. The test of main effects for the independent
variable, age difference, produced an F-ratio of 1.8900
(P = .1769). Since P was not .05 or less, the null hy-
pothesis stating that the means of the two age difference
. groups are equal could not be rejected.

H13: The mean rating of the learning cell exper-
ience by students whose sex factor is the
same will differ from the mean rating by
students whose sex factor is different.

Data relevant to this hypothesis are presented in

Table 4. The test of main effects for the independent
variable, sex factor, produced an F-ratio of .3105
(P = .5805). Since P was not .05 or less, the null hypoth-

esis stating that the means of the two sex factor groups

are equal could not be rejected.

H There will be a matching method by age dif—
ference interaction for the dependent vari-

able, rating of the learning cell experience.

14‘

Data relevant to this hypothesis are presented in
Table 4. The test of the interaction of the two independent

variables, matching method and age difference, produced an

69

F—ratio of 1.3724 (P = .2652). Since P was not .05 or
less, the null hypothesis stating that there is no inter-
action between matching method and age difference could
not be rejected.

H15: There will be a matching method by sex '
factor interaction for the dependent vari-
able, rating of the learning cell experience.

Data relevant to this hypothesis are presented in
Table 4. The test of the interaction of the two indepen-
dent variables, matching method and sex factor, produced
an F-ratio of .5365 (P = .5890). Since P was not .05 or
less, the null hypothesis stating that there is no inter-
action between matching method and sex factor could not
be rejected.

Although no hypotheses regarding the interaction
between age difference and sex factor or between matching
method, age difference and sex factor were included in
this study, data relevant to these interactions are in-
cluded in Table 4. The test of the interaction of the
two independent variables, age difference and sex factor,
produced an F-ratio of 3.3632 (P = .0742). The test of
the interaction of the three independent variables, match-
ing methOd, age difference and sex factor, produced an
F-ratio of .0810 (P = .9224). Neither interaction was

significant at the .05 level.

70

TABLE 4.--ANOVA for Rating of the Learning Cell Experience.

 

 

Source MS DF F P Decision

Bud 17.6804 2 .4073 .6682 Do not reject null
hypothesis 11

.AD 82.0310 1 1.8900 .1769 Do not reject null
hypothesis 12

SF 13.4770 1 .3105 .5805 Do not reject null
hypothesis l3

MMLx AD 59.5676 2 1.3724 .2652 Do not reject null
hypothesis l4

MMLx SF 23.2850 2 .5365 .5890 Do not reject null
hypothesis 15

AD x SF 145.9766 1 3.3632 .0742 --

MM x AD x SF 3.5164 2 .0810 .9224 --

Error 43.4037 40 —- -- --

 

Interpretation of the Data Analysis

The result of analyzing the data was that none of

the null hypotheses was rejected.

The obvious interpre-

tation is that the relationship between the independent

variable(s) and the dependent variable in a given null

hypothesis is supported by the data and that the relation-

ship described in the alternate hypothesis does not exist

under the conditions of this study.

The purpose of this

discussion is to present some factors that may help to

account for the failure to reject any null hypotheses.

71

Although there were one hundred four subjects who
finished the study, the total N was actually fifty-two.
The unit of analysis was the learning cell so the scores
for each pair were combined into a mean for each cell.
This had the effect of cutting the Npin half. In addition
nine of the original sixty-one learning cells did not com-
plete the study because at least one of the people in a
cell dropped the course or never took the final examina-
tion. The relatively small N of the study meant that
greater differences between the various groups would have
to exist to reach statistical significance than would be
needed if the N were larger. The inadequacy of the small
N appeared strongly in the analysis of the three-way
interaction. The interaction produced two cells with an
N of one, two cells with an N of two and two cells with
an N of three.

When the study was being planned, it was thought
that the four weeks of learning cell experience would be
considerable. Actually the total class time devoted to
the learning cell in those four weeks was four hours and
twenty minutes. It is suggested that perhaps this amount
of time spread over four weeks was insufficient to enable
learning cell partners to really establish productive
relationships. It is assumed that being in a learning
cell during class time was a novel experience for the

subjects. More time may have been needed to learn how

72

to function effectively in a learning cell. This point
might apply particularly to the subjects in the two
evening sections who met once a week. They had only
four class meetings that involved learning cells.

This study required that subjects who were paired
at the beginning would have to keep the same partners
throughout the study. It is certain that incompatible
pairs were formed. Such pairs may have viewed the learn—
ing cell with contempt instead of looking forward to each
session. The requirement that partners remain together
was deemed necessary because of the design of the study
but it may have been made at the expense of lowering the
quality of the learning cell experience for several sub-
jects.

There was no required pattern of communication
that the learning cells in the study were required to
follow. Three examples of learning cell patterns were
given and each pair was free to choose one of the three
or to invent a new communication pattern. The decision
to provide the options was made because it was thought
that the flexibility would make it more likely that the
subjects would willingly participate in learning cells.
There is the possibility that the different communication
patterns confounded the results by acting as uncontrolled

independent variables. The different patterns could have

73

interacted with the independent variables to cause effects
that were not investigated.

Another factor presented here that relates to the
results of the study is very fundamental. The independent
variables for the study (Matching Method, Age Difference,
Sex Factor) were all abstract concepts until definitions
for each were stated. The findings of this study relate
specifically to the relationships between the independent

variables as they are operationally defined and the de-

 

pendent variables. This point is made to suggest that
there might be other operational definitions of the ab—
stract concepts that when substituted for the independent
variables in the study would have statistically signifi—
cant relationships with the dependent variables.

There were three conditions for Matching Method:
Cognitive Style Similarity, Self Select, Random. The
cognitive style maps of two subjects had to be 86% alike
to be considered similar. Raising the criterion level
might create an operational definition for Cognitive
Style Similarity that would change the results of the
study. The formation of the Self Select group was de—
layed until the Cognitive Style Similarity group had been
made. The Random group was not formed until the other
two had been formed. The operational definitions of the
three matching groups were interdependent. Having the
matching groups formed independently could affect the

findings.

74

The mean of the distribution of the age differ-
ences of the fifty-two learning cells was used to split
all the learning cells into two groups, Wide and Narrow.
It is possible that real differences did exist between
those learning cells whose age differences were very
wide and those whose age differences were very narrow but
those differences could have been neutralized by the
learning cells whose age differences were nearer the
mean. The mean may not have adequately discriminated
the two age difference groups.

The last factor presented here that could have
influenced the results of the study concerns the reli-
ability and validity of the achievement measures. The
reliability and validity of the weekly tests and of
the final examination are unknown. It is possible that
the outcomes of the study across all the variables were

affected by these two unknown test characteristics.

Summary

This chapter contained an analysis of the data for
the study. Five research hypotheses were presented for
each of the three dependent variables, weekly achievement,
final achievement and rating of the learning cell experi-
ence. Each hypothesis was followed by a brief summary of
the data related to that hypothesis. A table containing
the data for the set of hypotheses for each dependent

variable was provided.

 

75

None of the fifteen null hypotheses could be
rejected. In addition, two additional interactions for
each dependent variable were investigated. There were
none of these six interactions which proved to be sig-
nificant at the .05 level.

Six factors that might have influenced the
results were proposed. They included: (1) the relatively
small N of the study, (2) the amount of time partners
spent working together, (3) the requirement that partners
could not change, (4) the possibility that the communica-
tion patterns acted as confounding variables, (5) the
Operational definitiOns of the independent variables,
and (6) the unknown reliability and validity of the

achievement measures.

CHAPTER V

SUMMARY, CONCLUSIONS, IMPLICATIONS

AND RECOMMENDATIONS

This final chapter contains four sections. The
first section reviews the purpose of the study and the pro-
cedures used to realize the purpose. Section two includes
the major conclusions of the study. The third section
suggests some implications resulting from the study for
those who plan to use learning cells. A statement of
recommendations for further research is presented in the

final section.

Summary

The primary purpose of this study was to compare
three methods of matching partners in a learning cell.
The three methods were to pair students whose cognitive
style maps were highly similar, to let students pair them—
selves and to pair students randomly. It was intended
that the achievement of students in the three conditions
be compared and that ratings of the learning cell exper—
ience by students in the three kinds of learning cells

be compared.

76

 

77

Secondary purposes of the study included exploring
the influence upon achievement and ratings of the learn-
ing cell experience of matching partners who were of the
same or different sex and of matching partners who were
relatively the same or different in terms of chronological
age.

Students in each of four sections of the intro-
ductory psychology course at Oakland Community College
were placed into one of three kinds of learning cells:
Cognitive Style Similarity, Self Select or Random. Each
learning cell met for sixty-five of the one hundred fifty
minutes of class time per week for four consecutive weeks
at the end of the term. Three examples of interaction
patterns for learning cells were given to all the learning
cells. A cell could choose one of the three or invent a
new pattern. A weekly study guide was provided for each
cell to help facilitate interaction with the content.

A mid—term examination, the covariable for weekly
achievement, was administered two weeks prior to the study.
The subjects took achievement tests each of the four weeks
of the study and they took a final examination. A ten
item instrument that involved the students' evaluation
of the learning cell experience was administered after the
final learning cell session. The data that were collected
using these instruments were analyzed by ANOVA and ANCOVA

with the computer program, Multivariance...Univariate and

 

78

Multivariate Analysis of Variance, Covariance and Re-

 

gression; a fortran IV program (Version 4, June, 1968),

 

on the Michigan State University CDC 6500.

Conclusions

 

Based on the analysis of data collected four major

conclusions of the study are listed below.

Conclusion 1.

 

The use of cognitive style similarity based upon
cognitive style mapping as a criterion for matching stu-
dents in learning cells was no better than the use of self
selection or randomization in terms of achievement and
rating of the learning cell experience. This conclusion
is based on the failure to reject: (l) the first null hy—
pothesis regarding matching method and weekly achievement,
(2) the sixth null hypothesis regarding matching method
and final achievement and (3) the eleventh null hypothesis
regarding matching method and rating of the learning cell

experience.

Conclusion 2.

 

The closeness of the chronological ages of cell
mates did not affect achievement or the rating of the learn-
ing cell experience by the subjects. This conclusion is
based on the failure to reject: (l) the second null hy—

pothesis regarding age difference and weekly achievement,

79

(2) the seventh null hypothesis regarding age difference
and final achievement and (3) the twelvth null hypothesis
regarding age difference and rating of the learning cell

experience.

Conclusion 3.

 

The sameness or difference of the sex of cell
mates did not affect achievement or the rating of the learn-
ing cell experience by the subjects. This conclusion is
based on the failure to reject: (l) the third null hypoth-
esis regarding sex factor and weekly achievement, (2) the
eighth null hypothesis regarding sex factor and final
achievement and (3) the thirteenth null hypothesis regard—

ing sex factor and rating of the learning cell experience.

Conclusion 4.

 

No combination of the three independent variables
in the study (Matching Method, Age Difference, Sex Factor)
affected achievement or the rating of the learning cell
experience by the subjects. This conclusion is based on
the failure to reject: (l) the fourth null hypothesis
regarding the matching method by age difference inter-
action for weekly achievement, (2) the fifth null hypoth-
esis regarding the matching method by sex factor inter-
action for weekly achievement, (3) the ninth null hypoth-
esis regarding the matching method by age difference

interaction for final achievement, (4) the tenth null

80

hypothesis regarding the matching method by sex factor
interaction for final achievement, (5) the fourteenth null
hypothesis regarding the matching method by age difference
interaction for rating of the learning cell experience and
(6) the fifteenth null hypothesis regarding the matching
method by sex factor interaction for rating of the learn-

ing cell experience.

Implications

 

No new idea can qualify for acceptance within a
given body of knowledge as the result of a single study.
Such acceptance takes a long time and follows numerous
replications of the research. Nevertheless, tentative
suggestions can be made based on a single study. The
purpose of this section is to present four such tentative
suggestions resulting from the study that the potential
user of the learning cell might consider.

1. In the delimitations of the study it was
noted that no control group representing regular instruc-
tion appeared in the study. The main intent was to com-
pare the effect of various matching variables on achieve-
ment and rating of the learning cell experience. The
findings of the study should not be construed to mean
that the learning cell itself is not an effective instruc-
tional mode. Rather, the findings support the idea that
none of the different ways of matching people influenced

achievement or rating of the learning cell experience.

 

Bl

This implies that if one were going to use learning cells
he could randomly assign partners or let them self select
without regard to their ages or sex. The use of cognitive
style similarity to match peOple would not be recommended
because the random or self select methods require much
less time and effort.

2. Both instructors reported that the subjects
responded very favorably to having the study guides avail-
able. The subjects said that they could use their time
more effectively and that they knew what the instructor
thought to be important. Perhaps other learning cell
participants would also benefit from getting study guides
from their instructors.

3.. As mentioned in Chapter IV, certain learning
cells contained incompatible partners but the design of
the study required that the partners remain together
during the study. Itjjssuggested that partners ought to
periodically be permitted to change if desired.

4. A final implication is that it may be desir-
able to use the learning cell as an alternative rather
than requiring every student to participate. Some of the
subjects resisted being in a learning cell. Some com-
plained that they could get the content on their own and
that a partner only slowed them. The learning cell, like
most every instructional mode, seems to be effective with

some students but not with others. The instructor who

 

82

plans to use learning cells might try to be sensitive to
those students who do not seem to want to work with a

partner and not force such students into learning cells.

Recommendations

 

Many endeavors that strive to find answers to
questions often result in the generation of new questions.
This Study has raised a number of questions which suggest
further directions for research dealing with the learning
cell. The following four recommendations are the result

of such questions.

Recommendation'l.

 

Further research should be done with modifications
to the independent variables of this study. The N should
be increased so that the design would require the indepen-
dent operational definitions of the three matching groups.
Consideration should be given to making the criteria for
establishing cognitive style similarity and the age dif-
ference groups more discriminating. The potential inter-
action of communication patterns and the other variables
should be investigated. It is possible that a learning
cell's communication pattern (method of working together)
may interact with the matching method, and/or age dif-
ference and/or sex factor associated with that learning
cell to affect the cell's level of achievement and rating

of the learning cell experience.

83

Recommendation 2.

 

Further research should be done to identify the
characteristics of those learners who appear to benefit
most from participating in a learning cell. Both instruc-
tors in the study agreed that certain subjects benefited
greatly from being in a learning cell while others did not.
If descriptive data regarding such things as personality,
academic ability and personal preference were available
about those students for whom the learning cell was bene-
ficial and about those for whom the learning cell was not
beneficial, one might use the learning cell more selec—

tively on a prescriptive basis.

Recommendation 3.

 

Further research should be done to identify and
describe "successful" learning cells. It would be desir—
able to generate a profile of such learning cells
including elements such as the communication patterns
and the frequency and length of sessions. The profile

could serve as a model for the users of learning cells.

Recommendation 4.

 

The learning cell in this study appeared as a
required experience replacing part of the regular class
lecture. Further research should compare the effect
of varying the context within which the learning cell

occurs. The learning cell could appear as the only

 

84

instructional mode available, as a supplement to regu-
lar instruction or as a totally extracurriculuar,
voluntary option. If it were a supplement to regular
instruction, it could be the only option (as in this

study) or one of several options.

 

BIBLIOGRAPHY

85

 

BIBLIOGRAPHY

Alexander, L. T., Gur, R., Gur. R., & Patterson, L. Peer-
assisted learning. Submitted for publication, December
12, 1973.

Amaria, R. R., Biran, L. A., & Leith, G. O. M. Individual
versus co-operative learning. I: Influence of intel-
ligence and sex. Educational Research, 1969, 11, 95-103.

 

Amaria, R. R., & Leith, G. O. M. Individual versus co-
operative learning. II: The influence of personality.
Educational Research, 1969, 11, 193-199.

 

Argyris, C. Interpersonal competence and organizational
effectiveness. Homewood, Illinois: Dorsey Press, 1962.

 

 

Blanzy, J. Cognitive style as an input to mathematics
curriculum systems and exploratory studies in the
educational sciences. Unpublished doctoral disserta-
tion, Wayne State University, 1970.

Borg, W. R., & Gall, M. D. Educational Research: An
introduction. New York: David McKay, 1971.

 

 

Bruner, J. S. On knowing: Essays for the left hand.
New York: Atheneum Press, 1965.

 

Bruner, J. S. Toward a theory of instruction. New York:
W. W. Norton, 1968.

 

Bugelski, B. R. Thegpsychology of learning applied to
teaching. Indianapolis and New York: Bobbs-Merrill,
1964.

 

 

Burton, W. H. Basic principles in a good teaching-learning
situation. In H. W. Bernard & W. C. Huckins (Eds.),
Readings in educational psychology. Cleveland and
New York: World Publishing, 1967.

 

86

87

Davis, R. H. Some psychological considerations in the
selection and design of instructional models. Paper
presented at the Joint UNESCO-IAV Research Program in
Higher Education Seminar on Methods and Programmes for
the Improvement of University Teaching, University of
Amsterdam, 1970.

DeLoach, J. F. An analysis of cognitive style disparity
as an antecedent of cognitive dissonance in instruc-
tional evaluation:' An exploratory study in the educa-
tional sciences. Unpublished doctoral dissertation,
Wayne State University, 1969.

Dewey, J. Experience and education. New York and London:
Macmillan, 1938.

Dick, W. Retention as a function of paired and individual
use of programmed instruction. The Mathematics Teacher,
1965, 58, 649—654.

Fragale, M. A pilot study of cognitive styles of selected
faculty members and students in a community college
setting. Unpublished doctoral dissertation, Wayne
State University, 1969.

Frazier, A. Learning in pairs. Grade Teacher, 1970, 81
(7), 95-96, 98, 100.

Gagne, R., and Rohwer, W. Instructional psychology. Annual
Review of Psychology, 1969, g9, 381-418.

Gartner, A., Kohler, M.C., and Riessman, F. Children teach
children: Learning by teaching. New York: Harper &
Row, 1971.

Goldschmid, M. L. Instructional options: Adapting the large
university course to individual differences. Learning
and Development, 1970, 1(5), 1-2.

 

Goldschmid, M. L. The learning cell: An instructional
innovation. Learning and Development, 1971, 2(5), 1-6.

Hapkiewicz, W. G. The application of contingency management
techniques to the teaching of teachers.' Paper presented
at the Midwestern Psychological Association Convention,
Cleveland, May, 1972.

Harrison, R. An introduction to nonverbal communication.
Prentice-Hall, 1973.

 

88

Hartley, J., & Hogarth, F. W. Programmed learning in
pairs. Educational Research, 1971, 13, 130-134.

 

Johnson, D. M. Systematic introduction to the psychology
of thinking. New York: Harper & Row, 1972.

 

 

Kingsbury, D. Project in course design. Report of the
McGill Students' Society. McGill University, 1968.

 

Lange, C. M. A study of the effects on learning of match-
ing the cognitive styles of students and instructors
in nursing education. Unpublished doctoral disserta-
tion, Michigan State University, 1972.

Mager, R. F. On the sequencing of instructional content.
Psychological Reports, 1961, 9, 405-413.

 

Mauer, D. C. Pair learning techniques in high school. Phi
Delta Kappan, 1968, 49, 609-610.

 

McGraw—Hill World News. Team learning tries out buddy
system. Nation's Schools, 1968, 81(5), 111.

 

Moore, 0. K., & Anderson, A. R. Some principles for the
design of clarifying educational environments. In D. A.
Goslin (Ed.), Handbook of socialization theory and
research. Chicago: Rand McNally, 1969.

 

Myers, K. E., Travers, R. M. W., & Sanford, M. E. Learning
and reinforcement in student pairs. Journal of Educa-
tional Psychology, 1965, 56, 67~72.

 

Oen, U. T. Memorandum to faculty of Fox Valley Technical
Institute regarding method of determining degree of cog-
nitive style match in research at that institution,
dated May 16, 1974. (a)

Oen, U. T. Memorandum to S. Spanbauer regarding prelim-
inary research findings of research at Fox Valley
Technical Institute, dated March 21, 1974. (b)

Rogers, C. R. The facilitation of significant learning.
In L. Siegel (Ed.), Instruction: Some contemporary
views. San Francisco: Chandler Publishing, 1967.

Rosenbaum, P. S. Peer-mediated instruction. New York:
Teachers College Press, 1973.

 

Schermerhorn, S. M. Learning principles of probability in
student dyads: A cross-age comparison. Unpublished
master's thesis, McGill University, 1972.

89

Schermerhorn, S. M. Peer teaching. Learning and Develop-
ment, 1973, 5(3), 1-5.

 

Schroeder, A. V. A study of the relationship between stu-
dent and teacher cognitive styles and student—derived
teacher evaluations. Unpublished doctoral dissertation,
Wayne State University, 1969.

Sheppard, W. C., & MacDermot, H. G. Design and evaluation
of a programmed course in introductory psychology.
Journal of Applied Behavior Analysis, 1970, 3, 5-11.

 

Stone, R. D. An experimental study investigating the
effectiveness of peer—assisted learning as an instruc-
tional strategy for teaching principles of learning.
Doctoral dissertation proposal, Michigan State Univer-
sity, 1974.

Thelen, H. A. Tutoring by students. The School Review,

 

Thompson, G. B. Effects of co-operation and competition
on pupil learning. Educational Research, 1972, 15, 28-36.

 

Torrance, E. P. Influence of dyadic interaction on
creative functioning. Psychological Reports, 1970,
.26, 391-394.

 

Torrance, E. P. Peer influence of preschool children's
willingness to try difficult tasks. The Journal of
Psychology, 1969, 12, 189-194.

 

 

Torrance, E. P. Stimulation, enjoyment, and originality
in dyadic creativity. Journal of Educational Psy-
chology, 1971, 62, 45-48.

Wasser, L. An investigation into cognitive style as a
facet of teachers' systems of student appraisal. Un-
published doctoral dissertation, University of Michigan,
1969.

Westmeyer, P. Some bases used by students in choosing
partners in the chemistry laboratory. The Journal of
Educational Research, 1965, 58, 355-357.

 

 

 

 

APPENDICES

90

APPENDIX A

During the last few weeks of the semester this
class will be participating in a study conducted within
the Department of Psychology at Oakland Community College.
The study is intended to examine an instructional innova—
tion known as the Learning Cell. A Learning Cell is
made-up of two students who work together to master co-
operatively some learning task. The members of this
class will be divided into pairs and part of the class
time will be devoted to Learning Cell activities. Your

cooperation in this study will be greatly appreciated.

91

APPENDIX B

 

 

 

Name

Two statements are given below. Decide which
statement you agree with and place a check (/)
in the space preceding that statement.

If our class were to work in pairs again, I
would be willing to work with my last
partner again.

If our class were to work in pairs again, I
would prefer not to work with my last
partner again.

 

92

 

 

APPENDIX C

DETERMINATION OF COGNITIVE STYLE

SIMILARITY: AN EXAMPLE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1. Display all of the cognitive style elements of the
potential pair. Show a major as a 2 and a minor as
a l.
T(AL) T(AQ) T(VL T(VQ) OM) OM CK)
Student A 2 2 1 l 1 2 -
Student B 2 2 2 l l — 2
2. Assign three points to each element present in a map
but give nine points for an L.
a. Student A has twelve elements but one of the
elements is an L worth nine points.
PointsA = 11(3) + 9 = 33 + 9 = 42
b. Student B has thirteen elements but one is an
L worth nine points.
PointsB = 12(3) + 9 = 36 + 9 = 45
3. Give the pair three points for each element for which

there is a direct match. Give nine points for an

L match.

Students A and B have a direct match for the follow-
ing elements: T(AL), T(AQ), T(VQ), Q(A), I, A, M,
D, L.

Points = 8(3) + 9 = 24 + 9 = 33

Direct

93

94

Give the pair two points for each element for which
there is a non-direct match.

Students A and B have a non—direct match for the
following elements: T(VL), R.

Points = 2(2) = 4

Non-Direct

The sum of the direct and non-direct match points
becomes the numerator of a cognitive style similarity
ratio.

The denominator of the ratio is the larger number of
points from step 2.

Numerator = 33 + 4 = 37
Denominator = 45

Make the ratio, perform the indicated division and
convert the answer to a percentage.

APPENDIX D

SUGGESTED LEARNING CELL PATTERNS

It is important that each learning cell establish

a pattern or method of working together. Four options are
explained below. Each cell should agree upon one of these
options before starting a session.

OPTIONS:

1.

One student asks his/her partner a question. The
partner responds and the questioner gives feedback.
Then the partners change roles and proceed as before.
This process is continued until both partners feel
they have mastered the content.

During the first few minutes of the learning cell
session, the partners divide the assigned content
in half. For the first half of the remaining time
one person explains the major points of his/her
assigned content and interacts with the partner for
clarification. Partners reverse their roles for
the remaining time.

Partners agree upon content that needs clarification
or review. They hold a discussion without either
partner being assigned a particular role.

Partners may invent a new pattern or adapt one of
the above.

95

 

APPENDIX E

STUDY GUIDE FOR CHAPTER 14

I. CAN YOU DEFINE THESE?

1. Four requirements of a formal psychological test

a. objectivity
b. standardization
c. reliability

d. validity
2. Mental Age
3. Chronological Age

4. Intelligence

II. CAN YOU EXPLAIN THESE?
l. The method used to compute a child's I.Q.
2. Importance of heredity
3. Importance of environment
4. Group test

5. Individual test

III. CAN YOU COMPARE THESE?
l. Aptitude tests and achievement tests
2. Stanford-Binet test and Wechsler test
3. Objective and projective tests

4. Group test and individual test

96

 

IV.

97

YOU KNOW THE SIGNIFICANCE OF THESE?

Environment on I.Q.

MMPI test

Heredity on IQ

Of the relationship of environment and heredity

General factor in intelligence

APPENDIX F

LEARNING CELL RATING FORM

Name:

 

Please be frank and honest in responding to each of the
following items. The information you provide will be the
most important factor in determining the future use of
learning cells. Please respond to every item. Your
responses will be kept confidential.

KEY: SA = you strongly agree
A = you agree
U = you are uncertain
D = you disagree
SD = you strongly disagree

1. I felt more comfortable
raising questions in my
learning cell than in
front of the whole class. SA 2: [J 1) SD

2. Learning cells should be
used in more courses. SA IX {1 I) SD

 

3. In general the learning
cell was a worthwhile
experience for me . SA A U D SD

4. The learning cell did not
improve my understanding
of the content of the
course. SA A U D SD

5. Being in a learning cell
caused me to prepare for
class more thoroughly. SA IX [1 I) SD

6. Learning was more
enjoyable because of
the learning cell. SA A U D SD

7. My learning cell met
outside of regular

class time at least
once . SA A U D SD

98

 

10.

ll.

99

The learning cell
sessions took too much
of the class time each

 

 

 

week. SA. .A U D

The learning cell activity
should have started

SD

 

 

earlier in the ternn SA. .A U D

I would avoid enrolling
in a section of a course
if the instructor planned

SD

 

 

to use learning cells. SA. .A U D

After reading the four learning cell patterns given

SD

below, indicate the one that most nearly explains how

your learning cell operated.

One student asks his/her partner a question. The

Then the partners change roles and proceed as
before. This process is continued until both
partners feel they have mastered the content.

partner responds and the questioner gives feedback.

During the first few minutes of the learning cell

session, the partners divide the assigned content

in half. For the first half of the remaining time

one person explains the major points of his/her
assigned content and interacts with the partner

for

clarification. Partners reverse their roles for

the remaining time.

Partners agree upon content that needs clarification

partner being assigned a particular role.

Other (please explain).

or review. They hold a discussion without either

 

 

 

 

 

 

 

 

“immugnm(MIMI)mama“