Self-directed learning and intellectual development : a correlation study
by DeAnna Melody Shaw

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Education
Montana State University
© Copyright by DeAnna Melody Shaw (1987)

Abstract:
The purpose of this study was to examine the relationship between self-directed learning and
intellectual development. A significant factor which influenced this research is the current influx of
adult learners in higher education. It was postulated that an analysis of self-directed learning and
intellectual development could enhance the theory base from which educators (in both adult education
and higher education) practice their profession. A packet of materials was sent to a random sample of
students attending Montana State University. The packet of materials included: a demographic
questionnaire; the Measure of Epistemological Reflection (MER) - an instrument designed to measure
intellectual development along the Perry Scheme: the Oddi Continuing Learning Inventory (OCLI) - an
instrument designed to measure self-directedness.

Results of the MER and OCLI were compared. A Pearson correlation indicated that there is a
statistically significant relationship between scores on the MER and OCLI. Age and class rank were
determined to correlate positively with each instrument. There was no statistically significant
difference between men and women on either instrument.

From the results of this study we can conclude that there is at least a tentative relationship between
self-directed learning and intellectual development. Further we can conclude that both age and class
rank are factors for both intellectual development and self-directed learning. 



SELF-DIRECTED LEARNING AND 
INTELLECTUAL DEVELOPMENT: 

A CORRELATION STUDY

by
DeAnna Melody Shaw

A thesis submitted in partial fulfillment 
of the requirements for the degree

of '
Master of Education

MONTANA STATE UNIVERSITY 
Bozeman, Montana

June 1987



MAIN UBAWf 
S jt <2/

I ii

APPROVAL

of a thesis submitted by

DeAnna Melody Shaw

This thesis has been read by each member of the 
thesis committee and has been found to be satisfactory 
regarding content, English usage, format, citations, 
bibliographic style, and consistency, and is ready for 
submission to the College of Graduate Studies.

Chairperson, Graduate 
CommitteeDate

Approved for the Major Department

Date Head, Major Department

Approved for the College of Graduate Studies

A
Graduate DeanDate



iii

STATEMENT.OF PERMISSION TO USE

In presenting this thesis in partial fulfillment 
of the requirements for a master's degree at Montana 
State University, I agree that the Library shall make 
it available to borrowers under rules of the Library. 
Brief quotations from this thesis are allowable without 
special permission, provided that accurate 
acknowledgment of source is made.

Permission for extensive quotation from or 
reproduction of this thesis may be granted by my major 
professor, or in his absence, by the Director of 
Libraries when, in the opinion of either, the proposed 
use of the material is for scholarly purposes. Any 
copying or use of the material in this thesis for 
financial gain shall not be allowed without my written 
permission.

Signature



iv

ACKNOWLEDGMENTS

Many individuals were" responsible for enabling me 
to "climb this mountain." I would like to take this 
opportunity to acknowledge their immeasureable 
assistance.

Dr. Ralph Brockett, without whom this thesis would 
never have been completed. His tireless patience 
throughout this project was greatly appreciated.

Dr. William Porterfield, whose advise and personal 
support inspired much of this work.

Dr. Michael Waldo, who was willing to take on the 
responsibilities of being a member of my committee lit­
erally minutes after arriving on campus.

Jo and Butch Nelson, whose establishment, the Leaf 
and Bean, provided a tranquil writing environment.

Garnet Shaw, my mother, and to the rest of my 
family for their emotional support.

Tony Berget, my husband-to-be, for the use of the 
computer; for making phone calls; for taking me out to 
dinner when the going got rough; for his constant love 
and support.



V

TABLE OF CONTENTS

Page
LIST OF TABLES............................. vii
ABSTRACT . ..... '...........■..................... viii

1 . INTRODUCTION .............................. •.......... I
Overview of Self-Directed Learning.............. 2
Intellectual Development and the Perry Scheme... 7 
Statement of the Problem and Research

Hypotheses................    11
Signficance of the Research Problem..-.......... 13
Assumptions and Limitations ......................15

2. REVIEW OF THE LITERATURE............. 17
Cognitive Developmental Theory and -

Intellectual Development...........'..........18
The Perry Scheme:

A Cognitive Developmental Theory........... 19
Assessment Along the Perry Scheme. .. ......... 22

Self-Directed Learning Theory and Research.....27
Measurement of Self-Directed Learning.... ..38

SeIf-Directed Learning and Intellectual
Development...........    41

Summary. .......... ................ '.............. 45
3. METHODOLOGY............................ 47

Population and Sample......................... '. .47
Instrumentation and Scoring..................... 50

The Measure of Epistemological Reflection....50
The Oddi Continuing Learning Inventory.......52

Research Design and Data Collection Procedure..54 .
Hypotheses...............      55
Summary.............. ,........................... 56

4. RESULTS....... ...................... ............: .58
Characteristics of the Sample........ . i .. 58
The Measure of Epistemological Reflection......62

Interrater Reliability for the MER........... 65
The Oddi Continuing Learning Inventory. .........66



vi
■ TABLE OF CONTENTS--Continued

Page
Results of the Hypotheses.............. . 66

Relationships Between the MER and OCLI..... 68
Age-Related Findings for the MER and OCLI... 68 
Gender-Related Findings for the MER
and OCLI................... '............ . 71

Relationships Between the MER, OCLI, and
Class Rank................  72

Interaction of Age and Class on the MER
and OCLI....... ................... ■......... 74

Summary....... ...........................'..... 75
5. DISCUSSION OF THE RESULTS...,....... '............76

Results Related to the MER..... 1...............76
Results Related to the OCLI.... ................ 80
Correlation Between Intellectual Development ■

and Se If-Directed Learning....... '.......... 81
Significance and Limitations of Applying

Results to Practice.......................... 83
Suggestions for Future Research....-........... 86
Summary. . . ..........................    .87

REFERENCES CITED .;...!................. '........ . 88
APPENDICES..........................  93

Appendix A - Sample Research Packet.......... .94
Letter to Participants....................... 95
Instructions........  96
Consent Form. ..................................97
Demographics Questionnaire..............    98
The Measure of Epistemological Ref lection... 99
The Oddi Continuing Learning Inventory.....106

Appendix B - Certification to Rate MER....... 109



vii

LIST OF TABLES

Table Page
1 ' Intellectual Development by .—  r - :— r- 7  • :--- - Ti

Curriculum.................. ............. . 43
2 Characteristics of Sample Population......... 59
3 Numerical Results for the MER. ................ 63
4 Positioning Results for the MER...............64
5 Results Related to the OCLI.............   67
6 Pearson Product Moment Correlation for

Scores on the MERf OCLI and Age.............  70
7 Differences for Individuals Under and ■

Over the Age of Twenty-five on the MER
and OCLI................. ...................... 70

8 Gender-Related Differences for Scores
on the MER and OCLI....................   72

9 Relationship Between Class Rank and
Scores on the MER and OCLI....... ........... 73

10 Differences Between Class Rank and
the MER................  ...... ................73

11 Differences Between Class Rank and
the OCLI........................................73

12 ANOVA Results for Interaction of Age and
Class on. the MER and OCLI..................... 75



viii

ABSTRACT

The purpose of this study was to examine the 
relationship between self-directed learning and 
intellectual development. A significant factor which 
influenced this- research is the current influx of. adult 
learners in higher education. It was postulated that 
an analysis of self-directed learning and intellectual 
development could enhance the theory base-from which 
educators (in both adult education and higher 
education) practice their profession. A packet of 
materials was sent to a random sample of students 
attending Montana State University. The packet of 
materials included: a demographic questionnaire; the
Measure of Epistemological Reflection (MER) - an 
instrument designed to measure intellectual development 
along the Perry Scheme: the Oddi Continuing Learning 
Inventory (OCLI) - an instrument designed to measure 
self-directedness.

Results of the MER and OCLI were compared. A 
Pearson correlation indicated that there is a 
statistically significant relationship between scores 
on the MER and OCLI. Age and class rank were 
determined to' correlate positively with each 
instrument. There was no statistically significant 
difference between men and women on either instrument.

From the results of this study we can conclude 
that there is at least a tentative relationship between 
self-directed learning and intellectual development. 
Further we can conclude that both age and class rank 
are'factors for both intellectual development and 
self-directed learning.



I

CHAPTER I 

INTRODUCTION

Adults older than the traditional age of college 
students (18-22) are returning to formal learning 
situations with rapidly increasing frequency.
Currently, 43 percent of students on our college and 
university campuses are over the age of 25 ("What 
They're Reading," 1985) . This increase in older 
students impacts our institutions of higher education. 
Some of the obvious areas of impact can be seen in the 
need for family housing, childcare, and special support 
Organizations which provide a social network. Academic 
assistance programs for the non-traditional student are 
also common on campuses. Not so apparent, however, is 
the difference in adult learning styles which necessi­
tates the need for a difference in the instructional 
method. The field of adult education is rooted in the 
conceptual and theoretical assumptions that adults 
learn differently from children or pre-adults. 
Theoretically, one way of considering the underpinnings 
of adult education can be found in the comparison of 
the pedagogical and andragogical models of education as 
described by Knowles (1985). The pedagogical model



2
assumes education to be teacher-centered, and that the 
learner is by and large a passive recipient of 
knowledge. Students enter into a learning situation 
motivated by external pressures of society, parents, 
teachers, competition for grades, and/or fear of 
failure. The andragogical model, however, assumes that 
learning is student-centered. The learner chooses the 
direction his or her education will take based on 
his/her life experiences and circumstance (Mezirow, 
1981; Spear and Mocker, 1984). The student has chosen 
to enter into the learning situation, and is motivated 
by a desire to address a problem or issue in his/her 
life and. by the assumption that the knowledge will 
enhance the quality of his/her life.

Given the andragogical model as a definition of 
adult learning, it would be easy to make the assump­
tions that all adults are self-directed in their 
learning, or that only adults are self-directed in ■ 
their learning. The following discussion on self- 
directed learning (and the findings of this study) 
should serve to clarify why these assumptions may be 
innaccurate when applied to the population as a whole.

Overview of Self-Directed Learning

Self-directed learning is an illusive concept be­
cause it has been given.many descriptive definitions.



3
but .little in the way of a theoretical focus. Is self- 
directed learning a personality construct, a learning 
style or a method of instruction? If it is an instruc­
tional method, are age, maturity, and intellectual 
development' factors for the appropriateness of this . • ■
particular method? Is it influenced by socio-economic 
factors?

There exists a multitude of definitions to 
describe self-directed learning. These definitions 
take their place on a continuum that has as .one extreme 
the notion that self-directed learning is a change in 
perspective or awareness (Mezirow, 1981; Tough, 1979)’; 
the other extreme of the continuum would suggest' that 
self-directed learning exists within a more rigid set 
of parameters (Penland, 1981). Knowles (1975) offers a 
more specific and less extreme definition of self- 
directed learning.

...'seIf-directed learning' describes a process in 
which individuals take the initiative, with or 
without the help of others, in diagnosing their 
learning needs, formulating learning goals,

. identifying human and material resources for . 
learning, choosing strategies, and evaluating 
learning outcomes, (p. 18)
This definition, while broad enough to.satisfy 

most adult educators, is merely descriptive of a self- 
directed learning process. It does not address why 
individuals choose this process. It does not explain 
what types of individuals choose this method nor does



4
it address the relative effectiveness of this learning 
style. The lack of a theoretical base to explain the ■ 
how's and why's of self-directedness is a concern of 
many researchers (e .g . , Gibbons f 1980; Mezirow, 1981; 
Penland, 1981; Cross, 1981; Oddi, 198.3).

The definitions which follow are related to the 
concept of self-directed learning and are presented as 
these terms are interpreted for this study.

A self-directed learner (for purposes of this 
study) is someone who possesses the personality traits 
which distinguish a self-directed learner from an 
other-directed learner.

An adult (for purposes of this study) is someone 
who is 25 chronological years of age or older. The age 
of 25 is an arbitrary number chosen to help define the 
population for this study. A person's individual 
experiences and maturity rather than his/her chrono­
logical age are believed to be a greater influence as 
to whether or not he/she is "adult" in his/her world 
perspective.

Pedagogy is'a process of education which is 
teacher-centered; the learner is a passive recipient of 
knowledge.

Andragogy is a process of education which is 
student-centered; the learner is responsible for



5
initiating, following through with, and evaluating 
his/her own learning/educational situation.

Thus far, self-directed learning as a concept with 
theoretical underpinnings is an enigma to the field of 
education because it is so many things to so many dif­
ferent individuals. Therefore one might speculate its 
relevance has not been proportional to the amount of 
attention it. has received. To individuals who take 
this view, the many definitions of self-directed 
learning suggest that it is merely a conceptual idea to 
be recognized, to be made aware of, but that the 
relevancy ends there with many descriptions of many 
instructional methods. Research can help to develop a 
theoretical foundation which in turn will establish 
self-directed learning as an instructional method.'

Self-directed learning can be an avenue for many 
people to get the kind of education they would like to 
have to enhance their lives, but might not access 
otherwise. Instead of guarding the educational process 
as if it had some mystical power, educators should be 
allowing education its own autonomy to empower the 
individual. The mystique surrounding education must be 
removed without removing this ability to empower. We 
are not so far removed from Paulo Freire's (1970) 
description in Pedagogy of the Oppressed of teachers, 
the oppressors, who pour the desired knowledge into the



6
receptacles who are their students, the oppressed.
This method distorts the world to be a set of absolutes 
waiting to be learned.

Researchers in education have failed,. thus far, to 
establish a strong theoretical base for self-directed 
learning; although some have attempted the development 
of theory (Mezirow, 1981; Penland 1981; Oddi, 1983); 
This failure has been a disservice to the field in a_ 
number of ways. Without a theory, any definition of 
self-directedness is tenuous and at best merely 
educated conjecture. Without theory, assessment of the
self-directed learner, or those projects which are.....-
considered self-directed learning is very difficult.
It is difficult to qualify something without a standard 
of measurement. Finally, without theory, research into 
the implications and application of self-directed 
learning is little more than an educated assumption.

The difficulty in establishing a theoretical base 
for self-directed learning is multi-faceted. There are 
some (e.g ., Oddi, 1985) who would theorize self- 
directed learning as primarily a personality construct. 
A person possesses more or less of the personality 
traits that allow him/her to be self-directed in 
his/her learning. Another school of thought (Spear and 
Mocker, 1984; Brookfield, 1984) suggests that a 
person's ability to be self-directed is influenced



7
primarily by his/her socio-economic circumstances.
This is a consideration which may be substantially 
supported by Maslow's (1970) hierarchy of needs. A ' 
third body of literature views self-directed learning 
as an instructional method related directly to . 
intellectual development. This study examines 
self-directed learning as a personality related concept 
and how this aspect of self-directed, learning is 
correlated to cognitive development. The following 
section will discuss intellectual developmental theory 
in more depth.

Intellectual Development and the Perry Scheme

How do adults make meaning out of their world 
around them? Developmental theory attempts to address 
this question. Adult developmental theory can be 
broken down into three sub-categories: typological
theory, psycho-social theory and cognitive develop­
mental theory. Typological theories concern 
temperament and socialization. The Myers-Briggs Type 
Indicator (Myers, 1980) and the work of Carl Jung 
(Jung, 1964) are two'examples of typological theory. 
Psychosocial theory describes how an individual reacts 
to and interacts with his environment. Chickering's 
vectors of development discusses this type of inter­
action (Chickering, 1972). Finally cognitive '



8
developmental theory describes the way we view our 
world and the way we reason about this world.
Kohlberg7 s theory of moral development (Kohlbergf 1980) 
and Perry's theory of intellectual development (Perry, 
1970) are examples of cognitive developmental theories.

This research study is concerned with this last 
sub-category of developmental theory. Cognitive 
developmental theory can be described by a number of 
characteristics. The following definitions are offered 
to further explicate the concepts of cognitive or 
intellectual theory.

Cognitive is a term used to describe the processes 
of thought - how we think, reason, understand. For 
purposes' of this study, cognitive and intellectual will 
often be used interchangeably (except where more 
specifically defined by the text).

Development suggests a progressive movement 
towards the more inclusive or complex.

Stage or position describes a particular set of 
assumptions about how we come to "know" our world. 
Stages or positions are hierarchical, sequential, they 
describe "how" we see not "what" we see, and finally 
each stage is qualitatively different from the 
preceding stage. What occurs before and after a stage 
or position could be considered periods of transition.



9
Growth describes, in general, progress in develop­

ment. It is generally valued positively. That is, 
growth is traditionally viewed as more- productive for 
the individual than is non-growth. Stages or positions 
are one way to delineate growth.

Experiential dissonance is an experience which 
when presented to an individual should cause a dilemma. 
The.dilemma is effectively solved when the individual's 
reasoning structure changes. Dissonance is used to 
promote growth in development.

Accommodation and assimilation are the two methods 
individual's use to make meaning from the experiential- 
dissonance. Assimilation is the process of trying to 
"fit" the dissonance caused by the experience into, 
current reasoning patterns.. This can be through 
selection, simplification, or distortion.
Accommodation is the process' of changing the reasoning 
patterns to "fit" the dissonance caused by the 
experience. This can occur through transforming or 
recombining previous reasoning structures to 
accommodate the new experience.

Lifespan development is a concept which suggests 
that development occurs from the moment of birth to the 
moment of death.

Student developmental theory looks specifically at 
the' growth in development during the college years.



10
Student development can be seen as a period of intense 
development in the course of an individual's lifespan 
development.

Often developmental theories are represented by a 
series of sequential stages in which a person's world 
view or perspective is enhanced by a series of experi­
ences which cause a degree of cognitive dissonance.
This dissonance challenges a person to consider more ■ 
options or different types of options in their problem 
solving.' The dissonance leads a person from one stage 
or position into the next. Because intellectual 
development is sequential in nature, each subsequent 
stage encompasses the perspective of the previous 
stage. Further, these stages are' hierarchical and ■ 
sequential and for the most part irreversible. Once an 
individual has fully established a new perspective and 
way of reasoning it is difficult for him/her to return 
to a more limited perspective and reasoning structure.

The Perry Scheme, published under the title.
Forms of Intellectual and Ethical Development in the 
College Years: A Scheme (Perry, 1970) is one theory
which describes the intellectual development of adults. 
The Bureau of Study Counsel at Harvard University, 
under the direction of William Perry, spent twenty 
years examining the'responses of college undergraduates 
to their learning environment. What emerged from this



11
study was a theoretical description of the develop­
mental stages individuals pass through during their 
college experience. There are nine positions on the 
Perry Scheme. A more comprehensive discussion of the 
Perry Scheme and each of its "positions" is offered in 
Chapter 2.

An understanding of how students reason about 
their world can inform educators about how to present 
information, interact with students, and create the 
most conducive learning environments. For example, the 
Perry Scheme examines how individuals view authority 
(the instructor), the nature of truth, and peers. It 
describes the decision-making process individuals 
progress through and how individuals view evaluation. 
With this information, instructors may more adequately 
challenge and support their students. This challenge 
and support can occur through learning activities per 
se or through the discussions about the learning 
activities.

Statement of the Problem and Research Hypotheses

Is intellectual development related to self- 
directed learning? Can we make assumptions regarding a 
person's ability/preference for self-directed learning 
based on their level of intellectual development? If 
individuals progress in intellectual development can



12
their ability/preference for self-directed learning 
also progress? - Can and should educators use this 
knowledge to enhance an individual's ability/preference 
for self-directed learning? These are only a few of 
the issues to consider when examining the relationship 
between cognitive development and self-directed 
learning. The purpose of this study is to empirically 
examine this relationship. It is not expected that 
this study will provide conclusive answers; rather, it 
is expected that this project will provide some useful 
data regarding the relationship between intellectual 
development and self-directed learning - data that will 
ultimately create more questions for investigation.

This study examines the relationship between a 
person's ability/preference for self-directed learning 
and that person's intellectual development as described 
by the Perry scheme. There are nine operational 
hypotheses tested, statistically, by this study. These 
are specifically presented in Chapter 3. It is' 
expected that a statistically significant correlation 
will be found between a person's ability/preference for 
self-directed learning as measured by the Oddi Continu­
ing Learning Inventory (Oddi, 1984) and a person's 
position on the Perry Scheme of Intellectual Develop­
ment as measured by the Measure of Epistemological 
Reflection (Baxter-Magolda and Porterfield, 1985).



13
Leeb (1983) found "no statistically significant 
associations" (p. 192) between positioning on the Perry 
Scheme (as measured by an essay protocol) and self- 
directed learning [as measured by the Self-Directed 
Learning Readiness Scale (Guglielmino, 1977)]; however, 
Leeb herself noted that her findings should be con­
sidered speculative as they were derived from 
tangential data from her study. This study will be 
considered in greater depth in Chapter 2.

Significance of the Research Problem

The significance of this study will be examined by 
considering some factors involved with this study.
These factors include: the relationship between
chronological age and ability/preference for self- 
directed learning and/or intellectual development; 
differences among individuals at different levels of 
educational attainment; and the differences.between the 
results for men and for women.

The purpose of this'study is to gather additional 
empirical data on self-directed learning and its rela­
tionship to intellectual development. To this point in 
time, much conjecture has been posited to explain and 
define self-directed learning. It is this author's 
belief that only through empirical data collection, 
will the field of adult education be able to form a



14
more homogeneous concept of self-directed learning and 
the factors affecting it.

The significance of this study is derived from the 
quality of data produced. This data provides informa­
tion to help test, empirically, a theoretical frame-. 
work, and to clarify the concept of self-directed 
learning. A "theory" per se has not been generated by 
the data of this study; however,, the data contributes 
to a body of knowledge that can be used to more clearly 
define theory; Another significant factor to consider 
in this study is whether or not self-directed learning 
is a developmental concept. It was presumed on the 
outset of this study that a positive relationship found, 
between self-directed learning and intellectual 
development could improve the instructional method by 
helping instructors to develop the self-directed 
learning tendency in the individuals they instruct. In 
addition, this study should help clarify for 
practitioners; in student affairs-related occupations, 
how they can better serve their new clientele - the 
older student. The information gained from this study- 
will also help clarify the findings of Leebf s study 
(1983). Finally, and from a pragmatic view, this study 
will help add to the validation of both the MER and
OCLI .■



15

Assumptions and Limitations

The construction of this research study has 
inherent assumptions. Primarily, it is assumed that 
self-directed learning has a developmental property. 
This developmental property is the degree to which 
persons are able to and choose to practice self- 
directed learning. •Self-directed learning is not 
viewed as a static method of learning, but rather on a 
continuum of personal involvement. Secondly> it is 
assumed that the relationship between self-directed 
learning and intellectual development can be 
quantified, and.this is reflected by the scores on the 
OCLI and the MER.

A major limitation of this study concerns the 
demographics of the population from which the sample is 
drawn. This study was conducted at Montana State 
University in Bozeman, Montana, and the sample, 
regardless of the random selection process, was fairly 
homogeneous. Statistically the largest portion of the 
university's population is from small towns in Montana 
(Astin.and Groseth, 1984). The population of students 
attending Montana State University could be described 
as largely white, middle class, rural individuals. .

Another possible limitation of this study is rela­
tive newness of both' instruments. There are not, at



16
this time, a .large number of published reports using 
either of these instruments; therefore, it is difficult 
to generalize about the validity of the instrument from 
study to study. There are specific limitations 
inherent in both instruments which will be more fully 
discussed in Chapter 3. There are also specific 
reasons for why these instruments were chosen, 
including cost of administration, validity and 
reliability,- ease of administration, and the author's 
familiarity with the instruments.



17

CHAPTER 2

REVIEW OF THE LITERATURE

This chapter will examine the literature base of 
cognitive developmental theory.and self-directed 
learning that is most relevant to this study. The 
literature related to cognitive development, particu­
larly as reflected by the Perry Scheme is quite 
extensive as researchers have investigated, many 
aspects of the theory (e .g ., Knefelkamp, Widick, and 
Parker, 1978; Knefelkamp, 1981; Baxter-Magolda and 
Porterfield, 1985) . The literature related to self- 
directed learning is also quite extensive (e.g ., 
Knowles,' 1975; Tough, 1979; Gibbons, 1980; Cross, 1981; 
Mezirow, 1978; Brockett,, 1983; Oddi, 1983; Leeb, 1983; 
Brookfield, 1984). One reason for this recent interest 
in adult education theory is the influx of 
non-traditional students (i.e ., students over the age 
of 25) on college campuses. Little research, however, 
has been conducted regarding the relationship between 
cognitive developmental theory and self-directed 
learning. Therefore, the literature base that overlaps 
both fields of study is limited.



• ' .18
The purpose of this chapter is to present selected 

literature on. both concepts so that.the question of the 
relationship between the two can be more fully 
explored. Cognitive developmental theory and the Perry 
scheme of intellectual development will be considered 
first. In addition to the discussion of theory, an 
analysis of assessment issues for intellectual 
development and.the instrumentation used will be 
presented. The second section of this literature 
review concerns self-directed learning's conceptual■ 
framework. Literature relevant to the research and 
theories of self-directed learning will be reviewed, as 
well as of the instrumentation for self-direct.edness. 
Finally, the.literature which explores the relationship 
between cognitive development and self-directed 
learning will be reviewed.

Cognitive Developmental Theory 
and Intellectual Development

Cognitive developmental theory suggests that 
individuals will change their patterns of reasoning.
It suggests that as current reasoning patterns cease to 
assimilate an issue or event (and making meaning from 
the experience is deemed necessary), then an 
individual's patterns of reasoning will expand to fit 
the experience. A classic example might be the child,



19
who formerly considered his parents to be infallible, 
suddenly realizes that a parent has made a mistake. 
Under the previous reasoning structure the mistake 
would never have been seen, but now the child is con­
fronted with the idea that perhaps this parent isn't 
infallible. The child has two choices: I) to ignore 
the mistake and continue to believe that the parent is 
infallible; 2) to consider the idea that, at least in 
this situation, the parent is fallible. The second 
choice would be a step towards growth in his/her 
development since the reasoning about parents/authority 

- change to explain the experience.

The Perry Scheme: A Cognitive Developmental Theory
The Bureau of Study Counsel at Harvard University, 

under the direction of William Perry, spent twenty 
years examining the reasoning of college undergraduates 
about their learning experiences. What emerged from 
this study was a theoretical description of the 
developmental stages individuals pass through to gain a 
more inclusive perspective of their world. This theory 
was published under the title Forms of Intellectual and 
Ethical Development In The College Years: A Scheme
.(Perry, 1970) but has since been more commonly referred 
to as "the Perry scheme". There are nine positions on 
the Perry scheme which represent growth in



20
development and can be divided under three broad 
categories: dualism, relativism, and commitment.

The first three positions of the Perry scheme fall 
■ under the category of "dualism". A person at the first 
position, "basic duality", views the world in terms of 
absolutes. The authorities have "the right" answer. 
Multiplicity is not perceived at this position. A 
person at the second position, "multiplicity pre­
legitimate", perceives multiple options; however, only 
one authority or option is considered to be valid. The 
third stage is entitled "multiplicity subordinate". At
this stage, a person recognizes that the authorities___
may not always have "the" answer. The right answer may 
be unattainable, but this is only a temporary situa­
tion. The right answer may be known at some point in 
the future.

Positions 4a, 4b, 5, and 6 all belong to the 
category entitled "relativism". A person at position 4 
may relate his/her experience in either or both of two 
ways. Position 4a, "multiplicity correlate", acknow­
ledges that when there are no absolutes, everyone has a 
right to his/her own opinion. Position 4b, "relativism 
subordinate", acknowledges that authorities can have 
multiple opinions/views because they want others to 
think in terms of multiple options. At position 4 a 
person can choose to be either adhering or



21
oppositional in his/her opinions. An adhering 4 is 
more likely to accept any idea or perception as equally 
accurate. An oppositional 4 is more likely to argue 
that regardless of what his/her opinion is, that • '
opinion is just as valid as anyone else's opinion. At 
position 5, a person moves into the "relativism 
correlate" position. Relativism is viewed not in terms 
of what the authority wants for us, but rather how the 
world and authorities really are. At position 5 a 
person views everything as relevant within a context. 
Each person's opinion may be relevant in the larger 
picture, but may not be relevant" in a ""particular 
context. At position 6, "commitment foreseen", a 
person begins to recognize the need for commitments in 
a relativistic world.

The final category of positions is the 
"commitments" category. Position 7,. "initial commit­
ment", is the stage at which a.person first actualizes 
and examines the implications of the first commitment. 
At position 8, "orientation in implications of commit­
ment", the complete implication of commitments is 
realized. Commitments need to be balanced, which lead 
a person to position 9, "developing commitments". At 
position 9, commitments and their full implications 
become a considered part of a person's reasoning and 
decision making process. There are some theorists



22
(e.g ., Kitchner and King, 1981; Baxter-Magolda and 
Porterfield, 1985) who would argue that no develop­
mental change occurs past position '5. This argument is 
based on the logic that no qualitative differences 
occur after position 5 or between positions 6,7,8, and 
9. In this' view positions 6-9 are seen as a process 
that reoccurs throughout adulthood.

• A person's perspective and reasoning structure is 
fairly consistent from subject to subject. How a 
person views authority, for example, should be fairly 
close in reasoning structure to how that person views 
the nature of truth. If a person has a dualistic view 
of God and human nature, he/she will probably have a 
dualistic, right/wrong, view of the nature of truth.

Assessment-Along the Perry Scheme.
The Perry Scheme was developed through the 

analysis of a longitudinal study which used an essay 
and interview format. Assessment of positioning along 
the Perry scheme can occur through an interview, an 
essay, a recognition instrument, or a short answer 
questionnaire. Assessment through interviews requires 
that a trained investigator ask a series of questions 
which are designed to establish that person's reasoning 
structure. Each question is followed by a series of 
follow-up questions which probe deeper into the 
response. The essay format is a standardized, written



23
prototype of the interview format. A recognition 
instrument asks the respondent to identify a 
perspective most like his/her own from a list of 
possibilities. • Each of these formats have been used to 
assess positioning along the Perry scheme.

More recently, Taylor and Porterfield (1983) have 
developed a short answer questionnaire as an assessment 
tool.- In a recent article (1985) they addressed some 
of the difficulties inherent to the other methods of 
assessment. These difficulties include:
1 . Although an interview technique is the most 
comprehensive form of assessment - it is the most time 
consuming method and requires highly trained 
interviewers and scorers. ■
2. Research by Rest (cited in Baxter-Magolda and
Porterfield, 1985) indicates that respondents may be'
able to recognize and prefer a cognitive stage which, is 
higher than the one they can produce (and presumably 
the one they can produce is their position).
Therefore, a .simple recognition instrument may be more 
likely to reflect where an individual would prefer to 
be rather than where he/she actually is.
3. There is a need to distinguish between content and 
justification for the response. Because the content 
can be the same for several stages on the scheme, the 
"why" behind the content is more enlightening. The



24
"why" is not reflected in a response format 
questionnaire.

With these difficulties in mind, Baxter-Magplda 
and Porterfield (1985) developed, an instrument entitled 
the "Measure of Epistemological Reflection" (MER)..
This instrument was developed, according to the 
authors, "to provide specific stimuli.and a standard 
scoring procedure to reduce the degree of inference 
necessary to assign a Perry level" (p. 343).

The MER is a short-answer response format . 
questionnaire. There are six domains which are rated 
separately: decision making, role of the learner,, role
of the instructor, role of peers, evaluation, and view 
of knowledge, truth, or reality. Individuals can be 
rated differently within the different domains. For 
example, a person may generate a position 2 response to 
the role of the instructor and a position 3 response to 
decision making. However, individuals will rarely vary 
more than one position level from domain to domain.
The total protocol rating (TPR) is obtained by looking 
at all 6 domains. .If the same position rating shows up 
in at least 2 of the domains, that position is 
reflected in the TPR. For example, if an individual 
has a position 4 rating on 2 of the domains and a 
position 5 rating on 3 of the domains and a position 3 
rating on I' domain, the TPR would be shown as 5(4) . In



25
this case 5 is the dominant position, but a tendency is 
still evident for position 4. (The position 3 rating 
is not reflected in the TPR because it occured only 
once.) This TPR suggests that this individual has just 
made the transition from position 4 to position 5 
reasoning. Position ratings that are equally distri­
buted among the domains are shown as 2-3. This TPR 
suggests that an individual is beginning to make the 
transition from position 2 reasoning to position 3 
reasoning, but has not fully moved into position 3.
The MER is based on the premise that no significant 
developmental changes occur past position 5. There­
fore, the highest positioning assigned through the MER 
assessment is 5.

The MER requires rating by certified raters. To 
be certified an individual must rate protocols for 20 
individuals and must obtain approximately a .80 corre­
lation on the TPRs with the expert raters. Reliability 
of the MER was tested for interrater agreement between 
highly trained raters, an expert on the Perry Scheme, 
and two self-trained raters. Comparison with the 
Measure of Intellectual Development (MID), a similar 
instrument measuring intellectual development along the 
Perry scheme (Kneflekamp, 1974; Widick, 1975 as cited 
in Baxter-Magolda and Porterfield, 1985.) was also



26
conducted to test the correlation between the two 
measurement tools.

In one research project (Baxter-Magolda and 
Porterfield, 1985), 155 participants (79 men and 76 
women) were sampled from four groups including college 
freshmen, seniors, graduate students under the age of 
30, and graduate students over the age of 30. These 
participants had been randomly selected (except for the 
freshmen who were students in a psychology class) from 
students at a large state university, in 1982. The 
groups were fairly homogeneous across socio-economic 
indicators. ......-

The MER and the MID were given to the sample 
group. The order of the administration of the instru­
ments was reversed for half of the respondents to avoid 
bias in administration of the instruments. Reliability 
of the MER was assessed internally by comparing the 
consistency between the raters. The MER was compared 
to the MID to establish concurrent validity.

The results of this initial study showed that the 
internal validity among the raters was quite high. 
However, there seemed to be quite a discrepancy between 
the MER and the MID. One reason cited for this low 
correlation was the lack of similar variability in the 
MID scores. Also the MER, according to the authors.



27
generated a greater amount of data. The difference in 
the type of response elicited could account for the low 
correlation. The authors cite four follow-up studies 
which they feel substantiate the validity of the MER. 
They also recognize the need for a longitudinal study 
to further their claims of validity for the MER.

The study of intellectual development and. applying 
the.Perry scheme to the practices of both adult educa­
tion and higher education is another avenue educators 
can use to increase their effectiveness.

Self-Directed Learnincr Theory and Research

In the literature, the term self-directed learning 
has become synonymous with self-planned (Tough, 1971), 
self-initiated (Pen!and, 1981), self-educated (Gibbons, 
1980), among other terms. As self-directed learning 
seems to be the term most commonly used, this will be 
the term used in this report. The definitions'-of 
self-directed learning are also many and varied. These 
definitions are spread across a continuum. This 
continuum represents degrees of self-directedness.
What qualifies as self-directed behavior (whether or 
not that behavior is personality based, instructionally 
based, or cognitively based) differs considerably from 
one theorist to the next. On one extreme, of the 
continuum is Mezirow (1981) who.links self-directed



28
learning to perspective transformation or a change in a 
person's awareness as an aspect of the emancipatory 
process of the adult learning process. Close to 
Mezirow on the continuum, but a little farther .towards . 
the middle is Tough (1971) who recognizes within broad 
parameters almost any deliberate effort to gain 
knowledge or create a change of view as self-directed 
learning. Somewhere in the middle of the continuum is 
Knowles (1975) who describes self-directed learning as 
a process of self-initiated, self-planned, self- 
implemented, and self-evaluated learning. On the 
opposite extreme of the continuum is the traditional 
school view of the independent study project - an 
academic project initiated and implemented through 
traditional, academic parameters.

The elusiveness of theory for both adult 
education, in the broader sense, and self-directed 
learning, in particular, is seen in.the literature as 
frustrating and defeating to this field of study 
(Tough, 1979, Mezirow, 1981) . Most practitioners and 
researchers in the field have recognized a need for 
theory (e.g,., Gibbons, 1980; Mezirow, 1981; Penland,/ 
1981; Cross, 1981).

Cross (1981) delineates three major stumbling 
blocks to theory building in adult education. First, 
she notes that adult education is seen by adult



29
educators as a consumer-oriented service. Thus the 
task is not always discovering what is most effective, 
but rather what is most wanted. Second, Cross notes 
that there are relatively few scholars in adult. ■ 
education. Most adult educators feel that their first 
obligation is to serve the needs of the adult learners. 
The strongest argument Cross makes is her third point. 
She suggests that the multidisciplinary nature of adult 
education (e.g ., pyschology, sociology, gerontology, 
and physiology) makes it difficult to assess where the 
focus of an adult education theory should be placed. 
Further, as Cross implies, it is not likely that one 
theory will ever be conclusive enough to address every 
essential aspect of adult education. She suggests that 
perhaps it would be more useful to have several 
theories, each examining in detail one of the disci­
plines of adult education.

In reviewing the work of several theorists who are 
attempting to create a theory on adult education. Cross 
discovered five similarities among theories and sug­
gests that they are very applicable to concepts ■ 
involved with self-directed learning. First, 
interaction between the person and the environment is 
seen as an important component. Each theory reviewed 
is grounded in the belief that each individual has some 
control over his/her environment. Second, a low



30
self-esteem is counter-productive to the processes of 
adult educationThird, incongruence and dissonance" 
are common themes throughout adult education. Forth, 
Maslow7 s (197 0) hierarchy of needs is also a common 
concept to adult education theories. Finally, ■ 
motivation is seen as being impacted by the perceptions 
of the positive and negative forces surrounding the . 
learning situation, and by the expected rewards of the 
learning situation.

This work by Cross (although more specifically 
related to adult education rather than self-directed 
learning) addresses issues very relevant to self- 
directed learning. The following is a review of the 
more prominent research projects in self-directed- 
learning that suggest some of the current theoretical 
assumptions.

Malcolm Knowles has contributed much to the 
literature base.for self-directed learning..
Self-Directed Learning: A Guide for Learners and
Teachers (Knowles, 1975) is considered by many a 
significant contribution to the literature. This is a 
how-to book that guides the learner towards 
self-directedness while at the same time guiding the 
teacher through the process of losing the center 
spotlight to become a facilitator of learning.
Recently, Knowles (1985) has delineated the differences•



31
between pedagogy and andragogy.. He describes pedagogy 
as a teacher centered monologic approach to learning. 
The learner is pictured as a blank slate to be written 
upon, or as Friere (1970) has described the traditional 
concept of the student, as a recepticle to be dumped 
on. Andragogy is described as a student-centered 
approach to learning. It is described in terms of the 
adult and the adult is described as one who by 
definition is self-directed.

Another significant contribution to the literature 
of seIf-directed learning is The Adult's Learning 
Projects. (Tough, 1979) conducted interviews with 66 
adults in 1970. This survey focused on the number, 
type and variety of learning projects the adults had 
undertaken in the preceeding year. A learning project 
was defined as, "simply a major, highly deliberate 
effort to gain certain knowledge and skills (or to 
change in some other way) "- (p. I) . All but one 
individual in Tough's study, had participated in at 
least one learning project (some were involved with as 
many as fifteen or twenty; the median number was 
eight). Further, he discovered that adults spend from 
100 to 2000 hours per year on learning projects (687 
hours was the median number of hours for the adults in 
this study). The estimated number of hours may be a 
conservative number as only one-third of the learning



32
projects were complete or'dormant at the time of the 
survey. Another important finding of this study was 
that 68% of the learning projects undertaken by the 
adults in this study were planned primarily by the 
learners themselves; they were self-planned/ 
self-initiated projects. This study supported, 
empirically, what most adult educators knew intuitively 
- that self-directed learning was a common practice 
among adults. Further, the evidence from this study 
would suggest that for many adults self-directed 
learning was the preferred method of study/instruction.

Gibbons, et al (1980) took a more qualitative 
approach to the research of self-directed learning.
They studied.the biographies or autobiographies of 
twenty individuals of prominence who, with one 
exception, had no formal education beyond high school. 
From this study. Gibbons and his associates were able 
to extrapolate fourteen principles and implications 
related to self-directed learning. Some of the more 
relevant factors are: I) the locus of control for a
self-educated person is himself/herself, and it is the 
task of the instructor to help students internalize 
control over their own learning; 2) self-directed 
learning is usually more focused,. and the instructor, 
should help students identify their areas of interest; 
3) learning is for immediate application (self-directed



33
learners tend to be pragmatic in their learning 
activities), and the instructor'should integrate theory 
with practice; 4) self-educated individuals tend to be 
self-motivated; instructors should encourage students 
to set their own goals and objectives; 5) 
self-educators are motivated by. the "rewards", of the 
project; instructors can help a student follow-through 
to the attainment of the reward; 6) self-educated 
people tend to find a method of study that is unique to 
them and their needs; instructors should encourage 
students to try a variety of learning styles to find 
the ones that work best for them; 7) self-educated 
adults use their skills to find the resources and 
direction they need; instructors can help encourage the 
drive and independence students need to seek out their 
own guidance. Gibbons states that, "teaching for 
self-education involves helping each student to become 
an expert, a participant, and a person" (p. 55).

Mezirow (1981) described a theory of adult 
education based on the work of Habermas (cited in 
Mezirow, 1981). He describes perspective transfor­
mation as central to the emancipatory aspect of adult' 
learning. The emanicpatory domain of adult learning is 
seen as a prerequisite for a person's ability to be 
self-directed in his/her learning. This observation 
seems consistent with the views of authors such as



34
Gibbons (1980) and Oddi (1983) who have concentrated on 
the personality constructs of self-directed learning, 
and by Knowles, (1970) when he describes the differ­
ences between andragogy and pedagogy). Perspective 
transformation, according to Mezirow, is the experience 
of processing critical events- in one's life by 
integrating the past and therefore moving to a new 
understanding, and a new perspective. The process of 
perspective transformation is delineated by the 
following events. First, a significant emotional 
experience occurs and disorients the individual which 
in turn causes self-examination leading towards a 
critical assessment of beliefs and values. Next, is 
the recognition that others have shared this 
disorientation which in turn allows the individual .to 
explore options with others. Finally, once a new 
option is acted upon and confidence in this new role is 
gained, a plan of action is created; new skills and 
knowledge are tried until the individual feels more 
comfortable with their new perspective. The process 
can be sudden if'the significant emotional experience 
is disorienting enough. Or the transition.to a new ■ 
perspective can - happen gradually over the course of 
time with several less demanding, less disorienting
events.



35
It is Mezirow''s contention that an understanding 

•of perspective transformation will enhance the adult 
educator's ability to integrate self-directed learning 
as an instructional model. Mezirow (1981) describes a

Oself-directed learner as one who:
...has access to alternative perspectives for 
understanding his or her situation and for giving 
meaning and direction to his or her life, has 
acquired sensitivity and competence in social 

• interaction and has the skills and competence
required to master the productive tasks associated 
with controlling and manipulating the environment, 
(p. 21)
Further Mezirow (1981) suggests that perspective, 

transformation in a related context with self-directed 
learning offers, "the essential elements of a compre­
hensive theory of adult learning and education" (p.
22).

Spear and Mocker (1984) took yet another approach 
towards looking at self-directed learning. Rather than 
looking at the internal characteristics of the indivi­
dual, they looked at the external characteristics of 
the individual's circumstance. "They conducted a 
secondary analysis of data collected previously and 
detected some environmental factors which describe the 
ways individuals organize and initiate learning 
projects. These factors become the organizing 
circumstance - the events which dictate the need, 
interest, motivation for the learning situation. Their 
qualitative analysis was taken from a survey-based



.36
research project which compared individuals who were 
involved in formal learning situations, with indivi­
duals involved in self-directed learning projects.
From this analysis they drew the following hypothesis:

The Organizing Circumstance, rather than 
preplanning by the individual, is the directing 
force behind much, perhaps most, self-directed ■ 
learning for this population, (p. 4) ■
Although the specific organizing circumstances

vary from individual to individual, there are some
themes that seem-common to most self-directed learners.
A most.central theme involves the concept of change.
Impetus for learning often stems from a positive o r '.
negative change in a person's life circumstances. The
change in circumstances typically does not provide very
many opportunities or resources for the learning situa-

< .tion; therefore, "the structure, methods, resources and
conditions for learning are provided or dictated by the 
circumstances" (p. 5). Also, learning is not neces­
sarily a linear progression, but rather circumstances 
seem to suggest the next logical step. The difference 
here seems to be that the individual may see a linear 
progression in the light of hindsight; however, this 
person did not nor could they have planned that linear 
progression proactively. ' ' .

Spear and Mocker are suggesting that circumstance 
plays"a large part in organizing the self-directed 
learner's choices about learning. The study they



37
analyzed seems to contradict Tough's (1979) research 
which stated that self—directed learners were very 
planful in their approach to learning. It is 
interesting to note the significance that Spear and 
Mocker found regarding the change in circumstance as an 
initiating factor for self-directed learning. This 
observation, in part, would support Mezirow's (1981) 
theory of "perspective transformation." Brookfield 
(1984) makes several critical comments regarding the 
existing research encompassing self-directed learning. 
Several of these points address the lack of research 
concerning some of the environmental factors of ~ " 
learning. He believes that the research has focused 
primarily on white, middle class America, and that the 
working classes and minorities have largely been 
ignored. Concentration on adults whose, educational 
attainment level is above the national norm is one 
characteristic Brookfield finds most alarming in the 
current research. Further, he notes that with only a 
few exceptions the research has been very ethnocentric, 
considering mainly the white population in America. 
Brookfield also states that self-directed learning 
research has also failed to adequately address the 
social climate surrounding the individual. Learning 
networks, informal learning exchanges require the 
individual to participate in his/her social setting.



38
Learning in this way is still self-directed rather than 
other-directed, but it necessitates an individual's 
interaction with others.

In response to Brookfield's critical assessment, 
Brockett (1985) suggests that Brookfield had a narrow 
view of "hard-to-reach" populations. Brockett cited 
his own research which targeted an elderly population 
as well as several others which he felt offered some of 
the kinds of data that Brookfield .has been quick to say 
doesn't exist, or only exists in relative scarcity.

Measurement of Self-Directed Learnincr
The importance of being able to measure 

self-directed learning lies in the need that educators 
have for knowing their students and how best to teach 
them. There exists, at this time, only two instruments 
for assessing self-directed learning: The Self-
Directed Learning Readiness Scale (Guglielmino, 1977), 
and The Oddi Continuing Learning Inventory (Oddi,
1983) .

The SeIf-Directed Learning Readiness Scale is an 
instrument which examines to what degree individuals 
perceive themselves to be self-directed in their 
learning. It is a 58 item, Likert scale design. The 
instrument has been used frequently in research on 
self-directed learning. (e.g ., Brockett, 1983; Leeb, 
1983)



39
Brbckett (1985) raised several concerns with the 

Self-Directed Learning Readiness Scale (SDLRS). He 
raised a concern with the appropriateness of the 
instrument with certain populations (i.e ., individuals 
with little formal education). Brockett also concluded 
that twelve, of the fifty-eight SDLRS items did not 
significantly correlate with the entire instrument.
The use of double negatives in some items added to the 
confusion of respondents - accentuating some of the 
correlation c o n c e r n s Brockett also noted that the 
SDLRS seemed to place specific value on book learning 
as a preferred tool for self-directed learning.

In response to some of the criticisms of the 
SDLRS, Oddi (1984) developed a new instrument to 
measure self-directed learning, the Oddi Continuing 
Learning Inventory (OCLI). Her study had three 
specific goals:

1. to describe a theory of the personality 
characteristics of the self-directed learner.

2. to develop an instrument to assess these 
learners.

3. to validate the instrument, empirically.
(p. 98)

She began by reviewing the relevant literature on 
the subject; and, by drawing on some common themes, 
derived three dimensions of self-directed learning.
Each dimension lies on a continuum with polar 
opposites. One pole describes a high inclination ■ 
towards a characteristic of self-directed learning; the



40
opposite pole, a low inclination for self-directed 
learning. These continuums are labeled as follows:

1. Pro-active Drive versus Reactive Drive (PD/RD)
2. Cognitive Openness versus Defensiveness (CO/D)
3. Commitment to Learning versus Apathy or 

Aversion to Learning (CL/AAL) (pp.98-99)
A pilot instrument was derived from the analysis 

of 100 items. This pilot instrument consisted of 31 
items and was given to 287 graduate students in law, 
nursing, and education. Five items were discerned to 
be unreliable in the item, analysis and were subse­
quently eliminated from the scale. The remaining 26 
items had a raw coefficient of .750. A validation 
study was conducted to obtain estimates of reliability 
and to discern internal and external validity... No less 
than 7 items were included for each dimension. 
Validation data were.obtained by administering.this 
instrument and a selected validation•instrument. Four 
separate validation instruments were used. Each 
instrument correlated to a particular component of the 
OCLI. The reliability of internal consistency was 
.875, while test/retest reliability was .893. The 
author noted that these were preliminary findings and 
that more data should be collected and that the 
instrument should be refined further.



41

Self-Directed Learning- and Intellectual Development

The concern for this study was the relationship 
between intellectual development and self-directed 
learning. Kasworm (1983) considers self-directed 
learning as part of a broader context which includes 
the cognitive and affective factors. She views change, 
growth, and development as intregal parts of the 
self-directed learning process. She outlined five 
considerations for a paradigm of self-directed.learning 
that assumes a developmental stance. These consider­
ations can be related to the definition of stages which 
was presented earlier; her considerations describe 
levels which are hierarchical, sequential, and qualita­
tively different levels of development. ' They are as 
follows:

1. Levels of development imply a qualitative 
difference in the individual's mode of 
thinking about him/herself in the personal 
world.

2. These levels represent a complex process 
including:
a. level of skiII/behavior for engagement in 

learning inquiry
b. cognitive capacities and competencies
c . affective and value orientations

3. There is an invariant sequence for these 
levels - one must logically follow another.

4 . Each level represents the individual's 
perceptual and cognitive structure of thought.



■ 42
5. ■ Each level is necessary to the total process 

of development and has both positive and 
negative potentials, (p. 33)

Further, Kasworm suggests a three dimensional 
framework for considering the relationship between 
self-directed learning and cognitive development. This 
framework includes: a) specific levels,of behavior/
skill to engage and complete the action of self- 
directed learning; b) specific levels of cognitive 
complexity necessary for specific nature of acts of 
learning; c) specific levels of affect/value towards 
orientation of knowledge and learning actions.

Kasworm suggests that considering self-directed 
learning within this framework presumes several points. 
It presumes that there exists both quantitative and 
qualitative differences in generic skills/knowledge, 
and values of self-directed learning. It presumes that 
skills, knowledge and values evolve over the lifespan 
and, "incorporate a person-environment referent in 
relation to depth and breadth of engagement in 
self-directed learning" (p. 45). Kasworm conludes her 
remarks by suggesting a manner in which instructors 
could create an environment more conducive to promoting 
self-directed learning, "Thus, instructors who wish to 
create self-directed learning environments would 
provide sufficient challenge for both the varied level's 
and depth of complexity for self-directed learning 
development" (p. 45).



43
Cameron (1984) has noted that the accelerating 

rate at which adults are returning to post-secondary 
institutions' gives relevance to a study on cognitive 
development and adult learning theory. Her study 
examined the cognitive positioning of adult students 
during their first year in a two year institution.
Adult learners were described, "as persons who enter 
college following several years of other activities, 
such, as jobs, military, mothering, and are therefore 
older than the traditional age student" (p. 2).

A professional group of raters for the Perry 
scheme developed and validated an essay test (protocol) 
for the study. Fifty students over the age of 22 were 
tested. In the first sample of 46 individuals there 
were 29 dualists (63.3%), 12 multiplists (26.0%), and 5 
relativists (10.8%). Table I indicates a. break down by 
curriculum.

Table I Intellectual Development by Curriculum

Business Liberal Arts and
Curriculum Human Services

D (7) 46.6% . . . (22) 70.9%
M (6) 40.0% (6) 19.3%
R • (2) 13.3% (3) 9.6%

This study supported previous data to suggest that
most adults were reported to be at the lower positions



44
of cognitive development. The age range for the 29 
respondents at the dualist stage was 22-50 years, which 
Cameron claims disputes a positive correlation between 
cognitive development and age. According to Cameron 
several assumptions about these adult learners can be 
made. We can expect them to be highly dependent on the 
instructor, to have an external orientation, to possess 
a high need for approval, to be less aware of their own 
learning needs. Further she states, "students who are 
dualists will not be capable of self-directed learning, 
even though they can be encouraged to be more and more 
self-directed" (p. 9). Cameron speculates that many of 
these adult learners would benefit from a 
support/challenge ratio which would possibly assist in 
their cognitive development towards a higher 
positioning.

In her study, Leeb (1983) examined the 
relationship between health conducive lifestyles and 
self-directed learning. She used essay protocols to 
assess the Perry scheme and to establish how adults 
reason about their health and health goals. Although 
it was not of central importance to her thesis, Leeb 
did examine the relationship between Perry positioning 
and results on the Self-Directed Learning Readiness 
Scale. She found no statistically significant rela­
tionship. However, she did note that her findings were



45
speculative as the study she conducted was not designed 
as a correlational.study.

Much of the literature regarding the relationship 
between self-directed learning and intellectual ' 
development is speculative at this point. Not enough 
research has been conducted to draw adequate conclu­
sions. This void in the research is.a significant 
reason for this study.

Summary

Cognitive developmental theory suggests that 
individuals reason about and react to. their environment 
throughout their lives. This reaction can be a dog­
matic denial- of any other perspective, or it can be a 
reaction to the dissonance caused by examining the 
potential for a different perspective.. Growth is 
characterized as a sequential process of .moving towards 
a broader perspective. Perry suggests that growth 
includes changing one's view of authorities, peers, and 
truth. The MER measures respondent's views of the 
learner, the instuctor, peers, decision making pro­
cesses, evaluation, and truth/knowledge. These six 
categories relate well to examining the differences 
between a self-directed learner and. an other-directed 
learner. Therefore, an examination of self-directed 
learners in a cognitive developmental framework can



46
provide valuable information for educators. As Kasworm 
suggests, the instructor could use such information to 
create an environment conducive to promoting 
self-directed learning in individuals. Thus far the ■ 
research examining the connection between the two 
concepts has been very limited.■ This study adds to 
literature base uniting the concepts. An underlying 
presumption of this study, at the onset, was that as 
persons achieve a higher level of intellectual 
reasoning they become more self-directed in their 
learning. Although Kasworm, Cameron, and Leeb have 
provided some information regarding the nature and 
importance of the relationship, there is much to learn 
about this relationship.



47

CHAPTER 3 

METHODOLOGY

The previous chapter of this research report 
outlined some of the more significant literature 
regarding intellectual development (as specifically 
related to the Perry scheme) and the theories and 
research regarding adult education self-directed 
learning. A limited selection of literature was also 
presented which suggested that a relationship may exist 
between self-directed learning and intellectual 
development. This chapter will outline the methodology 
used to examine this relationship. This research 
project was conceptualized, in part, as a response to a 
void in this literature base. The problem examined in 
this study is the nature of the relationship between a 
person's predisposition/preference for self-directed 
learning and that person's intellectual development.

Population and Sample

The population from which the sample for this 
study was taken is the total undergraduate and graduate 
enrollment for Montana-State University. A demographic 
description of this population is found in a 1984



48
survey of incoming freshmen (Astin and Groseth, 1984). 
This survey echoed, in many respects, the results of 
previous surveys taken in 1976 and in 1970. Approxi­
mately 97% of the freshmen population in 1984 described 
their race as white/caucasian. This figure is consis­
tent within I percentage point of the previous surveys. 
The only other' race of individuals with a significant 
percentage was Native Americans with almost 3% of the 
total enrollment (a figure which has also stayed fairly 
consistent). Increasingly, a growing number of 
attendees at MSU come from homes in which at least one 
parent is a college graduate. While in 1970, only 
23.9% of entering freshmen claimed that their fathers 
had completed a college education, the figure for 1984 
was 42%. Another interesting statistic reflects family 
income. In 1984, 46.9% came from households with an 
annual income in excess of $30,000. From these 
statistics we can assume that the population of MSU is 
largely white and middle class. Any results of this 
survey may only be applicable to other populations 
within higher education with similar demographics.

A random stratified sample was taken from the MSU 
population. Stratums identified for this study were 
based on the university's classification of students as 
outlined by the undergraduate catalog (1986-1988).
These classifications are:



‘49
Freshman: A student who is entitled to regular or

conditional admission with less than 45 credits.
Sophomore: A student.must have earned 45 or more

credits.
Junior: A student must have earned 90 or more

credits.
Senior: A student must have earned 135 or more

credits. ' .
Second bachelor's degree candidate: A student,

seeking a bachelor's degree who has already earned one 
or more baccalaureate degrees.

Nondegree student: A student with at least a
baccalaureate degree but not seeking another degree.

Graduate degree student: A student who has at
least.a baccalaureate degree and has been accepted into 
the College of Graduate Studies, (p. 20)

Initially a sample of 100 students from each of 
the five straturns was drawn randomly by the Registrar's 
office. One of the major difficulties of this study 
■was contacting the students in the sample. The list 
from the Registrar's office did not include phone 
numbers and since phoning the sample was an important 
part of the data collection it was necessary to obtain 
phone numbers from other sources. It was possible to 
obtain phone numbers for approximately 50%.of the 
sample. Another difficulty encountered was that the 
information contained on the list was inaccurate for at 
least 10% of the sample. Some individuals listed were 
no longer students; many had changed residences.
Because of these difficulties, it was necessary to draw



50
a second sample. This second sample included only 
those students in the post baccalaureate sample.

Instrumentation and Scoring

The instruments chosen for this study were both 
developed within the last four years. However, there 
is evidence to indicate that each has appropriate 
validity and reliability to contribute solid results to 
this study.

The Measure of Epistemological Reflection
The Measure of Epistemological Reflection (MER) 

was chosen for this study for several reasons. The 
major advantage of this instrument is its format. The 
(MER) is a short answer essay with follow-up questions. 
There are six domains investigated by the instrument: 
decision making; role of the learner; role of the 
instructor; role of peers; evaluation; and view of 
knowledge, truth, and/or reality. The standardized 
format and scoring reduces the degree of inference 
necessary to evaluate along the Perry scheme.

Another advantage of this instrument is its ease 
of administration and the nature of the responses. The 
instrument can be given to the respondent in any 
setting. The nature of the questions make it unlikely 
that the individual can manipulate his or her response.



51
According to Taylor (1983), reliability of the MER 

was based on the interrater agreement and bn internal 
consistancy among domain ratings. Interrater agreement 
on the individual domains ranged from 42% to 64%; 
however, ihterrater agreement for the total protocol 
rating (TPR) ranged from .75 to .84. As was noted 
earlier, in order to be certified to rate the MER, an 
individual must obtain approximately a .80 correlation 
with an expert rater on a collection of sample proto­
cols. Internal consistency among domains for the 
initial sample (79 men and 76 women) was found to be 
.76. Concurrent validity of the MER was established to 
be quite low when compared with another similar 
instrument - The Measure of Intellectual Development. 
Correlation of scores from both instruments ranged from 
.04 to -.11. A difference in the variability of ' 
reported scores may account, in part, for the low 
correlations. The MER is designed to obtain more 
informtion and assigns a more specific position rating 
than does the MID.

Scoring the instrument is accomplished by 
comparing the individual's responses to the instrument 
with the reasoning structures outlined in the scoring 
manual.- A score ref lecting the Perry positioning and a 
reasoning structure as outlined by the manual- is 
assigned to each domain rating; the dominant rating is



52
assessed. Any additional rating which occurs twice is 
shown in parenthesis [example, 2(3)]. .If the ratings 
are evenly split between two ratings the TPR is shown 
with a dash (example, 2-3). A rating which occurs only 
once is not reflected in the TPR. The TPR reflects a 
person's positioning along the Perry scheme. The Perry 
positioning assigned to a person reflects the way that 
person reasons about his or her world. For research 
purposes the total protocol rating can be converted 
into a continuous variable. This was done in order to 
complete the statistical analysis of this study.

The Oddi Continuing Learning Inventory "
The. Oddi Continuing Learning Inventory (OCLT) was 

chosen to evaluate self-directed learning. The instru­
ment was chosen for a number of reasons. First, the 
instrument is easy to administer; like the MER, the 
OCLI can be administered in almost any setting. 
Instructions for completing the instrument are fairly 
straightforward and easy to follow. Second, it is 
relatively inexpensive to use. .A licensing agreement 
with the author allows a researcher to use the 
instrument without royalty. Finally, the OCLI examines 
self-directed learning as a personality construct which 
is how the concept is operationalized for this study.. 
The OCLI is a 24 .item Likert scale. .' Each item consists 
of a statement to which the respondent is asked.to



53
react. The scale ranges from a "I" which reflects that 
the person strongly disagrees with the statement to a 
"7" which reflects strong agreement with the statement. 
There are five reverse items on the scale which guard 
against "out-guessing" the instrument.

The OCLI measures factors on three continuums: 
Cognitive openness versus defensiveness; proactive 
drive versus reactive drive; and, commitment to 
learning versus apathy or aversion to learning.. 
External validity for the OCLI was determined by 
comparing these different factors measured by the 
instrument with other instruments which measure the ■ 
same factors. Commitment to Learning vs. Apathy or 
Aversion to Learning (CL/AAL) was validated against the 
Leisure Activity Survey (r = .363)' and against the 
Affliation subscale of the the Adjective Check List (r 
=.265). The Internal-External Scale was measured 
against the Cognitive Openness vs. Defensiveness (CO/D) 
factor (r= -0.40); and against the Change subscale of 
the Adjective Check List (r = -.020). The Proactive 
Drive vs. Reactive Drive (PD/RD) factor was correlated 
with two sub-scales of the Adjective Check List. The 
Self-Confidence subscale correlated with an r = .551; 
the Endurance subscale correlated with an r = .539.
Oddi (1985) suggests that the .instrument is a valid 
instrument when considered in its entirety, but raises



54
cautions about the use of the factors. For this reason 
factor scores were not considered in this invest!- . 
gation. The reliability of internal consistency was 
.87 while test/retest reliability was determined to be 
.89. The size of the sample on which validity and 
reliability figures are based was 271 (Oddi, 1985).

As was noted, the instrument is fairly easy to 
administer. The individual is asked to circle the 
number which corresponds to his/her' reaction to the 
statement. Scoring the instrument is merely a matter 
of adding the numbered responses. The range of 
possible scores is from 24 (which represents a low 
inclination towards self-directed learning) to 168 
(which represents a high, inclination towards self- 
directed learning).

Research Design and Data Collection Procedure

The design for this study is an ex-post-facto, 
correlational design. It is an ex-post-facto design 
because the variables, intellectual development and 
self-directedness, are not manipulated in the research 
and because these factors are associated with the 
individual's life span development. It is a correla­
tional study because it is examining the relationship
between two factors.



55
A random stratified sample was obtained from the 

Registrar's office at Montana State University. From 
this list, phone calls were made to 158 individuals 
soliciting participation in this study. The phone 
calls offered a brief introduction to the study and 
asked for a verbal agreement for participation. 
Individuals who agreed to participate were sent a 
packet of materials which included the following: a
letter to participants; a consent form; instructions 
for completing the instruments; a demographics 
questionnaire; the Measure for Epistemological • 
Reflection; the OCLI; and a return envelope.
Indivduals were instructed on how to return the 
completed materials to the principal investigator. An 
example of this packet of materials is provided in 
Appendix A.

Hypotheses

To investigate thoroughly the relationship between 
self-directed learning and intellectual development, 
nine operational hypotheses were postulated. Stated in 
the null, they are:
1. There is no statistically significant relationship 

between scores on the OCLI and scores on the MER.
2. There is no statistically signfleant relationship 

between chronological age and scores on the MER.



56
3. There is no statistically significant relationship 

between chronological age and scores on the OCLI.
4. There is no statistically significant difference 

' between individuals under the age of twenty-five
and indivdiuals over the age of twenty-five on the 
MER.

5. There is no statistically significant difference 
between individuals under the age of twenty-five 
and individuals over the age of twenty-five on the 
OCLI.

6. There is no statistically significant difference 
between the scores of male's and females on the 
MER. '

7. There is no statistically significant difference 
between the scores of males and females on the 
OCLI.

8. There is no statistically significant relationship 
between grade level attainment and scores on the 
MER.

9. There is no statistically significant relationship 
between grade level attainment and scores on the 
OCLI.

Summary

The distribution and collection of data took 
approximately two and a half months to complete. The



57
data collected .are rich with information regarding 
self-directed learning and intellectual development.
The following chapter will discuss the statistical 
configurations of the sample population; the results of 
the OCLI and the distribution of scores; the results of 
the MER and the distribution of scores; the results of 
each tested hypothesis. These data offer a clearer 
picture of the relationship between these two factors - 
perhaps the clearest picture available in the current 
literature base. Additionally, the information 
provided by this study is potentially valuable to adult 
educators as well as to the instructor or administrator 
in a higher education setting.



58

CHAPTER 4 

.RESULTS

In Chapter 3, nine hypotheses, stated in the null, 
were postulated to examine the relationship between 
self-directed learning and intellectual development. •
In this chapter, the characteristics of the sample will 
be discussed; results of each instrument will be 
presented; and finally there is a discussion of the 
results as they pertain to each hypothesis.

Characteristics of the Sample

Respondents were asked to complete a demographic 
questionnaire. This questionnaire was designed to 
solicit information regarding the sample which could be 
used for descriptive purposes. Data were collected 
regarding gender, age, race, family income, whether or 
not the respondent attended high school in Montana, and 
whether or not the respondent's father and/or mother 
graduated from college'. Results of these data are 
presented in Table 2. •

Research materials were sent to 154 individuals 
who had indicated, over the phone, their willingness to 
parti- cipate in this study. One-hundred of these



59

Table 2 Characteristics of Sample Population
Characteristic_____________________ n____________percent

Gender
Males
Females
Total

50 50
_50 50
100 100

Age
Under 25 
25 and Over 
Total

61
39
100

61
39
100

Class
Freshman
Sophomore
Junior
Senior
Post-Baccalaureate 
Total

22
15
16 
15 
32
100

22
15
16 
15 
32
100

Attended Montana High School
Yes 71 71
No 29 29

100 100Total



60
Table 2 Continued
Characteristic_____________________ n____________percent

Parent's Income
10,000 - 19,000 5 5.6
20,000 - 29,000 2 31.1
30,000 - 39,000 20 22.2
40,000 - 49,000 18 20.0
50,000 - higher 11 21.1
Total 90 100.0

Race
Caucasian 97 98.0
Oriental I 1.0
Other I 1.0
Total 99 100.0



61
Individuals returned completed research materials for a 
response rate of 65%. Fifty-percent of the respondents 
were men and 50% were women. Respondents ranged in age 
from 17-54 with the mean age for the sample being 25.27 
(with a standard deviation of 8.18). Sixty-one 
percent of the respondents were under the age of 25. 
Ninty-eight percent of the sample reported their race 
as "Caucasian"; one percent indicated "oriental"; one 
percent indicated "other". • One individual did not 
respond to this question. Respondents were asked to 
approximate their parent's income. Of the 100 indivi­
duals in the sample, 10 chose not to respond to this 
question. Of the respondents who answered this 
question, 63.3% indicated that their parent's income 
was over $30,000 per year.

The results of this demographic questionnaire are - 
similar to results found by the surveys of incoming 
freshmen (Astin and Groseth, 1984) cited in Chapter 3. 
The sample is mostly white, middle class and most ' 
attended high school in•Montana. These individuals 
seem to come from homes which indicate a value for 
higher education as 50% of the fathers and 36% of the 
mothers had graduated from college.



62

The Measure of Epistemological Reflection

Ninety-eight respondents completed the Measure of 
Epistemological Reflection (MER) in a manner which 
provided rateable data. Table 3 reflects the numerical 
breakdown for scores on the MER. Table 4 reflects 
Perry'positioning for individuals who completed the 
MER.

The range of scores possible on the MER is 1.00 - 
5.00. The range of scores for individuals in this 
study was 1.80 - '4.66. The mean for the 98 respondents 
was 2.95. Sixty individuals under the age of 25 
responded (x = 2.69). Thirty-eight individuals 25 
years of age and older completed the questionnaires (x 
= 3.38). There was equal representation by men and 
women on the MER. Women scored slightly higher than 
men. The mean for the female sample was 2.97 and for 
the men was 2.94. Juniors appeared to have the lowest 
average for scores on the MER with a mean of 2.52; 
freshmen had a mean of 2.57; sophomores had a mean of 
2.75; seniors had a mean of 3.07; and post baccalaur­
eate students had a mean of 3.48. Except for the 
junior sample, a positive correlation between intellec­
tual development and educational attainment is shown by 
these findings. The small sample representing each 
class rank may be a factor in these results.



63

Table 3 Numerical Results for the MER
Characteristic n X sd
Acre
17-24 60 2.69 .49
25-54 38 3.38 . 66

Gender
Females 49 2.97 . 64
Males 49 2.94 . 67

Class Rank
Freshman 22 2.57 . 60
Sophomore 15 2.75 .58
Junior 15 2.52 .36
Senior 15 3.07 . 66
Post-Baccalaureate 31 3.48 .49

Total Sample 98 2.95 . 65



64

Table 4 Positioning Results for the MER
Characteristic Dualists Multiplists Relativists
Age
17-24 46 12 2
25-54 8 21 9

Gender
Females 29 15 5
Males 25 17 6

Class Rank
Freshman 19 3 0
Sophomore 10 4 I
Junior 13 2 0
Senior 7 6 2
Post-BaccaIaureate 5 18 8

Total Sample 54 33 11



65
There are 54 dualists, 33 multiplists and 11 

relativists in this sample. There is a great 
difference between individuals under the age of 25 and 
individuals 25 years of age and older. Women and men 
were dispersed comparatively across the positions. 
Again, aside from the sample of juniors, positioning 
seems to increase with educational attainment.

Interrater Reliability for the MER
In order to rate MER data, this researcher 

completed a self-paced training program. At the end of 
that training program, I evaluated twenty protocols. I 
attained a .78 correlation with the expert raters on 
this sample data (and therefore, gained certification 
to rate the MER data). See Appendix B for the letter 
confirming certification to rate MER data.

To add to the reliability of the findings of this 
study, one of the developers of the MER agreed to rate 
10 randomly selected protocols from the 100 protocols 
completed for this study. A Pearson product moment 
correlation was used to examine the relationship 
between the expert rater's findings and this re­
searcher's findings. ' The correlation on these 10 
protocols was r=.79.



66

The Oddi Continuing Learning Inventory

One-hundred respondents' completed the Oddi 
Continuing Learning Inventory (OCLI). Table 5 reflects 
the breakdown of scores for the OCLI. The possible 
range of scores for this instrument is 24 - 168. The 
range of scores for individuals involved with this 
study was 54 - 157 (x = 121). There is hardly any 
difference in scoring on the OCLI for men (x" = 122) and 
women ("x = 121); and, hardly any difference in scoring 
for individuals under the age of 25 (x = 120) and those 
25 and over (x = 124). The only notable difference in 
scores was between post baccalaureate students and the 
undergraduates. The means for freshmen and sophomores 
were both 116; juniors had a mean of 112 (the lowest 
mean for class rank cells); the mean for seniors was 
119; post baccalaureates had a mean of 126. It is 
interesting to note that scores for juniors on both the 
MER and OCLI were lower than the scores for the other 
class rank stratums on both instruments.

Results of the Hypotheses

Nine operational hypotheses were examined in this 
■ study. They concern the relationship between the two 
instruments; age-related considerations; gender-related 
considerations; and class rank-related considerations.



67

Table 5 Results Related to the OCLI
Characteristic n X sd
Aqe
17-24 61 120 17.74
25-54 39 124 17.04

Gender
Females 50 121 19.41
Males 50 122 15.61

Class Rank
Freshman 22 116 17.62
Sophomore 15 116 13.10
Junior 16 112 14.06
Senior 15 119 19.88
Post-Baccalaureate 32 126 18.87

Total Sample 100 121 17.53



68
The following material will present each hypotheses, 
the tested results, and whether or not the null 
hypotheses has been rejected. .

Relationships Between the MER and the OCLI
The first hypothesis of this study is the main 

focus of the study. It was presented in the null, in 
Chapter 3, as follows:

Hypothesis I: There is no statistically significant
relationship between scores on the OCLI and scores on 
the MER.

A Pearson product moment correlation was used to 
measure the relationship between the scores on the two 
instruments. A correlation coefficient of .21 (p< .05) 
was found which rejects the null hypothesis. Although 
this is not an overwelming relationship it does 
indicate that to some degree a person who has attained 
a higher level of intellectual development is more 
likely to be more self-directed in his/her learning.
It should be noted that the variance for this coef­
ficient is quite low (r2- = .044) . Therefore, while 
there exists a statistical relationship, the magnitude 
of that relationship is weak.

Age-Related Findings for the MER and OCLI
Hypotheses two, three, four and five each examine 

the relationship of age to the two instruments. They 
are presented in Chapter 3 as follows:



69
Hypothesis 2: There' is no statistically significant-
relationship between chronological age and scores on the MER.
Hypothesis 3: There is no statistically significant
relationship between chronological age and. scores on ■ the OCLI.
Hypothesis 4: There is no. statistically significant :
difference between individuals under the age of 
twenty-five and individuals over the age of twenty-five 
on the MER.
Hypothesis 5: There is no statistically significant
difference between individuals under the age of 
twenty-five and individuals over the age of twenty-five 
on the OCLI.'

A Pearson product moment correlation was used to - 
measure the relationships between age and scores on the 
two instruments. Statistically significant relation­
ships were found for both the MER and age and the OCLI 
and age. This relationship can be noted in Table 6.
The relationship between the MER and age' is quite 
strong with a correlation coefficient of .55 (p< .001), 
with a variance of .30; the relationship between the 
OCLI and age, while statistically significant, is not 
nearly as strong with a correlation coefficient of .17 
(p< .05) with a variance of .028. This finding 
indicates that while age and intellectual development 
have a strong positively-correlated relationship, age 
and self-directed learning have only a slight, though 
statistically significant positive relationship.

Table 7 indicates the differences between indivi­
duals, above and below the age of 25 on the two



70

Table 6 Pearson
Scores

Product Moment Correlations 
on the MER, OCLI and Age

for

Relationship Correlation
OCLI and MER r = .21***
MER and Age r = .55*
OCLI and Age r = . 17***

* (p< .001)
*** (p< .05)

Table 7 Differences for Individuals Under and Over
the Age of Twenty -five on the MER and OCLI

Instrument Aqe Range n X T-Value
MER 17-24 60 2.69

25-54 38 3.38 -5.90*
Total 17-54 98 2.95

OCLI 17-24 61 120
25-54 39 124 -1.27

Total 17-54 100 121

(p< .001)



71
instruments. A T-test was computed to examine these 
differences. Age differences on the MER appear, again, 
to be a significant factor. Individuals under the age 
of 25 are more likely to be dualists, while individuals 
over the age of twenty-five are more likely to be 
multiplists. Scores on the OCLI do not appear to be
significantly different for persons above and below the 
age of 25.

The results of these statistics indicate that 
hypotheses two, three, and four are rejected at the .05 
level of significance. .The statistics failed to reject 
hypothesis five.

Gender-Related Findings for the MER and OCLI
Hypotheses six and seven concern the differences 

between males and females on the two instruments. They 
were presented in Chapter 3 as follows:

Hypothesis 6: There is no statistically significant
difference between the.scores of males and females on 
the MER.
Hypothesis 7: There is no statistically significant
difference between the scores of males and females on 
the OCLI.

T-tests were computed to examine gender-related 
differences for scores on the MER and OCLI. As is 
illustrated by Table 8, there was little difference 
between the scores of men and women on either



72
instrument. Therefore/ the statistics fail to reject 
hypotheses 6 and 7.

Table 8 Gender-Related Differences for. Scores on
the MER and OCLI

Instrument Gender n X T-value.
MER 'Females 49 2.97 '■ .19

Males 49. 2.94

OCLI Females 50 • 121 -.16
Males 50 122

Relationships Between the MER, OCLI, and Class' Rank
Hypotheses eight and nine examine relationships 

between the instruments and class rank. These . 
hypotheses are presented in Chapter 3 as follows:

Hypothesis 8: There is no statistically significant
relationship between grade level attainment ..and scores 
on the MER.
Hypothesis 9: There is no statistically significant
relationship between grade level attainment and scores 
on the OCLI. '

A Spearman rank order correlation was calculated
to obtain the nature of the relationship between scores
on the MER and Class Rank and between scores on the
'OCLI and Class Rank. Tables 9, 10, and 11 illustrate
these relationships.

Class rank correlated significantly with both 
instruments. The correlation for the MER with class



73
Table 9 Relationship between Class Rank and Scores 
____________on the MER and OCLI_________________________

OCLI MER
Class Rank .26 ** .57 *

* (p< .001)
** (P< .01)

Table 10 Differences between Class Rank and the MER
Class Rank n X sd
Freshmen 22 2.57 . 60
Sophomore 15 2.75 .58
Junior 15 2.52 .36
Senior 15 3.07 .46
Post-Baccalaureate 31 3.48 .49

Total 98 2.95 . 65

Table 11 Differences between 
OCLI

Class Rank and the

Class Rank n X sd
Freshmen 22 116 17.62
Sophomore 15 116 13.10
Junior 16 119 19.88
Senior 15 112 14.06
Post Baccalaureate 32 126 18.87

Total 100 121 17.52



74
rank was quite strong, (r = .57; p< .01) with a 
variance of .32. The correlation of the OCLI with 
class rank was not quite as strong, (r = .26; p< .001) 
with a variance of .09. It can be concluded from these 
results that, to some degree, as individuals become 
more educated they become more self-directed and they 
obtain a higher level of intellectual development. 
Therefore, hypotheses eight and nine are rejected at 
the .01 level of significance.

Interaction of Age and Class on the HER and OCLI 
A two-way analysis of variance (ANOVA) was 

computed to examine the interaction of age and class 
and how that interaction was reflected for both the MER 
and the OCLI. Table 12 reflects the results of these 
ANOVAs. There does not appear to be a statistically 
significant interaction of age and class on either 
instrument. This is surprising, given the results of 
the Pearson correlation which showed a positive 
relationship between age and class on both instruments, 
and given the results of the Spearman rank order 
correlation which indicated a positive correlation 
between class rank and both instruments.



75
Table 12 ANOVA Results 

Class on the
for Interaction 

MER and OCLI
of Age and

Instrument Interaction SS df f P

HER Age / Class 4546.657 2 .806 . 45
OCLI Age / Class 876.404 2 1 .463 .24

Summary

From the results of this study, we can conclude 
several things about the relationship between self- 
directed learning and intellectual development. First, 
there is a positive relationship between the two 
factors. The relationship found in this study is not 
terribly strong, but it is statistically significant. 
Second, age correlates positively with both instru­
ments. The relationship between age and intellectual 
development is quite a bit stronger than the relation­
ship between age and self-directed learning. Third, 
there are no significant differences between males and 
females on either instrument. Forth, class-rank 
correlates positively with both instruments. Again, 
there is a stronger correlation with intellectual 
development and educational attainment than with 
self-directed learning and educational attainment. 
Chapter 5 will present a discussion of the implications 
and limitations of these findings.



76

CHAPTER 5

DISCUSSION OF THE RESULTS

The results presented in Chapter 4 provide solid 
evidence regarding the relationship between intellec­
tual development and self-directed learning. While the 
data indicated that a statistically significant, 
positive correlation exists between the two factors, 
the magnitude of that relationship is tenuous. This 
chapter will discuss these results and their implica­
tions to practice. In addition, it will examine how 
the results of this study impact the current literature 
on this subject. Finally, limitations of this study 
and suggestions for future research studies will be 
presented.

Results Related to the MER

This study examined age, gender, and class in 
relation to scores on the HER. As was reported in 
Chapter 4, a statistically significant relationship 
does exist between chronological age and class rank and 
scores on the HER. It is difficult to speculate if 
older individuals have higher scores because of life 
experiences or because of educational attainment



77
because, in general, the older individuals of this 
study had attained more formal education. The results 
of the ANOVA would lead us to believe that age and 
class rank impact scores independently. There is little 
interaction between these two factors. If age can be 
isolated as a factor, this information could greatly 
assist adult educators in maximizing their effec­
tiveness. For example, educators could make certain 
assumptions about their students based on age. There­
fore, they could provide educational opportunities 
which offer an appropriate level of experiential 
dissonance. The appropriate level is dissonance which 
addresses the student's developmental reasoning 
structure. Presumably teaching at this level will be 
most effective, because it will offer enough challenges 
to keep the student engaged in the learning process, 
but will not be too challenging. Too much challenge is 
not productive to the learning process, because the 
student feels overwhelmed by the situation; and, 
therefore, is not engaged in the learn