• TABLE OF CONTENTS
HIDE
 Title Page
 Acknowledgement
 Table of Contents
 List of Tables
 Abstract
 Introduction
 Review of the literature
 Procedures
 Results
 Conclusions and recommendation...
 Appendices
 References
 Biographical sketch














Title: Community educational processes
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098836/00001
 Material Information
Title: Community educational processes group perceptions of energy issues
Physical Description: xi, 137 leaves : ; 28 cm.
Language: English
Creator: Blalock, Carol Douglass, 1944-
Publication Date: 1980
Copyright Date: 1980
 Subjects
Subject: Educational sociology   ( lcsh )
Power resources -- Public opinion   ( lcsh )
Personal construct theory   ( lcsh )
Curriculum and Instruction thesis Ph. D
Dissertations, Academic -- Curriculum and Instruction -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 132-135.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Carol Douglass Blalock.
 Record Information
Bibliographic ID: UF00098836
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000098366
oclc - 06731880
notis - AAL3812

Downloads

This item has the following downloads:

PDF ( 5 MBs ) ( PDF )


Table of Contents
    Title Page
        Page i
        Page ii
    Acknowledgement
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
    List of Tables
        Page vii
        Page viii
    Abstract
        Page ix
        Page x
        Page xi
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
    Review of the literature
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
    Procedures
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
    Results
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
    Conclusions and recommendations
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
    Appendices
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
    References
        Page 132
        Page 133
        Page 134
        Page 135
    Biographical sketch
        Page 136
        Page 137
        Page 138
        Page 139
Full Text













COMMUNITY EDUCATIONAL PROCESSES:
GROUP PERCEPTIONS OF ENERGY ISSUES








BY

CAROL DOUGLASS BLALOCK


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY



UNIVERSITY OF FLORIDA

1980





















TO MY HUSBAND TONY

AND MY CHILDREN

JEANNE, PATRICIA, AND ELIZABETH

FOR THEIR LOVE AND ENCOURAGEMENT,

AND TO MY PARENTS

WHO STARTED IT ALL















ACKNOWLEDGMENTS


The present research could not have been accomplished

without the assistance and cooperation of many individuals.

The researcher wishes to express appreciation to her doc-

toral committee,Dr. Arthur J. Lewis, Chairman, Dr. Roderick

McDavis, and Dr. Lynn C. Oberlin,for the time and guidance

provided. In particular, she is indebted to Dr. Arthur J.

Lewis for his valuable advice at critical stages in the

design and writing of this dissertation.

The researcher would like to further recognize Dr.

James R. Kennedy for his skill in the development of the

software used for the repertory grid analysis and Dr. H.

Anthony Blalock for scientific consultation in the data

analysis of energy issues. In addition, special thanks are

extended to Dr. Eugene A. Todd, Dr. H. A. Ingley, III,

Mr. John Dykes, and Mr. Lem Lee for their assistance in

data collection.

The researcher also expresses gratitude to her husband,

Tony, and her daughters, Jeanne, Patricia, and Elizabeth, for

their love, encouragement, and patience over the years

required to complete her graduate work.

Finally, the researcher would like to acknowledge the

influence of the many theoreticians and researchers whose

work provides the conceptual base for this dissertation.
iii









They rightfully share any credit this dissertation reflects.

The researcher feels as Sir Isaac Newton did: "if I have

seen thus far, it is because I have stood on the shoulders

of giants."
















TABLE OF CONTENTS


CHAPTER PAGE

ACKNOWLEDGEMENTS . . . . . . 111

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

ABSTRACT . . . . .... .. . .. ix

I INTRODUCTION . . . .... .. . . .. 1

The Problem and Its Context . . . 1
Personal Construct Theory . . .... ... 5
Design of the Study . . . . . .. 11
Organization of Dissertation . . .. 16

II REVIEW OF LITERATURE ...... .. . . 17

Analysis of Social Perceptions . . .. 17
Cultural Foundations of the Study . . 31
Summary . . . . . . ....... 44

III PROCEDURES . . . .... .. . . . 46

Selection and Identification of
Respondent Groups . . . ... 46
Design and Administration of Repertory
Grid Instrument . . . . . . 47
Data Reduction and Analysis ...... 54
Statistical Procedures and Tests of
Hypotheses . . .... .. . . . 57

IV RESULTS . . . .... .. . . . 63

Frequency of Element and Construct
Citation . . . . . . . . 64
Numbers of Constructs Elicited . . .. 67
Degree of Construct Association . . .. 68
Results of Factor Analyses of Grids . . 72

V CONCLUSIONS AND RECOMMENDATIONS .... 99

Group Perceptions of Energy Issues . . 102









CHAPTER


PAGE


Implications for Educational Planning
and Curriculum Development . . ... .105
Implications for Other Areas of
Educational Planning . . . ... .107


APPENDICES

A REPERTORY GRID INSTRUMENT . . . . .. 110
B DEMOGRAPHIC CHARACTERISTICS OF GROUPS . 113
C MASTER LIST OF ELEMENTS . . . ... 114
D MASTER LIST OF CONSTRUCTS . . . ... 116
E TYPICAL SPSS FACTOR ANALYSIS OUTPUT ... 119


REFERENCES . . . . . . . . ... .. . 132


BIOGRAPHICAL SKETCH . . . . . . . ... 136















LIST OF TABLES


TABLE PAGE

1 Demographic Characteristics of Groups . . 48

2 The Most Commonly Cited Constructs by
Groups . . . . . . . ... 65

3 The Most Commonly Cited Elements by
Groups . . . . . . . ... 66

4 ANOVA Table of Number of Construct ..... 69

5 ANOVA Table for Associations per Construct 71

6 Summary of Factor Structure for Engineering
Student Group . . . . . ... .73

7 Summary of Factor Structure for Education
Student Group . . . . . . . 75

8 Summary of Factor Structure for Community
Group . . . . . . . . ... 77

9 Unrotated Principal Components for
Engineering Students . . . . . 78

10 Unrotated Principal Components for
Education Students . . . . ... 80

11 Unrotated Principal Components for
Community Group . . . . . ... 82

12 Varimax Rotated Factor Matrix for
Engineering Student Group . . . ... 84

13 Varimax Rotated Factor Matrix for
Education Student Group . . . ... 86

14 Varimax Rotated Factor Matrix for
Community Group . . . . . ... .88

15 Major Factors for Engineering Student Group 90









TABLE PAGE

16 Major Factors for Education Student
Group . . . . . . . . 91

17 Major Factors for Community Group . . 92


viii















Abstract of Dissertation Presented to the Graduate
Council of the University of Florida
in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



COMMUNITY EDUCATIONAL PROCESSES:
GROUP PERCEPTIONS OF ENERGY ISSUES



By


CAROL DOUGLASS BLALOCK



JUNE, 1980

Chairman: Arthur J. Lewis
Major Department: Curriculum and Instruction


The major challenge to the planning of community educa-

tional processes lies in the pluralistic nature of society.

Educational theorists agree as to the critical need for an

emerging role for education in providing linkage between the

various formal and functional groups in society. In accept-

ing this challenge educators must find tools for mapping and

quantifying similarities and differences for values and per-

ceptions across groups. The purpose of this study was to

demonstrate the utility of a method based on Kelly's per-

sonal construct theory for mapping and quantifying attitudes

about energy issues.









Kelly's personal construct theory views human thought

and learning as a process of construing. In this view indi-

viduals view ideas about the world which give meaning to

information and experiences. These constructs are bipolar

so that new information is seen as like one end of the polar-

ity and unlike the other. There are also constructs about

constructs at various levels. Thus, people develop systems

of constructs relating to any particular area of their lives.

One measure of a person's effectiveness is how congruent his/

her construct system is with reality and consequently how

well it permits prediction of events. In this sense, Kelly

saw all persons as scientists making and testing models of

reality.

A major purpose of the study was to demonstrate the

utility of repertory grid analysis in identifying and quan-

tifying perceptual differences among groups. Repertory grid

questionnaires were administered to three groups: a group

of senior education majors at the University of Florida, a

group of senior mechanical engineering majors at the Univer-

sity of Florida, and a group of adult learners enrolled in

energy awareness or radio/T.V. repair classes at Santa Fe

Community College. These groups were expected to have sub-

stantially different energy-related perceptions because of

their widely different educational experiences. Results of

this study demonstrated that all three groups shared similar

awareness of energy issues, but that they construed these

awarenesses differently. Differences were also noted with









regard to perceptual complexity and connectedness but not

centrality. Factor analysis of the pooled repertory grid

data for each group permitted the identification of common

dimensionalities of construing among the group. The use

of factor loading scores as potential indicators of values

is described and an example illustrated. Suggestions are

made for the incorporation of information derived from the

repertory grid studies into community educational planning.

Strategies are described for directing dialog among dimen-

sionalities identified as critical by factor analysis. The

extension of the method to a broad range of educational

activities is discussed.















CHAPTER I
INTRODUCTION


The Problem and Its Context

The purpose of this study was to examine the perceptions

of undergraduate students in engineering, education, and the

community about energy issues in order to map these percep-

tions and quantify similarities and differences among groups.

Data from this study provide groundwork for energy education

and the design of the study is a paradigm for studying group

perceptions about a variety of issues.

It has long been recognized that education is a life-

long process which takes place throughout the community

(Dewey, 1916; Cremin, 1976; Cross, 1979). Many educators

have identified the multiplicity of educative forces and

institutions in modern society. Illich (1971, p. 47) was

perhaps the first to propose the development of "learning

webs--voluntary networks that would permit any student at

any time to gain access to any resource that might help him

define and achieve his goals." Cremin (1976, p. 27) sug-

gests that such networks already exist to some extent and

expands the definition of education to include any "deliber-

ate systematic and sustained effort to transmit, evoke, or

acquire knowledge, attitudes, values, skills, or sensibili-

ties as well as any outcomes of that effort."









Society and communities in the United States are replete

with efforts to inform and persuade which are both deliber-

ate and sustained. However, the extent that these efforts

are broad and systematic in nature is more a reflection of

cultural homogeneity than the result of planning and coordin-

ation. What may be needed for both efficiency and greater

effectiveness is not more educational efforts, but coordin-

ation of educational efforts to be mutually supportive rather

than duplicative or divergent.

The anthropological literature describes an institution-

alized network for the dissemination of information which

provides most of the formal ("technical" in the sense of

Hall, 1959) communication in our society. Major components

of this network are public agencies and institutions, the

media, and specific public interest groups. This network

is characterized by information that is processed in a cen-

tral locus and disseminated through a multibranching series

of pathways. This type of network is designated as dendritic.

The nature of a dendritic network is such that it is possible

to coordinate the flow and control the quality of information.

It should be noted that information flow is bidirectional

while control and coordination goes downward.

People living in communities develop a series of net-

works of communication which are characterized by random

paths in which information flows out from a locus of percep-

tion. This type of network is designated as radiative. A

radiative network operates in the spreading of rumors and









the development of "grass-roots" movements. Because of the

dynamic and continuously changing nature of this network

information flows cannot be coordinated nor can information

quality be controlled. Still this radiative type of network

is very important in the life of a community and probably

operates strongly to condition values and perceptions. In

fact, it may be this type of network that drives the statis-

tical convergence in the Delphi technique (Mayer, 1964;

Myers, 1979).

It would seem reasonable that the values and perceptions

of an individual tend to be determined by those implicit in

the communication in the dendritic network unless a differ-

ent perception has been spread through the radiative network.

Consequently, each type of communication will affect a given

individual to a varying extent at different times on a single

issue and on different issues at the same time (Robinson,

1950).

Concepts of network types and influences can be brought

to bear on a variety of current issues. Many studies have

demonstrated that under certain conditions (which are poorly

understood) people behave in a way which is inconsistent with

their announced values. For example, studies sponsored by

the United States Department of Energy have shown that while

most people strongly support energy conservation they tend

not to support this behaviorally (Public Energy Education,

1979). Thus while we can make inferences about the values

and perceptions an individual holds from an analysis of that









person's ethnic and socioeconomic background, educational

status, and/or formal associations, these influences are

likely to be useful only in the absence of other strong

influences conditioning that individual's behavior. In

particular the influence of a radiative network on specific

perceptions would appear to be exceptionally strong (Mayer,

1964). This may account for the type of observed discrep-

ancies referred to above.

It is possible to determine functional groupings based

on clustering of like perceptions and values (Kluckholm and

Strodtbeck, 1961). Correlating these functional groups with

institutional associations can provide insight regarding the

value structure of a community (Kluckholm and Strodtbeck,

1961; Robinson, 1950). All managed networks by necessity are

dendritic; thus, it may be argued that managed networking will

be most effective where radiative networks are weak or absent.

Because they lack lateral communication capabilities, dendritic

networks cannot provide the value reinforcement which takes

place in radiative networks. This effect may be approximated

by the coordinated actions of several dendritic networks on an

individual belonging to all of these networks.

Agencies assuming responsibility for providing public

information and clarifying values on issues of social impor-

tance can improve their effectiveness by precisely adapting

their program content and delivery methods to the character-

istics of target groups (Torda, 1978; Rosenstein, 1978;

Cremin, 1976; Dewey, 1938). Unfortunately, there is at this






5


time no generally accepted methodology for mapping the values

and perceptions of such target groups. Were such a method-

ology available, its use could lead to the effective deline-

ation of target groups and their characteristics.

If educational activities relating to areas of social

concern are to be effective in facilitating participatory

decision-making, they must meet four criteria: relevance,

timeliness, accessibility, and compatibility with group

values and perceptions (Public Energy Education, 1979).

Much research has been done regarding the meeting of the

first three criteria; relatively little work,however, is

available in the area of mapping and analyzing perceptions.

It is the purpose of this dissertation to demonstrate the

use of personal construct theory as developed by George

Kelly (1955) as a tool for mapping and contrasting percep-

tual and evaluative activities of individuals and groups.

Personal Construct Theory

George Kelly's (1955) introduction and development of

personal construct theory is based on a unique set of assump-

tions regarding the nature of mankind and the human universe.

Three assertions are made in relation to the universe.

Humanity knows the universe to be real and not the result of

someone's imagination; humanity knows that the universe can

be understood only by repeated encounters over a period of

time; and humanity knows that all events are interrelated.










Kelly's assumption that the universe contains real

events and objects is complemented by his assumption that a

person's internal events are as real. Thoughts about exter-

nal objects or events also have an internal reality as

authentic as the happenings themselves. Kelly believed our

knowledge about the universe is determined by the extent to

which we can interpret it. For individuals, this interpre-

tation gradually approaches a true picture of events by

successive approximations. Unlike the purely subjective

existential or phenomenological theorists, Kelly holds to

the possibility of accurate knowledge of objective reality.

Mankind in Kelly's theory is trying out his interpretations

or constructions for their accuracy in predicting the world

he is beginning to understand.

Secondly, Kelly believes that some aspects of existence

can only be made comprehensible by the notion of time.

Behavior is explicable only in the context of the future as

well as the present and the past in Kelly's world view (1955).

The third aspect of Kelly's personal construct theory, the

notion of an integral universe, is consistent with the

development of a system of ideas to bring coherence to one's

world view. Kelly tried to describe how systems like this

operate. The concept of an integral universe is both a phil-

osophical and a psychological statement. People are seen as

continually striving to construe relationships where there

were none perceived before (Bannister, 1968).









Kelly visualizes humans as distinguished by their ability

to not only respond to life's events but able to represent

their environment. Personal construct theory is based on

people knowing the world by means of the constructions they

place upon it. According to Kelly humans are bound by events

to the degree that they have the ability to construe them.

All present perceptions are open to question and reconsidera-

tion. Kelly's philosophical position has been called construc-

tive alternativism (Bannister, 1968).

Personal construct theory (Kelly, 1955) can be applied

to the analysis of physical environments as well as personal

perceptions. Kelly postulated that knowledge precedes behavior

and that a person's knowledge is developed over time by accu-

mulating and categorizing information. People develop images

of their environment in this manner. Kelly calls these images

constructs. Construing means to Kelly placing an interpreta-

tion upon. Moreover, "in construing, the person notes features

in a series of elements which characterize some of the elements

and are particularly uncharacteristic of others" (Bannister,

1968, p. 7).

The technique used to visualize an individual's constructs

contains four steps, each of which will be described in detail

in a subsequent section of this work. The first task is the

listing of relevant elements, i.e., observations; the second

is the generation of bipolar constructs from these elements;

the third is the location of each element in total construct

space; and the fourth is the rotation of axes to identify









factors, combinations of constructs which divide element

groups. The first two tasks require only listing, the third

requires generation of a grid or matrix of element-construct

relationships, and the fourth a computerized analysis.

The analysis of a grid is to make clear the structure

in the grid; analytical techniques therefore should not

impose the structure of the experimenter. If each factor in

the grid represented were a separate unrelated opinion of

the subject, no structure would emerge. However, all grids

developed by people have inherent structure. In other words,

the relationships among individual elements which are sus-

ceptible to mathematical analysis are representative of

basic psychological processes. Grid analysis, then, should

be applicable in principle to any area of perception and

cognition.

Application of Personal Construct Theory to the Study of
Environmental Perceptions

Deutsch (1972) used Kelly's repertory grid test to ask

subjects to apply a group of their own constructs to a set

of known environments. Graduate students in architecture

were compared with graduate students in other fields. Sig-

nificant differences between groups were found in verbal

content and organizational structure. The results of this

study suggested further work could be done on the factor

analytic procedures used as well as on the reliability of

the coding methods used.









Stringer (1976) used personal construct theory (Kelly,

1955) to facilitate public participation in planning to re-

build Wistow Hill "Triangle," a London neighborhood, in 1970.

The general feeling from the public was that some form of

remodeling should be attempted. Stringer was seeking a

general understanding of environmental value systems. Using

personal construct theory to allow individual differences to

emerge was essential to obtaining a group of elements repre-

sentative of individual construct systems. Instead of using

simple preference orderings, Stringer believed the meanings

of elements are more significant when observed in relation

to each other. Fewer assumptions by the researcher then need

to be made. Stringer felt the repertory grid technique

"integrated perception and evaluation" at a time when politi-

cal rallying made participation a reality. Attitudes are

changing according to Stringer and "doing" has become more

important than having. Change has also become of paramount

importance in the process of environmental perception (Kelly,

1977).

Application of Construct Theory to Energy Education

The energy issue, like many other issues of our time,

presents both technological and philosophical problems. As

a case in point, the population of Sweden enjoys a comparable

standard of living to that of the United States at about 60

percent the energy cost using substantially the same tech-

nology (Schipper, 1976). As important to discovering addi-

tional oil and alternate energy resources is the development









of cooperative group processes to deal with the critical

social issues. The role of education is to facilitate the

decision making process by communicating information and

clarifying values. Personal construct theory can provide

specific information to deal with perceptions and values of

the population as a whole and certain groups specifically.

The methodology used in this study has yet to be demon-

strated in the context of energy perceptions or other criti-

cal social issues. The purpose of the study was to demon-

strate that personal construct theory in this form is useful

in delineating the energy perceptions of three groups which

can be expected a prior to differ in attitudes. A group

was chosen with obviously different background so that

there would be a maximal likelihood that each might look at

the energy problem differently. A group of senior under-

graduate engineers specializing in energy conversion was

selected to represent a group of individuals knowledgeable

in energy use and transfer. The second group was composed

of senior secondary education majors chosen to represent

lay knowledge on the energy problem as well as specialized

awareness of the participatory nature of our society with

certain social and political orientations. The last group

were representatives of the community at large. These

students were registered at Santa Fe Community College

receiving credit for either a radio/T.V. or an energy cor-

respondence course. It was believed they demonstrated a

wide spectrum of attitudes.









Design of the Study

The sample population included approximately 75 under-

graduate students of the University of Florida and Santa Fe

Community College. The first group of 28 were senior mechan-

ical engineering majors studying energy conversion. The

second group of 32 were University of Florida seniors also

majoring in secondary education. The third group of 14 com-

munity members were chosen from two classes at Santa Fe

Community College which were known to be heterogeneous in

terms of age and occupations. All of the subjects were asked

to complete a three part listing based on Kelly's repertory

grid method. In the first part the subject was to list 20

elements related to the energy problem. In the second part

the subject chose from the first part pairs of factors which

were the same and one opposite forming a triad. Subjects

were encouraged to state how factors were the same or oppo-

site (from constructs). The third part consisted of a grid

with elements along the X axis numbered one to twenty and

constructs located on the Y axis lettered A through Z.

Respondents were asked to correlate each construct with each

element.

The repertory grid was administered to the engineering,

teacher, and radio/T.V. classes allowing approximately an

hour for completion. If more time was required for comple-

tion, the researcher extended the time as needed. The reper-

tory grid was mailed along with the mid-term exam to the









community group studying energy. The researcher's phone

number was included to encourage questions.

Analysis of the data included a number of steps. First,

all element responses were tabulated to ascertain how many

different elements exist and how many times each element is

repeated. Secondly, constructs were tabulated to determine

how many constructs exist and their repetitions. An element

by construct grid was developed using factor analysis for

each subject. All grid data for a particular group were

combined to form a composite grid of elements, constructs,

and their relationships for each group. Finally, the three

groups were compared for similarities and differences

(Bannister, 1968).

The following hypotheses were tested:

1. The three groups have not chosen different ele-

ments which can be tested by analysis of the frequency

distribution.

2. The three groups have not developed different

numbers of constructs which can be tested by an analysis

of construct responses.

3. The three groups did not show different numbers

of positive relationships between elements and constructs

which can be tested by grid.

4. Analysis of group data does not lead to different

principal components which can be tested by factor analysis.









Definition of Terms

Fundamental postulate "A person's processes are psy-

chologically channelized by the ways in which he or she

anticipates events" (Kelly, 1955, p. 46).

Construction corollary "A person anticipates events

by construing their replications" (Kelly, 1955, p. 50).

Kelly expressed his assumption that all men act as scientists

in this world (Ryle, 1975). Leman (1970, p. 65) examined

this corollary from a linguistic philosophy and suggested

that 'the characteristically scientific activity is an oper-

ation with language . and that the scientist's most

important problems have to do with the relationship between

language and extra-linguistic reality . ." Leman empha-

sized the 'making-sense-of' aspect of men's constructing of

themselves and reality.

Individuality corollary "Persons differ from each

other in their constructions of events" (Kelly, 1955, p.

55).

Organization corollary "Each person characteristi-

cally evolves for his/her convenience in anticipating events

a construction system embracing ordinal relationships between

constructs" (Kelly, 1955, p. 56).

Dochotomy corollary "A person's construction system

is composed of a finite number of dichotomous constructs"

(Kelly, 1955, p. 59).









Choice corollary "A person chooses for herself or

himself that alternative in a dichotomized construct through

which he or she anticipates the greater possibility for the

elaboration of her system" (Kelly, 1955, p. 64).

Range corollary "A construct is convenient for the

anticipation of a finite range of events only" (Kelly, 1955,

p. 68).

Experience corollary "A persons's construction system

varies as he construes the replication of events" (Kelly,

1955, p. 72).

Elements "The things or events which are abstracted

by a person's use of a construct are called elements. In

some systems these are called object's (Bannister, 1968,

p. 219).

Core construct 'A core construct is one which governs

an individual's maintenance processes" (Bannister, 1968,

p. 221).

Factor analysis 'Factor analysis is a multivariable

method that has as its aim the explanation of relationships

among several difficult-to-interpret, correlated variables

in terms of a few conceptually meaningful, relatively inde-

pendent factors" (Kleinbaum, 1978, p. 276).

Assumptions

It is assumed that the diversity within the sample

groups and among the sample groups is a minimum estimator

of community diversity. It is also assumed that it will be

possible to functionally define groups in later work by









identifying clusters of individuals having between group

differences equal to or greater than those demonstrated by

the groups in this study.

Delimitations

The scope of the study included three well-defined

groups with clearly identifiable educational characteristics.

Thus, in no way could the groups be construed as either a

random or a polar representation of the community at large.

Rather, the value of the study lies in demonstrating a novel

and effective perception mapping technique and demonstrating

the application of this technique to the identification of

functional groupings.

Limitations

The selection of a repertory grid with subjects respond-

ing with negative or positive answers limits the sensitivity

of perceptions of relationships. Using an intensity scale

for each relationship, however, would have made administra-

tion of the repertory grid an extremely tedious process.

Administration of the repertory grid format to groups pre-

vented acquisition of contextual information by the researcher

which might help clarify the dimensionality of constructs.

However, the avoidance of the interview technique reduced

the possibility of experimenter contamination of results.

Interpretation of principal components is a qualitative

rather than a quantitative measure of group differences.

By this method it will be possible to identify how groups

differ in perceptions, but not possible to directly compare

two sets of between group differences.









Organization of Dissertation

In the remaining four chapters of this dissertation

the literature pertinent to this study is reviewed, the

experimental and statistical methodology are described, the

results reported, and the significant implications of those

results discussed. A review of literature pertinent to

social perception research, personal construct theory, fac-

tor analysis, and cultural foundations of this study is

presented in Chapter II. The rationale for the selection of

personal construct theory and repertory grid analysis as

the method of choice is described in Chapter III. A

description of the groups studied and a discussion of the

statistical methodology used to analyze and interpret the

results also are included in this chapter. The numerical

results of the study are presented in Chapter IV. Raw data

summaries, descriptive statistics, and analytical results

were tabulated and interpreted. A discussion of the find-

ings in light of the original experimental intent is pre-

sented in Chapter V. Additionally, a review of the impli-

cations of this study will suggest extensions into network

identification and applications to other areas of educa-

tional research.

For the sake of brevity and convenience, the groups of

engineering students, education students, and community

college students will be referred to from time to time in

the text and tables as "engineers," "teachers," and "commu-

nity groups," respectively.















CHAPTER II
REVIEW OF LITERATURE


In developing and demonstrating the methodology of this

study, it has been necessary to draw upon the fields of phil-

osophical foundations of education, cultural foundations of

education, social perception research, personal construct

theory, and matrix methods of statistical analysis. The pur-

pose of this study differs from the work previously done in

that it was designed to provide a vehicle for meaningful user

input to the development of educational programs. Of course,

the problems of relevancy and of the roles of education have

been of concern to many educators. In the following section,

a summary of the philosophical, educational, and methodolog-

ical bases of this work is presented.

Analysis of Social Perception

If culture is communication as Hall (1959) has said,

then education may be thought of as the summated social tech-

nology of communication. Examination of this assertion leads

us to ask three questions: (1) What are the activities and

institutions which communicate and educate in a modern

society? (2) Who decides what is to be communicated and

how? (3) How do the people in a community perceive and

influence the educational process? In answer to these ques-

tions a map of the network of educative forces and activities









occurring in society will be found. Identification will

also be made as to where that network breaks down: in iden-

tifying and incorporating the perceptions of the groups it

serves.

Educational Roles in Social Attitudes and Perceptions

Education has grown from providing only literacy and

vocational skills to modifying social values through the

broad development of all functional aspects of citizenship.

It is important to view education in its broadest possible

context: in its role as interpreter and modifier of social

attitudes. It will also be necessary to identify the role

of education in serving the needs of non-traditional learners.

In analysis of social perceptions a number of educa-

tional philosophers have interpreted and conditioned social

attitudes and values. Dewey (1916) realized that cultural

transmission was education in its broadest sense. He pointed

to the difference between the education most individuals get

simply from the process of living and the deliberate educa-

tion offered by the schools. For Dewey, the advancement of

civilization was a process of complexification. While the

young of savage groups could participate in society by mere

incidental learning, the young of complex cultures could

gain the same degree of participation only with intentional

learning. Dewey's theory of learning was essentially a

theory of the school as agent of society. Dewey worked on

reconciling the polarity between school and society. This

gap has remained. Even in the educational reform of the










1960's, there was ambivalence as to whether schools should

be improved or abolished altogether. In the 1970's, Ameri-

can opinion swung from an overreliance on the school as an

agent of socialization to a widespread disenchantment of

community support for schooling. Dewey anticipated this

phenomenon in the year 1933 when all public institutions

were suffering from a lack of public confidence.

Silberman (1970, p. 2) was aware of the same dilemma.

"If our concern is with education, we cannot restrict our

attention to the schools, for education is not synonymous

with schooling." He saw the need to emphasize the many edu-

cating aspects in American society other than schools.

Silberman's task was to make recommendations on the educa-

tion of educators, but he found this difficult without an

understanding of what education would be like in the years

ahead. Thus, Silberman supported by the Carnegie Corpora-

tion, undertook a four-year study of schools and other educa-

tional agencies. This study, the purpose of which was to

understand basic directional developments and synthesize

these into a coherent program, is today viewed by some as a

theoretical statement of educational progressivism.

Illich (1971) recognized, as had Dewey, that many insti-

tutions and situations educate and that the school is only

one such. He proposed the development of "convivial" educa-

tional institutions called learning webs which were networks

permitting free access to any resource that might help a

student achieve individual goals. He suggested four networks:









reference services in educational subjects, skill exchanges,

peer matching, and reference services to education at large.

Illich was thought to have achieved in design the Dewian

utopia of 1933 in which no schools existed at all. "The

most utopian thing about utopia is that there are no schools

at all" (Dewey, 1933, p. 236). All people would learn what

they needed to know from informal association with others.

Cremin (1976) suggested that Dewey's formulations of

democracy and education be revisited in order to attempt a

redefinition of education. He questioned education as a

result of which Dewey emphasized institutional origins rather

than functions. The theory of education thus created posits

major educative agencies performing a linking role with

respect to other agencies and society. Cremin (1976, p. 27)

has defined education . as the deliberate, systematic,

and sustained effort to transmit, evoke, or acquire knowledge,

attitudes, values, skills, or sensibilities, as well as any

outcomes of that effort." He saw education as a process more

limited than socialization (as perceived by sociologists) or

enculturation (as perceived by anthropologists). This def-

inition nonetheless has moved beyond focusing on the schools

and colleges to identifying all persons and institutions

that educate--siblings, churches, family, friends, libraries,

museums, and others. Finally, the definition allows that

education may produce outcomes which may be anticipated or

may be unintended, with the possibility that the latter could

be the more significant outcomes.









Cremin (1976) defined the role of education by his con-

cept of configurations of education. He envisioned a multi-

plicity of institutions relating to one another within the

larger society. Relationships between and among institutions

he saw as political, pedagogical, or personal. He predicted

a correlative relationship between configurations of educa-

tion and social change or stability. Through the encultura-

tion of the young, configurations of education have maintained

social continuity and stability (Cremin, 1976). Educational

institutions also have played their part in catalyzing social

change according to Cremin.

At the individual level, persons have related uniquely

to configurations of education (Cremin, 1976) according to

individual experience and perspective. Cremin (1976) held

that educational life history of an individual begins with

the efforts of critical others in creating appropriate atti-

tudes and behaviors and results in the individual becoming

a self-directed learner.

Fromm (1968) held a compatible view in that he saw each

person to be sacred and (ideally) to be united with his/her

world. He expressed concern that our educational system

lacked quality despite its great institutionalization and

despite the large numbers of college graduates. According

to Fromm our educational system must become alive and re-

sponsive to each participant. He believed that it is time

for man to assert himself and make the technological society

human.









Leichter (1974) visualized education as a lifelong

process which can take place in a variety of settings and

needs to be understood in each of the settings. She assumed

education took place on numerous levels often simultaneously

and both learning content and process need understanding.

Leichter (1974, p. 239) conceptualized educative style as a

"set of characteristic ways in which an individual engages

in, moves through, and combines educational experiences over

a lifetime." Here it is assumed that educationally signifi-

cant others taught in childhood such relevant educational

attitudes and that these are supported or modified by addi-

tional experience. Leichter's notion (1974) of educative

style focused on continual change linked with the continuity

of an individual moving from institution to institution

within configurations and among configurations. Movement

of this kind by each individual resulted in the development

of individualized networks of education for each person and

resulted in the individual's selection of a variety of

learning activities which reflected that individual's con-

structs pertaining to knowledge and growth.

Kohlberg (1966) realized the difficulty of understand-

ing the educational process without a thorough picture of

maturation, learning, and development. He believed like

Dewey (1933) and Piaget (1973) that understanding of intel-

lectual content and cognitive processes was essential to

the development of moral judgment. Kohlberg's (1966) formu-

lation of six developmental stages of moral judgment









described the potential for continuous growth in individuals

over a lifetime. Kohlberg's moral stages were redefined in

1975 with each of them examined in terms of operating "moral

motives." This categorization was supported by longitudinal

and cross-cultural studies. Kohlberg believed that moral

reasoning was clearly reasoning and advanced moral reasoning

depended upon advanced logical reasoning. A person's logical

stage created a ceiling on the attainment of moral stage. To

understand the learner and the learning process, we must

observe reasoning about choice.

This kind of reasoning defined the structure an individ-

ual chooses and was reflected in configurations of critically

relevant other persons or institutions. Thus, if educators

can understand what an individual finds valuable and why he/

she finds it valuable, they can begin to conceptualize not

only the content of his/her moral judgment but also how he/

she is motivated to learn. On this basis it may be possible

in theory to provide a more supportive climate in which

learning can occur.

Education has now recognized its obligation to meet

needs of the non-traditional learner and is beginning to

address these needs. Knowles (1977) described a new tech-

nology of androgogy or education of adults based on the

premise that adults are critically different than children

in experiencing learning. Cross (1979) emphasized the need

for equal education for 30 to 80 year olds. Researchers

estimated that between 80 and 90 percent of the adult









population carried out one self-directed learning experience

every year. Surveys show between 17 and 32 million adults

are now participating in classes, workshops, groups, or

other organized educational structure. By the year 2000

Cross (1979) forecasts the United States will become an

adult culture with 57 percent of the population over thirty

years of age.

Cross (1979) has shown that education for adults is

elitist with certain populations significantly underrepre-

sented in organized learning activities. She argued that

brokering services and education information centers repre-

sented the greatest hope for shaping the "learning society."

The development of education and information services is

perceived by Cross to involve three steps: (1) collecting

information about the educational resources available,

(2) reaching the intended audience, and (3) assisting clients

to identify and obtain the appropriate learning opportuni-

ties. A strengthening of these three linking functions

will thus contribute immeasurably to the success of lifelong

learning.

The recognition that the learner, especially the adult

learner, must be viewed contextually has led to the important

conclusions by Cremin (1976), Leichter (1974), and Illich

(1971). It is reasonable to expect this matrix of inter-

relationships in which individuals are embedded to influence

and condition values and perceptions. Clearly values and

perceptions affect not only the choices of educational









participation but also the motivation for participation.

These influences are not generally seen as falling within

the purview of education.

Sociocultural Aspects of Attitudes and Perceptions

If, however, one could define functional groupings

within a community based on values and perceptions with

regard to a particular area of interest, it would be pos-

sible to meet the needs of that group for education and

information. A variety of anthropological studies have

demonstrated methodologies for identifying and mapping

social influences as well as for mapping attitudes.

Mayer (1964) directed his efforts toward describing

the networks of relationships among individuals in urban

settings. He observed that people living in communities

developed a series of networks of communication which were

characterized by random paths in which information flowed

out from a locus of perception. He saw such a network

operating in the spreading of rumors and the development

of "grass-roots" movements. Mayer (1964) knew that be-

cause of the dynamic and continuously changing nature of

this network, information flow could not be coordinated

nor could information quality be controlled. Still this

type of network, according to Mayer (1964), was very impor-

tant in the life of a community and probably operated

strongly to condition values and perceptions. In fact,

it may be this type of network that moved the statistical

convergence in the Delphi technique (Myers, 1979).









While Mayer emphasized intra-group networks, Kluckholm

and Strodtbeck (1961) looked at the problem from the perspec-

tive of entire social groups in his study of five South-

western cultures. Using interview techniques informants from

five juxtaposed cultures were questioned about their percep-

tions of social and physical environments. Group mean scores

for the different cultures were compared by a one-way analysis

of variance for several predetermined dimensions. The results

obtained included an accurate map of perceptions along the

predetermined axes. This mapping elucidated the causes of

a great deal of conflict in values among the various groups.

The need for better tools in observing individual

values was expressed by Craik (1970), in his comprehensive

review of environmental psychology, who devoted a section to

the discussion of personality inventories. He pointed out

the neglect by psychologists of items or scales for assess-

ing environmental dispositions in spite of individuals' strong

orientations toward the physical environment, He suggested

a number of environmental dispositions, such as Pastoralism

scale, an Eccological Perspective scale, a Luddite scale,

an Urbanite scale, and measures of environmental sensitivity.

McKechnie (1970, p. 320) developed the Environmental

Response Inventory (ERI) for measuring environmental disposi-

tions, which he defines as "the configuration of attitudes,

beliefs, values, and sentiments .. ." of the individuals

being tested. The Environmental Response Inventory included

items relating to the areas of pastoralism, conservation,









science and technology, urban life, rural life, stimulus

preferences, cultural life, leisure activities, the outdoors,

geographic and architectural preferences, and environmental

memories and knowledge. The responses were factor analyzed

and a series of different factors for men and women surfaced.

The ERI scales did have an interesting correlation with

numerous traditional personality measures, and with environ-

ment related behaviors such as membership in conservation

and agricultural organizations. This result was consistent

with the results of the anthropological work of Kluckholm and

Strodtbeck (1961) and Mayer (1964).

Numerous studies of energy perceptions have been done

which demonstrated that under certain conditions (which are

poorly understood) people behave in ways inconsistent with

their announced values. For example, studies sponsored by

the United States Department of Energy (Public Energy Educa-

tion, 1979) have shown that while most people avowedly sup-

port energy conservation, they tend not to demonstrate this

support behaviorally. Thus while inferences can be made

about the values and perceptions an individual holds from an

analysis of that person's ethnic and socioeconomic background,

educational status, and/or formal associations, these in-

ferences are likely to be useful only in the absence of

information regarding other strong influences conditioning

that individual's behavior. In particular the influence of an

associational network on specific perceptions appeared to be










exceptionally strong to Mayer (1964). This may account for

the observed discrepancies referred to above.

Hall (1959) explored culture by means of communication

theory (information theory) of the electronics laboratory.

Hall (1959) likened communication theory to shorthand for

talking about communication events such as phonetics of

language, orthographics, and telephone signals. He noticed

that the process proceeded in one direction--toward symboli-

zation. He saw an individual's speech as an arbitrary vocal

symbol used to describe something that had taken place or

might have taken place with possibly no actual connection

between occurrences and symbols. Because of the workings of

culture, Hall (1959) saw talking as a highly selective

process. He believed that no culture had discovered a means

for talking without emphasizing some events at the expense

of others. He saw writing, then, as a symbol of a symbol.

Using communication theory (Hall, 1959) took this process

still further. He noticed entire messages of various dura-

tions with some less than a minute and others extending

over years. The study of culture thus could include events

of short duration whereas the study of individuals or govern-

ments involved communication over longer duration. He

developed a system based on tripartite theory which included

three kinds of time: formal time, informal time, and tech-

nical time. Hall (1959) discovered that man had three modes

of behavior and that at any point in time one of the three

orientations would dominate although all three would be









present. Formal activities were taught by a mistake being

made and a correction suggested. Informal learning empha-

sized a model for imitation in which thousands of details

could be passed through generations without specific under-

standing of the rules. Technical learning was shared in

explicit terms from teacher to student usually preceded by

a logical analysis. In summary, the formal mode was a two-

way process while informal learning involved identification

of a model to follow. The technical learning rested with

the teacher.

Numerous trends of social perceptions have been influ-

ential in the evolution of social communication. Joseph

(1979) forecasted a transformation in education through

technology to accommodate the tremendous mass of information

necessary for adaption to a highly complex society. The

information explosion has forced the emerging development

of a "technology for education which included imbedding

increasingly capable, but physically small, micro-processor

logic, digital storage/memory, sensors, communications cir-

cuits and links, and eventually voice actuated and reply

mechanisms for creating convivially smart machines--which

are more humanistic for students" (Joseph, 1979, p. 1).

Meadows (1974) emphasized the information explosion as

it related to the future course of human society or human

survival. He saw that human survival could depend on the

effectiveness with which the population related to the

world's problems or solutions. Meadows (1974) has plotted









the dimensions of time and space and demonstrated how every

human concern can be located at a point on a graph. He

felt life was a challenge for people, many of whom use total

effort to provide daily for their families. Other individuals

act on problems further out in time and space so the pressure

they feel is of a community nature.

Meadows (1974) observed that a person's use of time and

space dependson his/her cultural orientation, immediate prob-

lems, and past experience. Before an individual moved into

a larger space he/she must have solved the more immediate

problems. He further concluded that the more difficult prob-

lems involved longer time commitments made by smaller numbers

of people. The danger perceived by Meadows (1974) was that

by limiting their perspective too much, individuals lose the

ability to cope with problems at state, national, or global

levels.

McLuhan's (1967) global village concept accentuated the

current use of a community of information. He saw the entire

world knowing about other lifestyles through the perceptions

created by television, movies, radio, and all forms of tele-

communications. A hamlet in the Andes or a traffic jam in

Los Angeles has been made the immediate experience of each

individual. Drucker (1969) compared the closeness between

continents today with the less relative closeness between

mansions and slums during the eighteenth century. The apt-

ness of the concept of a global village in terms of communi-

cation can readily be observed in current world events.









Cultural Foundations of the Study

The work reviewed in the preceding sections has traced

the role of education in modern societies, identified the

need for better understanding the role of values and percep-

tions in the learning process, and delineated evolutionary

trends in the expanding mission of education to serve the

needs of all groups in society. In this section we present

a rationale for utilizing knowledge of learner attitudes and

perceptions in the design and development of innovative

learning networks.

Human Cognition and Perceptions in Social Decision Making

Personal construct theory (Kelly, 1966) has demonstrated

the usefulness of looking at human and scientific endeavors

as sharing relevant similarities in that in both, people

pinpoint issues, observe issues, become intimate with prob-

lems, form hypotheses, test hypotheses, relate results to

expectancies, control investments so they can understand

what leads to what outcomes, carefully generalize, and broaden

dogma in view of experience. Kelly (1955) considered man in

this context and delineated the unifying concepts brought

forth through explaining and charting strategies of human

experience, both over individual lifespans, and over centur-

ies. Kelly's (1977) unique contribution to psychology was

the introduction of a single language for explaining human

process making the psychological nature of scientific in-

quiry able to provide new insights into man's potential and

the nature of science. Kelly (1955) did not claim to reach









a total understanding of human process but he did feel the

idea of man-the-scientist was worth exploring.

Kelly (1955) assumed humans were real and not just

existing in fantasy, that the human universe could be under-

stood only through a time perspective, and that the universe

was integral such that given complete knowledge and a wide

enough viewpoint, all events would be seen as interrelated.

Kelly (1966) further accepted that not only was the universe

real but also that human internal events were also real.

Therefore, an individual could come to understand his/her

world only to the extent he/she could interpret it by moving

toward an accurate awareness of events through successive

approximations. This theory bypasses the groundlessness and

subjectivity of phenomenological or existential analysis and

views people as able to test out their own constructs for

completeness of world prediction. Kelly argued that human

life events could be understood only as people acted on them

in view of the present, past, and future. He saw man in a

continual effort of construing relationships where none were

observed before in an attempt to incorporate what was for-

merly diverse into a more integrated universe. This repre-

sentational model of the world allowed humans, according to

Kelly, to make sense from the world and choose behavior in

relation to it. Therefore, Kelly (1966) refused to accept

any once and for all construction of the universe because

he viewed all current interpretations as subject to revision

or replacement.









In a further exploration of the characteristics and

nature of constructs, Bannister (1968) demonstrated that a

construct is a way in which some things are seen as alike

and others as different. For each person the basis of dif-

ference can only be appreciated when the contrast is under-

stood. The range of usefulness of an applied construct is

also necessary for complete comprehension as it is possible

to use similar discrimination while using different ranges

of convenience. A construct is an interpretation imposed

on events, not part of events themselves and thus constructs

are useful inventions, not a part of nature. A construct

was for Kelly (1955) a tool of discrimination and structur-

ing of events in anticipation of future possibilities.

Kelly (1966) suggested that a person could be under-

stood by clearly seeing his/her construction system. Each

person's constructs represented a network of avenues along

which he/she can move. When movement was necessary, each

person was presented with a number of dichotomous choices

each channeled by a construct. Therefore, each construct

represented a pair of rival hypotheses within a system.

Consider, for example, the construct represented by the

triad: automobile, airplane, and railroad train. A hypo-

thetical respondent might suggest that airplanes are like

trains and unlike automobiles in that the former are

usually scheduled and the latter usually not. This construct

clearly dichotomizes travel decisions according to scheduled

and unscheduled arrangements. Note that the same triad









could have led to an entirely different construct; e.g.,

one in which ground transportation was constrasted with

more rapid air transportation. This illustrates the need

for careful interpretation of the like-unlike axis elicited

from the respondent.

Each construct system limits a person's perceptions

beyond which he/she cannot perceive and, therefore, controls

his/her behavior. In relation to sociological cultural con-

structs, decisions arise from individual behavior and per-

ceptions held by groups of people. Two individuals holding

similar construct systems in both discrimination and range

can be seen as having similar sets of constructs. A role

of education is to create values and facilitate participa-

tion of individuals in their life. Education has tradi-

tionally provided information and identified desirable be-

haviors while ignoring the critical link provided between

individuals of similar construction systems. Furthermore,

if education could change perceptions or concepts, our

society as a whole might be more effective in meeting the

challenge of survival. An example of the usefulness of per-

sonal construct theory is in accounting for the difficulties

we find in convincing other cultures to make good use of

transported commodities such as surplus food grain. The

range of convenience of their constructs regarding food may

not include the particular material provided. A perhaps

apocryphal example is the resistance of Central and South

American Indians to the use of high-lysine corn in tortillas.









Their perception of quality corn was that it was yellow; thus

they rejected the white imported corn even though it was

nutritionally superior.

Before education can deal with the development of use-

ful constructs, it must map the perceptions and values of the

target groups. Otherwise, it will be extremely difficult to

enlist the active support of members of the target group who

do not see relevance in the educational program goals. Many

learning theorists have developed hypotheses with related

models. Rogers (1970, p. 158) believed "significant learning

takes place when the subject matter is perceived by the stu-

dent as having relevance for his/her own purpose. Speed of

learning is also influenced by relevance. Probably one-third

to one-fifth of the present time allotment would be sufficient

if material were perceived by the learner as related to his

own purpose." Combs (1974) believed human beings had a

natural ability to learn. He saw motivation in people to

better understand themselves and their world. This fascina-

tion with learning, for experimentation, and extension of

cognitive limits could be released under ideal environmental

conditions according to Combs (1974).

Piaget (1973, p. 70) stated "the general culture which

education is to transmit to the student cannot be restricted

to abstract formation without roots in the structure and real

life of the society as a whole but must consolidate the dif-

ferent practical, technical, scientific, and artistic aspects

of social intercourse into a more organic whole." Piaget









(1973) supported Kelly's (1966) observation that the con-

structs of each individual are real for that person and

that the universe with its system of interrelationships is

also real.

Combs (1974, p. 126) stated "people do not behave in

response to stimuli but, rather, to the meanings these stim-

uli hold for them." Kelly (1966) further described stimuli

as constructs being individually formed and meaningfully used

as guides to further behavior. Kelly's (1955) personal con-

struct theory can be considered a metatheory or a theory

about theories in that it accounts not only for the behavior

of observed individuals, but also simultaneously it accounts

for the activities of the observer-theorist. Kelly's use of

language to structure human approaches to understand events

is unique.

Kelly (Bannister, 1968) developed techniques for elicit-

ing and measuring personal construct systems leading to his

repertory grid as the most sophisticated. The idea that con-

structs are individual bipolar abstractions with precise

ranges of convenience used for an individual's world struc-

ture is considered in the procedure for eliciting constructs.

The significance of exploring and understanding an individual's

system of constructs was recognized by Kelly (1955) in his

elicitation of numerous constructs and his design of statis-

tical techniques in which assessment of links among constructs

was possible.









The repertory grid technique of Kelly (1955) to be

described more fully in the next section lends itself not

only to the mapping of the construct fields of individuals

but also to that of groups. Although architectural and

urban planners are thus far the only non-psychologists to

make use of the method, the limited data available reveal

much promise for its application to a broad range of social

research fields.

Stringer (1976) discovered some important moral and

political implications as well as technical and theoretical

ones while working in the fields of planning and architec-

tural design. He was placed in charge of a study to examine

people's perceptions of alternative environmental aspects of

planning proposals to rebuild a decaying Victorian shopping

center in South London. In seeking general understanding

of environmental value systems, he felt his success would

depend upon not pre-empting the focus of inquiry too soon.

A comparable set of environmental elements would enable re-

spondents to give a clearer picture than one element or a

disparate set. Stringer (1976) saw a range of responses as

more richly definable by their observed relation to each

other. He further envisioned the repertory grid as preserv-

ing individual construct systems by integrating perceptions

and evaluations. Meaning was defined by the grid structure.

The planning context provided a future orientation to a

person's construing which conformed with personal construct

theory. Public participation has been a potential rallying









cry lately but it is in the planning field rather than edu-

cation or social services that most examples can be found

according to Stringer (1976). He examined 200 individual

grids to determine which redevelopment proposal drawn up

in different map formats constituted what respondents con-

sidered "adequate publicity" for local planning.

Leff and Deutsch(1972) did a pilot study using a modi-

fication of Kelly's (1955) repertory grid in which they re-

quested individuals to apply a set of their own constructs

to a set of environments known to them. Graduate students

in architecture, urban planning or studies, and graduate

students in other fields construed their physical environ-

ments. The two groups had significant organizational and

content differences between the environmental verbal con-

struct systems and between the results of this study and

semantic differential studies. The results of the study

suggested that there were two types of subjects in the pro-

fessional group, ones who construed environments in terms

of more qualities than lay persons and ones who construed

environments in terms of fewer qualities. "The finding

that professionals bracket lay persons suggests that there

might be a cognitive source of interactional problems in

that they are concerned with different numbers of environ-

mental aspects" (Leff and Deutsch,1972, p. 289). The find-

ing that some architects use fewer constructs than lay per-

sons and non-architects could suggest that when professionals

with backgrounds interact with lay persons most probably









conflict will occur because the lay persons most probably are

not concerned with as many environmental aspects. The find-

ing also suggested architects will have problems interacting

with environmental planners and designers. Alexander (1964)

saw that professionals are socialized by their educations to

think about environments very differently than non-professionals

and that such a cognitive gap could be a major hindrance to

cooperation. He saw cognitive differences translated readily

into interactional problems when groups think differently

about problems.

Hershberger (1969) carried out a study to compare the

environmental cognitions of students in architecture and

other fields using the semantic differential method. Even

though the study did show group differences the semantic

differential method required all students to use the same

set of scales provided by the experimenter and as a result

inter-individual differences were masked. Furthermore,

associations at the aggregate level did not necessarily

reflect relationships existing at the individual level

(Robinson, 1950).

Repertory Grid Methods

The work described in the previous section has demon-

strated the utility of personal construct theory and the

repertory grid method derived from it for describing the

perceptions of individuals and groups. In this section

will be described the methodology used in relevant work

and the appropriate statistical measures will be discussed.









Certain essential components of the repertory grid and

its conventions should be considered first (Ryle, 1975).

Among them are a list of things to be compared, the ele-

ments, and a list of terms used to compare and contrast

them, the constructs. The first list was compared systema-

tically against the second list to create a grid of figures

which is where the name "repertory grid" comes from. Con-

structs are bipolar according to Kelly (1966); therefore,

constructs should be elicited in the bipolar form. The sub-

ject has to have a list of elements which can be compared

to a list of constructs and vice versa. A researcher's

first responsibility in carrying out the repertory grid is

to elicit from the subject a list of elements and constructs.

The more subjects provide their own elements and constructs

the more they reveal about themselves according to Kelly

(Ryle, 1975).

The first stage in constructing a test is to put together

a set of elements. The next step is to obtain constructs.

The more freedom that is allowed the respondent the more

valid their constructs. The classical method has the tester

randomly choosing sets of three elements from the element

list and asking the subject to describe all the similarities

and differences in the triad. The tester then writes down

these descriptions. Triads are used until no new constructs

are formed. Three elements are considered rather than two

based on Kelly's (1955) stand on the bipolar nature of con-

structs and the fact that this is the minimum to allow for









obvious similarities and differences and to define both poles

of the construct. Constructs are recorded in each subject's

words unless constructs are supplied to describe often used

judgments. If the respondent did not feel comfortable with

the supplied construct, it could be modified.

The tester now has a list of elements and a list of con-

structs from the subject. To complete the test a respondent

is asked to rate each element with each construct by dichoto-

mising, by ranking, or by rating. The form which the test

thus takes is the form simply of rows and columns. The form

of the grid described can be called a "standard grid."

Other modifications can be made as required by the mathema-

tical analysis of the repertory grid.

Let us consider the following hypothetical case in

which a graduate student is asked to respond to the follow-

ing elements each of which is a potential topic for a doc-

toral dissertation:

1. Correlation of Myers-Briggs Extraversion Scales

with height.

2. Comparison of bilingual and monolingual children

on performance in a new language.

3. Correlation of Myers-Briggs thinking, feeling

category with computer literacy.

4. Use of computer-aided instruction to teach mathe-

matics in the middle school.

Our hypothetical respondent is asked to create triads

from the above elements. In each case, one pair of elements









will be alike and the third will be different in the same

way the pair is alike.

Alike Different How

A. 1 and 3 2 1 and 3 used Myers-Briggs

B. 3 and 4 2 3 and 4 involve computers
in education

C. 3 and 4 2 3 and 4 approved of by
dissertation chairman
and 2 is not

Having created the above constructs, the respondent is asked

to fill in the grid using a 2-point rating scale. A + will

be placed in a cell representing a construct and element

having a strong association. A will be placed in a cell

for the construct and element having a strong negative asso-

ciation. The responses given by our hypothetical graduate

student appear below.
Elements

1 2 3 4

A + --

Constructs B + +

C + +


This simple grid represents a mapping of the values and per-

ceptions about the various potential dissertation topics.

The reason for the use of Kelly's (1955) repertory

grid in specific is to facilitate the communication of con-

cepts relevant to understanding individuals and groups

(Ryle, 1975). Ultimately a grid of rows and columns is too

complex to be intelligible to inspection; therefore some









form of analysis and display is required before any conclu-

sions can be drawn. The purpose of mathematical analysis

of a grid is to make obvious the structure in the grid, not

to impose the experimenter's expected structure upon it.

If each point of the grid represents an independent judgment

of the respondent, there can be no structure to reveal. Ryle

(1975) stated that all grids created by humans are character-

ized by visible structure such that there are relationships

between individual ratings which are open to mathematical

analysis and which also relate to psychological process.

Complete analysis of essential properties of a grid (or the

relationships between elements and interaction of elements

and constructs) demands, according to that author, computer

analysis. Bannister and Mair (1968) described methods which

can easily be applied to grids based on dichotomization of

elements. Computer analysis also allows for examination of

element relationships and construct-element interactions

which are only fully explored by such an analysis. Problems

of centrality versus extremity of ratings and a skewed dis-

tribution of elements on constructs can also be addressed.

Two elements receiving precisely the same rating on

every construct must be perceived by the subject as the same

or indistinguishable in terms of constructs chosen. Two

elements rated at opposite ends of a construct must be per-

ceived as highly dissimilar. Similarity of any two elements

can be estimated by the measure of the distance between any









two elements and also the degree to which they are related

as similar in relation to the constructs used.

Ryle (1975) treated analysis of constructs as similar

or different in the same way as the elements. The associa-

tion between two constructs based on all the elements is

given with a value of -1 or +1 in the table of construct

correlations. Assuming enough elements have been rated

against constructs, the correlations between constructs

could understand his world. Knowledge of these could help

the researcher explain future behavior.

A graph can be created from an analysis of elements

to represent conceptual space in which the meaning is indi-

cated by the constructs and the location of each element

in relation to these constructs and to the other elements.

Summary

The literature review of this section has demonstrated

the need for educators to accept a role in the development

of society to serve non-traditional learners outside of

school settings. Very little is known about the values

and perceptions of these non-traditional learners with re-

gard to a variety of social issues which education will

need to address. In fact, very little is known about

methodology for acquiring information about values and

perceptions in a reliable way. In limited studies done

by architects and planners, the repertory grid method of

personal construct theory appears to have much promise for

mapping values and perceptions of individuals and groups.





45


This study demonstrates its utility for such a mapping with

regard to energy issues and suggests ways the results can

be used to plan curricula for energy education activities.















CHAPTER III
PROCEDURES


Kelly's theory of personal constructs and recent appli-

cations of its associated repertory grid method to map envi-

ronmental perception have provided a potentially powerful

tool for analyzing and interpreting group perceptual struc-

ture. Any technique for mapping complex perceptions is it-

self necessarily complex because of the many decisions

involved in its application.

The construction, administration, and analysis of the

repertory grid instrument are described in this chapter.

The makeup of respondent groups and the reduction of the

acquired data into forms suitable for hypothesis testing

are reported. Finally, the statistical procedures for

analyzing the data and testing the hypotheses are described.

Selection and Identification of Respondent Groups

The rationale for group selection was to find groups

of individuals who might be expected a prior to be differ-

ent in energy values and perceptions. Further, a group was

selected to represent a community sample with no prediction

made about member's homogeneity of values and perceptions.

A class of senior education majors at the University

of Florida was selected to represent a group with socio-

logical sophistication but limited technological expertise.









To represent a group with technological and energy exper-

tise, a class of senior mechanical engineering majors at

the University of Florida was selected. Community members

were sampled by soliciting responses from participants in

two classes at Santa Fe Community College which were known

to be heterogeneous in terms of ages and occupations One

class consisted of individuals participating in a "course

by newspaper" on energy; the second was made up of persons

studying radio-TV repair.

Valid responses were acquired from 28 engineering stu-

dents, 32 education majors, and 14 community students. The

average age of the groups was 24 for the engineering students,

25 for the students in education, and 31 for the community

group. Table 1 summarizes the demographic information avail-

able for each group. Appendix B lists majors and/or occupa-

tions for the groups.

Design and Administration of Repertory Grid Instrument

The instrument used to elicit repertory grid information

is found in Appendix A. Respondents were asked to list fac-

tors they believed were important in understanding the energy

problem. Several forces were implicit in the design of this

form. First among these was the decision not to provide

respondents with a list of elements, but rather to permit

free-form generation of element lists by each individual.

Providing a list of elements forces respondents to limit

their thinking to the researcher's supplied scope. The

eliciting of free-form lists, on the other hand, makes












Demographic


Average Range
Age Age


Engineering 24 21-36

Teacher 25 21-50

Community Group* 31 19-68



*Respondents from energy class = 4;


Table 1

SCharacteristics of Groups



Male Female Caucasian


28 0 27

7 25 32

12 2 12



from radio-TV class 10.


Black


0

0

1


Oriental Indian


1 0

0 0

0 1









direct comparison of individual grids impractical. This

latter difficulty can be dealt with, in both concept and

practice, by a procedure which will be demonstrated later

in this section.

The choice of requiring exactly 20 element responses

was made for the following reasons:

1. It was necessary to have uniformity of element

list size in order to meaningfully compare the resulting

construct list sizes.

2. Leff and Deutsch (1972) found 20 elements to be a

workable number of elements in their study.

3. Twenty three-digit element identification fields

leave a reasonable amount of space on an 80-column IBM

card for demographic and identification information.

Instructions for the first part of the response were

made as brief as possible. Subjects were asked to "List

20 factors you believe are important to understanding cur-

rent world energy problems." No attempt was made to define

"factor" nor was an energy related example provided. Indi-

vidual responses ranged from single words such as names of

oil companies to complex relational phrases such as economic

or political influences. In most cases individuals were

able to complete this listing in 20 to 30 minutes.

Below is a sample of a single individual's response.

Note that two numbers appear to the left of each response.

The first is the assigned element code from the categorical

list of Appendix C which is described in the next section









of this chapter. The second is just the number of the ele-

ment in the list. The category number was assigned by the

researcher on two criteria. The first was the meaning of

the listed element. The second was derived from examining

the grid column of construct associations to achieve con-

textual understanding of the element's meaning. For example,

note that element 8 "rising gas prices" could have been

recorded as a sociopolitical factor, a resource/distribution

factor, or an economic factor. Column 8 of the association

grid below, however, shows positive associations with con-

structs B, C, and E, all of which suggest economic dimension-

alities. Negative associations are recorded for constructs

A and D which suggest resource and technology dimension-

alities. Thus the element was assigned to category 80

"economics." Note that the underlying structure of the

grid would have been presented equally well had the category

assigned been 30 or 31, providing all other similar responses

by other individuals were treated consistently. This is true

because the purpose of the study was to analyze the response

structure of the groups rather than to interpret the precise

meaning of the responses.

Provision was made for the elicitation of from 0 to 26

constructs on the form. In the study reported by Deutsch

(1972) a maximum of 29 responses were found. It was antici-

pated that because the list of elements in this study was

not provided, fewer constructs would be produced. This was,

in fact, the case with 17 being the maximum number from a









single individual. Each construct was defined by informa-

tion of two types: the triad of elements which provided

the poles of the dichotomy, and the organizing concept

("How") which provided the dimensionality of the dichotomy.

Thus, the instrument could distinguish between constructs

using the same triads, but different dimensionalities. For

example, "car A is like car B and unlike car C." In one

case the dimensionality might be: "Because A and B are

sports cars while C is a station wagon"; in another "Because

A and B will run while C will not run." No responses were

considered valid for constructs if both triad and dimension-

ality was supplied.

Survey of Energy Problems

I. List 20 factors you believe are important to under-
standing current world energy problems.

73 1. Arab oil problems
72 2. Nuclear energy processing
74 3. Gasohol
80 4. Rising electric bills
68 5. Cleaning up coal use
06 6. Windmill generators
42 7. "Turn out the lights" syndrome
80 8. Rising gas prices
69 9. Use of natural gas
74 10. Use of geothermal energy
06 11. Use of solar energy
30 12. 55 m.p.h. speed limit
34 13. World politics
73 14. Third world's desire for energy
21 15. Rising rate of energy consumption/per cap
72 16. Difficulty of finding new oil
30 17. Emergency building temperature restrictions
52 18. Building insulation
51. 19. Car pooling and car pool lanes
80 20. Increased costs of all transportation









An example of the instructions and response to the con-

struct section of Part II using the same single individual

as in Part I follows:

II. Find pairs of factors which are alike. For each
pair find a third factor that is different in the
same way as the pair is alike. Continue to do
this until you can find no more contrasts. (Most
people will have more than 5 and fewer than 25
contrasts.) For your convenience, simply record
the numbers of the factors in each contrast in
the blanks provided below. In the space along-
side each triplet briefly describe how the factors
are similar or different.

Alike Different How

A. 10 11 5 018 Ways to find new energy

B. 1 3 15 031 Politics of the situation

C. 12 17 18 028 Federal mandate

D. 6 11 9 018 Still experimental

E. 4 20 18 009 Price-to-pay increase

Most respondents required 20 minutes or so to complete this

portion.

Constructs were examined for triad content. First,

the triad was recorded as in the element list. If a partic-

ular element code occurred twice other than in the alike

pair, the element coding was reexamined. The original ele-

ment triad was looked at, and the dimensionality axis ("How")

applied. The construct was then assigned to a number on the

construct category list. This list appears as Appendix D.

If an appropriate category did not appear on the list, one

was created and a new number was assigned.









For example, construct C above has the original ele-

ments of "55 m.p.h. speed limit" (12) and "emergency build-

ing temperature restrictions" (17) contrasted with "building

insulation" (18). The dimensionality ("How") is "federal

mandate." What this is interpreted to mean is "12 and 17

represent politically-motivated conservation requirements

while 18 is a conservation measure which is not required

politically but may be of value." Construct number 28 which

has the dimensionality "conservation versus politics" was

assigned to this response.

In filling out the association grid, the choice was

made to seek only a + or a response for each construct-

element association. This was done for reasons of expedi-

ency as well as for theoretical reasons. The use of a rat-

ing scale to estimate the degree of association (relevance)

between each element and each construct would be useful in

observing the evolution of attitudes of an individual over

time. In this case, however, it was felt the time required

for filling out the grid would be inordinate for an indi-

vidual who had developed possibly 20 constructs. Most

persons were able to complete the grid in 15 to 20 minutes.

An example grid for the above elements and constructs follows.

Interpretation of the simple dichotomous responses

elicited is relatively easy. "Construct X relates in some

way to element K" (+) or "Construct X does not relate in any

I think significant to element K" (-). Note that the latter

interpretation is difficult to distinguish from "I didn't









III. Look at the grid on the page. Each square in the
grid has a small letter and number on it. The
numbers refer to your list of factors from Part I
and the letters to your list of contrasts from
Part II. Start with box Al. Does contrast A
apply to factor 1? If so, place a + in box Al;
if not, place a in box Al. Repeat the process
for box A2 checking whether contrast A applies
to factor 2. Continue until you have checked
your contrasts against all 20 factors.

Factors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

A - + + - + + - - -

B + + + + -- + + + + + + -

C + +- +- -+ + + + + -

D + ++ - + + + + -

E ++ + + + + + ++ + + + + + + + ++ + +


consider element K (construct X, or both) very important,

and therefore, didn't mention them." This point is concep-

tually central to the use of individual grids to study group

values and perceptions. It is on this basis that we can

merge non-overlapping element and construct responses into

group grids as discussed in the next section.

Data Reduction and Analysis

Inspection of the received responses revealed that

there was a great many repetitions of elements and, to a

lesser degree, of constructs. In many cases, generic ele-

ments were represented by specific examples; e.g., Texaco,

Exxon (oil companies) and Iran, Saudia Arabia (O.P.E.C.).

Both these phenomena were anticipated, and in fact, neces-

sary for meaningful analysis to occur.









Because of the redundancies, it was possible to com-

press the approximately 600 element responses and 250 con-

struct responses per group into an orderly and manageable

group grid. To accomplish this, a taxonomic listing of

constructs was developed and each individual's responses

were recorded according to that taxonomy. The taxonomic

listing is found in Appendices C and D. The more complex

constructs were less easily compressed by categorization;

nonetheless, the same approach was used to categorize these

responses. A listing of construct categories is found in

Appendix C.

Care was taken to retain the original meaning of the

responses. Thus, tests were designed to ensure minimum

information loss in the recoding of the responses. To

assure fidelity in the recoding of elements, recorded triads

were examined for cases in which a single element occurs

both as a member of the like pair and as the unlike element

as was described in the previous section. In such cases,

the coding was reexamined to ascertain if a better assign-

ment could be made. If not, a subdivision was created in

the taxonomy so as to preserve the degree of discrimination

in the original response. After recoding, less than 1 per-

cent of element responses contained the same element in

more than one occurrence.

Consistency of interpretation of constructs was assured

by inspection of the row vectors of associations. In cases









where a single individual used the same construct two or

more times, the grid scores of each occurrence were com-

pared. If more than two grid cells were scored differently,

the constructs were reexamined, and, if necessary, reassigned.

A further check was the inspection of combined scores on

group grids to assure non-randomness of scoring distributions

across members within a group for a single construct.

The scoring of individual grids was merged into group

grid scores. While individual grids were no larger than

20 x 17, the group grids included all constructs found in

the group and thus were grids of 20 x 81 for the engineer-

ing class, 20 x 77 for the education class, and 20 x 44

for the community group. It is from the group girds that

the intercorrelation matrix of construct correlations was

calculated to be placed in the SPSS factor analysis routine.

Ordinarily, one would provide raw group data to the

SPSS factor analysis package. This program, in turn, cal-

culates the correlation coefficients between all construct

pairs and constructs the correlation matrix (Nie et al.,

1970). In this study, however, the raw data were submitted

to several purely data processing steps to permit checking

of group response consistency prior to calculating correla-

tions coefficients. A second reason for computing correla-

tions coefficient matrices for input to the SPSS routine is

that this calculation, which needs to be done only once, is

the most expensive calculation in the routine. By inputing









the correlation matrices, a variety of alternative computer

runs could be achieved relatively inexpensively.

Statistical Procedures and Tests of Hypotheses

The data acquired from the respondent groups were

analyzed and statistics calculated to test the four hypotheses

stated in Chapter I on page 12. It was anticipated that dif-

ferent groups would focus on different aspects of energy

concerns. One indication of such a difference is the range

of element responses elicited from a single group. A hypoth-

esis which can be easily tested is that three groups would

each contribute the same categorical elements with the same

relative frequency. The element frequency distributions

were printed out by the computer program and appear in

Chapter IV. Note that this hypothesis can be tested by super-

position of normalized frequency distributions.

It is not reasonable to believe that reference to a

particular element would be a randomly distributed variable

in the population as a whole. On the contrary, Kelly's

theory suggests that the element list an individual creates

is very much the result of their way of construing reality.

This individual set of characteristics is in turn a product

of a person's experience and learning. Consequently, it is

expected that common citation of elements should occur in a

group with some commonalities of experience and training.

It would thus be inappropriate to attempt to compute the

usual parametric test statistics and confidence intervals.









The data reported in Chapter IV do, in fact, support the

above prediction as will be described in detail later.

The number of constructs generated by individuals in a

group is a measure of the dimensional complexity of thought

in that group. The more constructs reported, the more ways

the respondent has of organizing perceptions. Consequently

one would wish to examine group differences as to the num-

bers of constructs generated by their members.

A convenient testable hypothesis is that the group

means are equal to each other. The implication of this

hypothesis is that its acceptance suggests all three groups

are representatives of a single population. This hypothesis

was tested by a one-way analysis of variance and the results

discussed in Chapter IV.

The underlying assumption of repertory grid analysis

is that the pattern of element-construct associations re-

flects the structure of the respondent's perceptions. An

individual who sees the world as a series of relatively

isolated events would be expected to have few positive

association responses. Thus the grid of such an individual

would be expected to show large numbers of (-) responses

and few (+'s). Conversely an individual who perceived a

great deal of interconnectedness would be expected to

report many positive associations. Such a person's grid

response should reflect a preponderance of (+) answers.

As a hypothesis susceptible to testing, it was pro-

posed that the percent of (+) responses of individuals in









each group were equal. This hypothesis was tested by a

one-way analysis of variance and the results are presented

and discussed in Chapter IV.

If it is desirable to assess the degree of interre-

latedness of perceptions of individuals, it is difficult to

interpret the structure of highly associated perceptions.

Interpretation of the structure of group perceptions becomes

almost impossible from simple inspection of grids. Fortu-

nately, methodology exists which can combine groups of

variables into composite variables called factors which

permit simplification of structural complexity. This method-

ology is termed factor analysis.

Factor analysis, as used here, proceeds through three

stages: computation of correlation matrix, interactive

estimation of communalities, and extraction and rotation

factors. Each of these steps and their associated analyti-

cal options will be discussed in turn.

In the simplest form, factor analysis attempts to

clarify structure in a set of data by creating linear

combinations of the variables involved. The proportion

of the group variance accounted for by these combinations

can be computed and, if the linear combinations are uncor-

related orthogonall), their relative significance is esti-

mated. The SPSS factor analysis routine operates on a

correlation matrix to find the single linear combination of

the variables present which maximally accounts for the total

group variance. This is called the first factor. A second









combination is then sought which will meet two criteria:

it must be orthogonal to the first factor, and it must

account for the maximum amount of the remaining variance.

A third factor is then generated with the same criteria,

i.e., orthogonal to the first two factors and accounting

for the maximum amount of remaining variance. This process

continues until there are as many factors as there are

variables.

The process described above suffers with regard to

interpretability of the factors derived. The correlation

matrix is generated by calculating the Pearson product

moment intercorrelation between the pooled rows of construct

pair responses. This generates a square matrix of construct

by construct correlations with l's on the diagonal. If we

are looking for structure in the matrix, we can focus on

the significance of the correlations. The diagonal row of

unit correlations reflects the fact that each construct has

reported for it both variance common to the whole group of

responses, and unique variance due to its own nature

(Harris, 1975). It is the former variance in which we are

interested. This quality is called the communality of the

variable. If accurate estimates of communality could be

used to adjust the matrix, then the factor scores would be

directly interpretable as reflecting structure.

The program used in this study accomplishes this com-

munality estimation by an iterative process (Nie et al.,

1970). On the first pass, the l's are replaced by the









single highest correlation in the column and the values of

all other correlations adjusted. Then the diagonal values

are replaced with the highest correlation in each adjusted

column. This process is repeated until there is no change

in the communalities between two passes.

Factors are then generated by the program until they

no longer explain a significant amount of variance. In the

option selected for this study, only factors with eigenvalues

of one or greater were printed. An eigenvalue of one is

equivalent to the normalized variance of one of the original

variables. Thus each factor printed is at least as useful

in accounting for group variance as any one of the original

variables.

The interpretation of the factors can proceed at this

point, but visualization of the significance of the factor

loading scores is difficult. A rotation algorithm is pro-

vided by the program to produce factors with loadings as

close to 0 or 1 as possible for each variable. The program

performs rotations of pairs of factors, retaining orthogo-

nality. If the rotation results in one or both of the

factor axes becoming colinear with a structural dimension

of the data, the projection of the data points on the other

axis becomes 0 as does the factor loading score. This is

easily interpretable; the colinear axis has no influence on

the variance in the data.

The program performs rotations maintaining orthogo-

nality and maximizing the numbers of O's and l's in the






62


factor loadings, thus generating more easily interpretable

factors. Further, it prints the distribution of data

points along each pair of factor axes. This information

is discussed in detail in Chapter IV and program output is

found in Appendix E.















CHAPTER IV
RESULTS


This study was designed to utilize data from three

groups of college students to demonstrate the use of reper-

tory grid methods in mapping values and perceptions concern-

ing energy issues. Two of the groups--education majors and

mechanical engineering majors--were relatively homogeneous

with regard to age, sex, and educational background. The

third group was made up of individuals taking a course by

newspaper on energy offered by Santa Fe Community College

and of persons taking a course in radio-television repair.

This last group was quite heterogeneous in composition

(for more detail, see Chapter III).

Repertory grids were elicited, constructs and elements

categorized, and individual responses merged into group

data. Pearson product moment correlations were calculated

between constructs and the intercorrelation matrices sub-

jected to factor analysis by means of the SPSS subprogram

FACTOR. Average number of constructs, means and standard

deviations, and one-way analysis of variance were performed

for each group using the SPSS subprogram ONEWAY. This pro-

gram was also used to compare number of positive associations

per construct for each group. Frequency of citation was

calculated for each construct and element for each group.









The products of the computations described above were

utilized in testing the following hypotheses:

1. The same elements and constructs are cited fre-

quently by different groups.

2. Groups generate the same number of constructs per

response.

3. Groups generate the same number of positive associ-

ations between elements and constructs per response.

4. Groups generate the same factors.

Each of these hypotheses will be analyzed, tested, and

discussed in turn.

Frequency of Element and Construct Citation

As might be expected, not all elements and constructs

occurred with similar frequencies in grid responses. In

fact, a few elements and constructs were very frequently

cited by members of a particular group, others were cited

much less frequently, and most were cited rarely. Tables 2

and 3 show the ten most frequently occurring elements and

constructs for the three groups. It is clear from the

tables that there is substantial agreement among the groups

on the importance of certain elements. Five elements--numbers

34, 41, 70, 80, and 81--appear in all five groups' listings.

A sixth--number 30--occurs in the top ten for the engineer-

ing and education groups.

It is interesting to note that these elements vary

widely in content. Elements 30, 34, and 41 relate to social

and political dimensions of the energy problem, while elements









Table 2

The Most Commonly Cited Constructs by Groups


Number
Construct Cited Meaning
Engineers

18 59 Conventional Technology/Alternate
Technology
87 44 Resources/Population Growth
44 43 Conservation/Public Attitudes
74 42 Education/Resources
10 36 Economics Cost/Public Needs
106 35 Politics/Industry
9 34 Economics Cost/Conservation
45 30 Conservation/Consumption
60 29 Attitudes/Planning
74 28 Pressure for/against International
Welfare

Teachers

9 104 Economics Cost/Conservation
18 75 Conventional Technology/Alternate
Technology
31 72 Politics/Resources
67 64 Individual Behavior/Public Politics
7 49 Economics Cost/Available Resources
28 48 Conservation/Politics
75 47 Education/Resources
44 43 Conservation/Public Attitudes
11 34 Economics Cost/Policies
5 34 Economics Lifestyle/Resource Limit

Communi-ty Groups

60 28 Attitudes/Planning
48 23 Conservation/Regulation
18 23 Conventional Technology/Alternate
Technology
37 21 Technology-practical/unrealistic
115 17 Government/Population
45 16 Conservation/Consumption
81 16 Energy production/Consumption
65 14 Credibility of Information/Attitudes
80 13 Energy Needs/Luxuries
2 12 Economic Factors/Technological Forces









Table 3

The Most Commonly Cited Elements by Groups


Number
Element Cited Meaning
Engineers

80 83 Economics
34 72 Politics
81 63 Industry
100 62 Efficiency of Current
Technology
70 59 Resources
6 58 "Soft" Technology
43 55 Credibility of Information
30 54 Government Regulation
41 51 Social Attitudes
50 50 Conservation


Teachers

70 119 Resources
80 112 Economics
34 111 Politics
51 89 Transportation
69 78 Consumption
41 76 Social Attitudes
81 75 Industry
42 68 Lifestyle Expectancies
74 65 New Types of Fuel
30 65 Government Regulation

Community Groups

34 31 Politics
68 28 Pollution
80 26 Economics
70 26 Resources
75 24 Safety
69 22 Consumption
73 19 Import/Export
81 17 Industry
51 17 Social Attitudes
74 15 New Types fuel










70, 80, and 81 are associated with resources, economics,

and industry. Surprisingly, none of the groups was strongly

aware of technical factors such as energy efficiency, emerg-

ing technology such as fusion, or radical lifestyle change

options as determined by frequency of citation.

Construct citations can be interpreted as indicative

of group values. Since constructs are both dichotomous and

polar, there is a "preferred" pole implicit as well as an

"undesirable" emergent pole. Recall the example of Chapter

II "approved by the dissertation chairman." The preferred

pole is clearly the approved pole, while the emergent pole

is the not approved pole. The data in Table 2 reveal that

only one construct, number 18 (conventional vs. alternate

technology), appears among those cited most frequently by

all three groups. One interpretation of the combined fre-

quencies is that while members of all three groups perceive

similar factors as relevant to the energy problem, they

place distinctly different constructions on those factors.

The implication of this is that the groups hold different

values and priorities regarding this problem.

Numbers of Constructs Elicited

Since each respondent provided the same number of ele-

ments (20), it is possible to compare the numbers of con-

structs produced from the element sets and attach significance

to the result. The number of constructs produced is indica-

tive of the complexity which the respondent perceives in the

area under study. An individual showing few constructs thus









can be said to have a relatively simplistic view while an

individual who responds with many constructs either views

the area as highly complex or perceives a variety of unre-

lated bits and pieces of observations. This ambiguity will

be resolved below.

Table 4 presents the mean numbers of constructs for

the three groups along with ranges and standard deviations.

In this table are also found the analysis of variance data

for numbers of constructs. Note that the F value is signif-

icant at the 0.0000 level. Thus we can reject the null

hypothesis that all group means are equal with less than a

0.01 percent chance of error. To determine whether or not

all pairs of means are significantly different, a least

significant difference follow-up test was performed. This

particular follow-up was selected because of its exactness

for uneven cell sizes. Although it suffers from multiple

levels of criterion variables, this particular application

with only three levels of group membership is immune from

that weakness. The table reveals that all pairs of means

are unequal at the 95 percent level of significance. It

is thus possible to state with confidence that insofar as

number of constructs are concerned, the three groups

examined are significantly different.

Degree of Construct Association

It was noted above that a relatively large number of

constructs could be interpreted in two ways. One means of

resolving this ambiguity is to achieve a measure of the









Table 4

ANOVA Table for Number of Constructs


Source D.F. S.S. M.S. F F Prob.

Between Groups 2 180.6951 90.3475 11.462 0.0000

Within Groups 71 559.6424 7.8823

Total 73 740.3374


Group N Mean S.D. S.E. Min. Max. 95% C.I. for Mean

Education 32 8.5000 3.1315 0.5536 4.0 16.0 7.3710 to 9.6290

Engineering 28 6.3571 2.7650 0.5225 3.0 17.0 5.2850 to 7.4293

Community 14 4.3571 1.9457 0.5200 3.0 9.0 3.2337 to 5.4805

Total 74 6.9054 3.1846 0.3702 3.0 17.0 6.1676 to 7.6432


Multiple Range Test
(*) Denotes pairs of
Mean Group

4.3571 C

6.3571 E

8.5000 T


- LSD At the 0.050 Level Ranges: 2.82 2.82
groups significantly different at the 0.050 level
C E T


*

*


C = Community

E = Engineering majors

T = Education majors









"connectedness" of the respondents' perceptions. It will

be recalled that each construct was scored against all ele-

ments for association. For strict consistency, only three

associations are required as a minimum for each construct.

Positive associations in excess of this minimum number are

indicative of a perception of connectedness by the respon-

dent. It is this feature of the grid that permits calcula-

tion of the construct intercorrelations. An alternative

measure is to determine the average number of positive

associations per construct for each individual. To the

extent that persons report positive associations, they indi-

cate the degree of interrelationships they perceive. This

degree of interrelationship perceived is useful in designing

learning models. A more complete discussion will follow in

Chapter V.

Table 5 also reflects results of the one-way analysis

of variance for this parameter. The F ratio is indicative

of significance at a level of 0.0030. The null hypothesis

that the group means are equal can thus be rejected with

only a 0.3 percent probability of error. The LSD follow-up

tests show that all pairs of means are different at a con-

fidence level of 95 percent. From the information displayed

in Table 4, we can then conclude that (1) all groups show

a degree of interconnectedness in excess of the minimum

required for definition of constructs, and (2) the groups

are all significantly different with regard to the degree









Table 5

ANOVA Table for Associations per Construct


Source D.F. S.S. M.S. F F Prob.

Between Groups 2 326.6074 163.3037 6.318 0.8030

Within Groups 71 1836.1949 25.8462

Total 73 2162.8023


Group Count Mean S.D. S.E. Min. Max. 95% C.I. for Mean

Education 32 10.0470 5.3352 0.9431 1.7000 27.5002 8.1234 to 11.9705

Engineering 28 7.3679 5'.7138 1.0798 1.2000 30.7002 5.1524 to 9.5835

Community 14 4.4001 2.3406 0.6256 0.6000 6.9001 3.0487 to 5.7515

Total 74 7.9650 5.4417 0.6326 0.6000 30.7002 6.7042 to 9.2257


Multiple Range Test
(*) Denotes pairs of
Mean Group

4.4001 C
7.3679 E
10.0470 T


- LSD At the 0.050 Level Ranges: 2.82
groups significantly different at the 0.050 level
C E T


2.82


C = Community
E = Engineering majors
T = Education majors


* *









of interconnectedness of perceptions demonstrated. The impli-

cations of this observation for curriculum development are

discussed in Chapter V.

Results of Factor Analyses of Grids

Table 6 lists the unrotated principal components (fac-

tors) for the engineering student group, the eigenvalue

associated with each factor, the percent of the grid value

accounted for by that factor, and the cumulative variance

accounted for. Because the group grid is structured, only

41 factors are required to account for 100 percent of the

variance. Had there been no structure, all 81 potential

factors would have been required. Note that the first fac-

tor accounts for far more variance than any other. Tables 7

and 8 display the same information for the education student

and community groups, respectively. Again, perceptual struc-

ture is indicated by much of the variance being accounted

for by relatively few factors. The amount of variance

accounted for by the first factor can be interpreted as a

measure of centrality of thought, that is, the degree to

which a single dimensionality is perceived to structure the

perceptual field. Note that on this criterion engineering

students and community members score similarly while educa-

tion students are significantly different.

Tables 9, 10, and 11 display the factor structures for

the first ten factors for each group. Each row of these

tables can be thought of as the row of regression coefficients

for predicting the score of one of the variables (constructs)









TABLE 6
SUMMARY OF FACTOR STRUCTURE FUR ENGIHEERING STUDENT CROUP
C"JSTRJCT ESTIMATED FACTOR EIENVALUE PERCENT OF CUMULATIVE
CDMUNALITY URIANCE PERCENT
C002 0.61552 1 17.492 21.9 21.9
C004 0.64939 2 7.57143 9.5 31.3
C005 0.82320 3 6.65219 8.3 39.6
C006 0.50018 4 5.06291 6.3 46.0
C007 0.83722 5 4.34184 5.4 51.4
CO08 0.86618 6 3.65097 4.6 56.0
C009 0.47453 7 3.37728 4.2 60.2
C010 0.9331 8 3.23098 4.0 64.2
COil 0.67399 9 2.73076 3.4 67.6
C012 0.74915 10 2.62047 3.3 70.9
C013 0.69756 11 2.25849 2.8 73.7
C014 0.86618 12 1.99421 2.5 76.2
C015 0.49723 13 1.92815 2.4 78.7
C017 0.79582 14 1.71036 2.1 80.8
C018 0.64405 15 1.58909 2.0 82.8
C019 0.85540 16 1.47138 1.8 84.6
C020 0.48224 17 1.36750 1.7 86,3
C021 0.48244 18 1.26276 1.6 87.9
C022 0.49246 19 1.09909 1.4 89.3
C023 0.52421 20 0.91551 1.2 90.5
C024 0.46245 21 0.84537 1.1 11.5
C025 0.74742 22 0.80370 1.0 92.5
C026 0.37538 23 0.72572 0.9 93.5
C027 0.67399 24 0.64331 0.8 94.3
C029 0.79741 25 0.d0929 0.8 95.0
C030 0.79512 26 0.54703 0.7 95.7
C031 0.70119 27 0.52099 0.7 96.4
C035 0.80277 28 0.43228 0.5 96.9
C040 0.63522 29 0.364d9 0.5 97.3
C041 0.82320 30 0.31827 0.4 97.7
CO42 0.79823 31 0.27886 0.3 98.1
CO4 0.79741 32 0.25853 0.3 98.4
C044 0.72941 33 0.21631 0.3 98.7
C045 0.72843 34 0.19824 0.2 98.9
C046 0.88561 35 0.18460 0.2 99,2
C047 0.66023 36 0.15452 0.2 99.4
C08 0.81409 37 0.13924 0.2 99.5
COIf 0.48244 38 0.13128 0.2 99.7
C051 0.57327 39 0.11085 0.1 99.8
COdO 0.70498 40 0.0538 0.1 99.9
C063 0.59395 41 0.04677 0.1 100.0
C064 0.70680 42 0.01982 0.0 100.0
COd5 0.70371 43 0.01039 0.0 100.0
C066 0.72018 44 0.00002 0.0 100.0
CON4 0.65523 45 0.00001 0.0 100.0
C070 0.38378 46 0.00001 0.0 100.0
C071 0.94281 47 0.00001 0.0 100.0
C072 0.70680 48 0.00001 0.0 100.0
C073 0.85540 49 0.00001 0.0 100.0
C074 0.62088 50 0.00000 0.0 100.0
C075 0.72941 51 0.00000 0.0 100.0
C076 0.93934 52 0.00000 0.0 100.0
C077 0.42591 53 0.00000 0.0 100.0
CC80 0.38378 54 0.00000 0.0 100.0
C081 0.67198 55 0.00000 0.0 100.0
C082 0.74742 56 0.00000 0.0 100.0
C084 0.83722 57 0.00000 0.0 100.0
CC07 0.71039 58 0.00000 0.0 100.0
C091 0.63296 55 0.00000 0.0 100.0









TABLE 6
SUnMARY OF FACTOR STRUCTURE FOR ENGINEERING STUDENT GRUP


CONSTRUCT ESTIMATED
CDIOUNALITY

CO92 0.57793
C093 0.54763
C094 0.49024
C095 0.75037
CO0 0.58287
C0o8 0.4268
C101 0.75879
C102 0.72018
C103 0.80277
C104 0.62088
ClOd 0.75037
C107 0.54763
C108 0.48596
C109 0.42591
C111 0.69756
C112 0.72862
C113 0.81409
C11i 0.61942
C115 0.71039
Ci d 0.61489
Cl17 0.94281


FACTOR EIGENUALUE PERCENT OF
VARIANCE


0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.0000
0.09000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
-.00001
-.00001
-.00001
-.00001


CUMULATIVE
PERCENT


0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0
0.0 100.0








TABLE 7
SUMMARY OF FACTOR STRUCTURE FOR EDUCATION STUDENT GROUP


CONSTRUCT ESTIMATED
CODMUNALITY
C001 0.71852
C002 0.61368
C004 0.80156
C005 0.82333
C007 0.83039
C008 0.71852
C009 0.83039
C01o 0.72850
C01 0.82158
C01l 0.48052
C014 0.82031
C015 0.54364
Co01 0.70771
C017 0.49397
C018 0.72105
C019 0.74675
C021 0.70895
C022 0.67306
C023 0.68494
C024 0.43252
C025 0.51929
C026 0.89301
C027 0.86017
C028 0.73976
C029 0.46857
C030 0.68052
C031 0.77447
CO32 0.72231
C033 0.70388
C034 0.53491
C035 0.85391
C036 0.68268
C038 0.76571
Co04 0.58779
C042 0.45196
C013 0.61020
C044 0.69601
C045 0.74289
C046 0.67976
CO49 0.67976
C050 0.73377
C051 0.70895
CO0O 0.65499
CC62 0.49799
CC06 0.71967
CC06 0.49156
CC67 0.77447
C069 0.67949
C073 0.74616
C074 0.85178
C075 0.69922
C076 0.72805
C077 0.49229
C078 0.74504
C079 0.85391
C080 0.66051
CC81 0.63038
CC82 0.73976
C083 0.86017


FACTOR EIGE1iVALUE PERCENT OF
UARIANCE

1 22.2013 28.8
2 7.25164 9.4
3 5.82969 7.6
4 4.60265 6.0
5 4.04365 5.3
6 3.74168 4.9
7 2.99490 3.9
8 2.92703 3.8
9 2.55866 3.3
10 2.28323 3.0
11 2.05944 2.7
12 1.87529 2.4
13 1.59244 2.1
14 1.44934 1.9
15 1.39991 1.8
16 1.28586 1.7
17 1.14581 1.5
18 1.04233 1.4
19 0.97265 1.3
20 0.71313 0.9
21 0.66515 0.9
22 0.58672 0.8
23 0.53504 0.7
24 0.48325 0.6
25 0.40610 0.5
26 0.38434 0.5
27 0.36624 0.5
28 0.29558 0.4
29 0.25154 0.3
30 0.22411 0.3
31 0.17917 0.2
32 0.16439 0.2
33 0.14679 0.2
34 0.11022 0.1
35 0.09746 0.1
36 0.08717 0.1
37 0.03317 0.0
38 0.01244 0.0
39 0.00005 0.0
40 0.00001 0.0
41 0.00001 0.0
42 0.00001 0.0
43 0.00000 0.0
44 0.00000 0.0
45 0.00000 0.0
46 0.00000 0.0
47 0.00000 0.0
48 0.00000 0.0
49 0.00000 0.0
50 0.00000 0.0
51 0.00000 0.0
52 0.00000 0.0
53 0.00000 0.0
54 0.00000 0.0
55 0.00000 0.0
56 0.00000 0.0
57 0.00000 0.0
58 0.00000 0.0
59 0.00000 0.0


CUMULATIVE
PERCENT
28.8
38.3
45.8
51.8
57.1
61.9
65.8
69.6
72.9
75.9
78.6
81.0
83.1
85.0
86.8
88.4
89.9
91.3
92.5
93.5
94.3
95.1
95.8
96.4
96.9
97.4
97.9
98.3
98.6
98.9
99.2
99.4
99.6
99.7
99.8
99.9
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0









TARLE 7
SUMMARY OF FACTOR STRUCTURE FOR EDUCATION SrUDEHT CROUP


CONSTIRCT ESTIMATED
COiMUNALITY

C084 0.82333
C087 0.49405
C088 0.41020
CO 0.49400
C091. 0.72281
COT9 0.80766
C094 0.82158
C094 0.82619
C097 0.65928
C098 0.82496
C099 0.76787
C101 0.dl0o5
C102 0.55699
C103 0.82619
C104 0.89301
C15 0.55554
C106 0.45196
C110 0.8650q


FACTOR E[GtEUALUE FERCEHT OF
UARIANCE


CUMULATIUE
PERCENT


40 0.00000 0.0 100.0
41 0.00000 0.0 100.0
62 0.00000 0.0 100.0
63 0.00000 0.0 100.0
64 0.00000 0.0 100.0
65 0.00000 0.0 100.0
6d 0.00000 0.0 100.0
67 0.0000 0.0 100.0
64 0.00000 0.0 100.0
69 0.00000 0.0 100.0
70 0.00000 0.0 100.0
71 0.00000 0.0 100.0
72 0.00000 0.0 100.0
73 -.00001 0.0 100.0
74 -.00001 0.0 100.0
75 -.00041 0.0 100.0
76 -.00001 0.0 100.0
77 -.00002 0.0 100.0










TARLE 8
SUMMARY OF FACTOR STRUCTURE FOR COMfUNITY GROUP
CONSTRUCT ESTIMATED FACTOR EICENUALUE PERCENT OF CUIULATIVE
CMUNALITY UARIANCE PERCENT
C002 1.00000 1 9.88800 22.5 22.5
C007 1.00000 2 4.62006 10.5 33.0
C009 1.00000 3 3.47582 7.9 40.9
C010 1.00000 4 3.25159 7.4 48.3
COll 1.00000 5 2.82219 6.4 54.7
C012 1.00000 6 2.63603 6.0 60.7
C013 1.00000 7 2.41758 5.5 66.2
C015 1.00000 8 1.92819 4.4 70.6
CO01 1.00000 9 1.65041 3.8 74.3
C017 1.00000 10 1.39879 3.2 77.5
C018 1.00000 11 1.27767 2.9 80.4
C021 1.00000 12 1.23360 2.8 83.2
C022 1.00000 13 1.07283 2.4 85.6
C023 1.00000 14 0.94358 2.1 87.8
C025 1.00000 15 0.89499 2.0 89.8
C027 1.00000 16 0.83737 1.9 91.7
C02q 1.00000 17 0.55595 1.3 93.0
C030 1.00000 18 0.52031 1.2 94.2
CO31 1.00000 14 0.46591 1.1 95.2
CO33 1.00000 20 0.38131 0.9 96.1
C03' 1.00000 21 0.31750 0.7 96.8
C035 1.00000 22 0.30508 0.7 97.5
C037 1.00000 23 0.25142 0.6 98.1
C040 1.00000 24 0.11222 0.4 98.5
C045 1.00000 25 0.17063 0.4 98.9
C048 1.00000 26 0.12654 0.3 99.2
C050 1.00000 27 0.10725 0.2 99.4
C040 1.00000 28 0.08092 0.2 99.6
CC65 1.00000 29 0.07204 0.2 99.8
Co66 1.00000 30 0.03859 0.1 99.9
C067 1.00000 31 0.03279 0.1 99.9
C075 1.00000 32 0.01545 0.0 100.0
C080 1.00000 33 0.01130 0.0 100.0
C081 1.00000 34 0.00311 0.0 100.0
C082 1.00000 35 0.00000 0.0 100.0
C083 1.00000 36 0.00000 0.0 100.0
C088 1.00000 37 0.00000 0.0 100.0
C092 1.00000 38 0.00000 0.0 100.0
C098 1.00000 39 0.00000 0.0 100.0
C099 1.00000 40 0.00000 0.0 100.0
C106 1.00000 41 0.00000 0.0 100.0
C115 1.00000 42 0.00000 0.0 100.0
C113 1.00000 43 0.00000 0.0 100.0
C119 1.00000 44 -.00001 0.0 100.0









TABLE 9
UNROTATED PRINCIF L COMHPNENTS FOR ENGINEERING STUDENTS
1 2 3 4 5 6 7 8 9 10


COSTRUCT
C002
C004
C005
COO

C008
009
C010
coil
C012
C013
C014
015
C017
Co018
C019
C020
C021
C022
C023
co24
C025
C026
C027
C029
C030
C031
C035
COIM
C040
C042
CO43
044
C045
C046
C047
C048
C049
C051
C060

Codi
COd5
Co46
C069
C070
C071
C072
C073
C074
COTS
C075
C076
C077
C081
COB1
C082
C087
C091
C092


0.475 0.496 -0.192 0.046 -0.071 0.156 -0.224
0.509 -0.007 -0.081 -0.460 0.288 0.083 0.085
0.425 0.293 -0.407 -0.327 0.262 -0.198 -0.158
0.325 0.097 -0.253 -0.035 0.121 0.187 0.039
0.578 0.370 0.026 -0.326 -0.030 -0.075 -0.008
0.560 0.259 0.536 -0.066 0.008 -0.281 -0.181
0.648 0.440 -0.119 -0.153 -0.304 -0.11 -0.045
0.715 -0.278 0.213 -0.462 -0.154 -0.001 -0.089
0.340 -0.216 -0.261 0.351 0.361 -0.151 -0.021
0.726 -0.306 -0.143 0.155 -0.197 0.297 -0.304
0.456 0.254 -0.243 0.340 0.290 0.089 -0.183
0.484 0.142 0.528 0.185 -0.088 -0.225 -0.174
0.241 0.461 0.092 -0.270 -0.213 -0.011 0.043
0.400 0.513 -0.165 0.139 -0.148 0.427 -0.315
0.622 0.186 0.330 0.367 0.084 0.207 -0.020
0.177 0.510 -4.024 0.341 -0.281 0.292 -0.229
0.014 -0.282 -0.O99 0.042 -0.080 0.102 -0.132
0.288 0.264 0.083 -0.334 -0.115 0.114 0.156
0.101 0.146 0.374 0.022 0.461 -0.030 0.105
0.444 -0.073 0.008 0.262 0.007 0.166 0.244
0.125 -0.119 0.296 0.379 0.313 -0.246 -0.096
0.285 0.371 0.442 0.206 -0.106 0.021 -0.012
-0.051 0.144 0.224 0.130 0.215 -0.091 0.050
0.245 -0.243 -0.193 0.372 0.260 -0.236 0.262
0.548 -0.139 -0.179 0.171 -0.355 -0.086 0.393
0.639 0.243 0.471 -0.099 -0.216 -0.224 -0.061
0.643 0.184 -0.173 -0.460 0.028 -0.086 0.134
0.368 0.121 -0.378 0.151 0.095 -0.390 -0.048
0.343 0.455 -0.337 0.323 -0.276 -0.123 -0.273
0.536 0.052 -0.327 -0.552 0.148 -0.098 -0.118
0.592 -0.371 0.442 0.176 -0.164 -0.154 -0.147
0.441 -0.171 -0.225 0.293 -0.205 -0.102 0.397
0.698 0.038 0.074 -0.043 -0.162 -0.163 0.161
0.678 0.025 -0.134 0.353 -0.120 -0.136 0.166
0.438 -0.610 -0.009 -0.174 -0.365 0.136 -0.175
0.324 -0.074 -0.312 0.243 -0.232 -0.229 0.284
0.450 -0.060 -0.336 0.345 0.009 -0.047 0.294
0.404 0.171 0.191 -0.113 -0.080 -0.020 0.246
0.104 0.042 0.646 0.225 0.408 0.021 0.042
0.506 0.160 0.592 -0.023 0.164 -0.124 -0.166
0.325 0.253 0.414 -0.286 -0.024 -0.251 0.140
0.062 -0.341 0.079 -0.187 0.015 0.494 0.110
0.393 -0.132 -0.310 -0.355 0.249 -0.188 -0.213
0.700 -0.277 0.072 0.222 0.022 0.067 0.011
0.415 -0.547 0.402 0.223 0.105 0.004 0.016
-0.041 -0.038 -0.049 -0.028 0.157 -0.088 0.058
0.484 -0.729 0.087 -0.017 -0.212 0.057 -0.274
0.145 -0.323 0.070 -0.374 -0.042 0.260 0.066
0.283 0.557 -0.187 0.254 -0.174 0.392 -0.175
0.630 0.057 0.359 -0.035 0.347 0.239 0.059
0.833 -0.159 -0.017 -0.024 -0.129 -0.042 -0.086
0.511 -0.705 0.028 0.008 -0.116 0.060 -0.243
0.154 -0.180 0.018 0.210 -0.433 -0.331 0.041
-0.115 -0.050 0.02q -0.047 0.391 -0.229 -0.107
0.493 0.078 0.359 -0.254 -0.261 -0.102 0.225
0.291 0.491 0.354 0.153 -0.162 0.158 -0.211
0.564 0.289 0.036 -0.277 -0.003 0.015 0.252
0.721 -0.121 -0.112 -0.095 0.056 -0.232 0.060
0.347 0.279 0.216 -0.402 0.146 -0.181 -0.074
0.454 0.292 0.416 -0.120 -0.030 0.237 -0.054


-0.176 -0.104 -0.024
0.113 0.101 0.116
0.150 0.041 -0.231
-0.242 -0.355 -0.272
-0.312 -0.068 -0.196
0.051 -0.217 -0.003
0.119 0.120 -0.090
0.083 0.114 0.053
-0.396 0.011 0.123
0.007 -0.045 0.022
0.205 -0.224 0.058
0.215 -0.108 0.029
-0.151 0.029 0.192
0.043 0.119 0.008
-0.073 0.033 -0.119
0.062 0.092 0.359
-0.019 -0.182 0.071
0.134 -0.012 0.210
-0.098 -0.080 0.148
-0.242 0.018 -0.238
-0.293 0.216 -0.089
-0.049 -0.017 -0.024
0.216 -0.007 0.034
-0.360 0.218 0.107
0.355 -0.124 0.037
0.195 -0.062 -0.107
-0.339 0.153 0.152
0.031 -0.132 -0.161
0.078 0.059 -0.140
0.263 0.252 -0.176
0.175 -0.080 0.107
0.218 -4.368 -0.116
0.109 -0.006 0.119
0.102 -0.216 -0.035
0.083 -4.250 0.051
0.347 0.240 0.119
-0.027 0.266 -0.144
0.101 0.064 0.392
0.028 0.046 -0.082
0.176 0.109 -0.120
0.167 -0.046 0.081
0.128 0.183 -0.185
0.212 0.270 -0.089
-0.286 0.138 0.377
0.134 0.131 -0.030
-0.120 0.112 0.371
-0.066 -0.080 0.082
0.208 0.355 -0.076
0.160 0.090 0.185
0.032 -0.047 0.154
-0.130 -0.062 0.141
-0.147 0.021 -0.040
0.225 -0.081 -0.194
0.077 -0.056 0.241
-0.065 0.330 0.034
-0.174 0.044 0.066
-0.328 -0.224 -0.187
-0.017 -0.145 -0.132
-0.227 -0.230 -0.048
-0.162 -0.063 -0.164









TABLE 9
UNROTATED FPRICIPAL CIMPXMENTS FUR ENGINEERING STUDENTS
CONSTRUCT 1 2 3 4 5 6 7 8 9 10


0.350 -0.190 0.327
0.262 -0.177 0.284
0.390 -0.328 -0.394
0.294 0.076 -0.473
0.298 0.105 -0.0O8
0.579 -0.523 0.039
0.621 -0.300 -0.24q
0.494 -0.074 -0.573
0.440 0.142 -0.207
0.548 -0.131 -0.267
0.16 -0.074 0.318
0.304 0.072 0.260
0.054 -0.169 0.10
0.334 0.273 -0.148
0.564 0.380 -0.157
0.647 0.177 -0.270
0.560 0.192 -0.252
0.707 -0.248 0.064
0.407 0.213 -0.320
0.543 -0.729 0.103


0.292 0.352 0.390 0.161
0.148 0.337 -0.006 0.155
-0.049 0.048 0.434 0.186
0.028 0.325 -0.138 -0.238
-0.218 0.334 -0.010 0.406
-0.269 0.135 -0.046 -0.213
0.154 0.228 -0.046 -0.294
0.055 0.166 -0.20 -0.108
0.075 0.358 0.211 -0.115
-0.134 0.311 0.316 0.198
0.015 0.224 0.334 -0.006
0.288 0.198 0.091 0.438
0.297 -0.144 -0.325 0.248
0.150 0.508 -0.204 -0.335
0.320 0.019 0.007 -0.073
0.293 -0.218 -0.115 0.095
0.067 -0.045 0.431 0.254
-0.222 -0.103 -0.033 0.097
-0.237 -0.061 0.055 0.364
-0.013 -0.202 0.029 -0.251


0.106 -0.093
0.020 0.261
0.099 -O.lb1
0.194. 0.138
0.134 -0.021
0.008 0.114
-0.344 0.037
0.075 -0.076
0.112 -0.321
0.322 -0.143
0.221 0.336
0.196 -0.270
-0.086 0.205
0.344 -0.114
-0.152 .3719
-0.027 0.400
-0.004 0.143
-0.351 -0.210
-0.364 -0.087
-0.043 -0.043


-0.154
-0.251
0.028
0.323
0.322
0.143
0.174
-0.048
-0.048
-0.137
-0.399
0.293
-0.176
-0.026
-0.094
-0.092
0.0998
-0.132
-0.084
0.081








TABLE 10

UNROTATED PRINCIPAL CifPONENTS FOR EDUCATION STUDENTS


CONSTRUCT
C001
C004
C005
C007
C008


C013
C001

C013
C014
C015
C014
C017
018
C019
C021
C022
C023
C024
C025
C026
C02?
C029
C029
C031
C032
C033
C034
C035
C036
C039
C040
C042
C043
C044
C045
C046
C044
C051
C060
C062
C06O
CO66

C073
C074
CO75
C076
C077
C078
C079
COSO
CC080
C082
C083
C034


2 3 4 5
-0.215 -0.304 0.333 -0.139
0.145 -0.340 0.139 -0.233
-0.389 0.154 -0.242 0.180
0.489 -0.299 0.246 0.227
-0.116 -0.147 -0.252 -0.063
0.040 -0.396 0.172 -0.209
0.007 -0.316 0.036 0.132
-0.365 -0.066 0.133 -0.117
0.095 -0.136 0.026 0.016
0.026 -0.262 0.181 0.109
0.572 0.347 -0.022 -0.349
-0.231 -0.420 -0.218 -0.045
0.310 -0.179 -0.080 -0.070
0.101 0.436 -0.01 0.296
0.235 0.031 -0.164 0.217
0.363 -0.270 0.100 0.436
-0.136 0.135 0.517 -0.014
0.357 0.484 0.222 0.357
0.314 0.268 0.024 0.604
0.042 -0.083 0.020 0.234
0.059 0.033 -0.075 0.434
0.488 0.544 0.035 -0.250
-0.366 0.229 -0.412 0.202
0.268 -0.1! -0.002 0.053
0.255 -0.503 -0.018 -0.104
0.253 -0.242 0.155 -0.042
-0.255 -0.032 -0.135 0.061
0.255 0.28, -0.083 0.781
-0.079 -0.069 -4.231 -0.037
-0.083 0.181 0.004 0.116
0.245 -0.011 -.13 -0.200
-0.159 0.108 -0.076 -0.278
-0.419 0.184 -0.133 -0.092
-0.004 -0.362 0.197 0.161
-0.155 -0.070 -0.290 0.168
-0.067 -0.134 0.115 -0.146
-0.257 0.291 0.652 -0.213
0.434 0.137 -0.095 -0.277
0.134 0.160 0.172 -0.276
0.427 -0.022 0.071 -0.154
0.343 -0.071 -0.193 -0.136
0.050 0.200 0.374 -0.060
-0.508 -0.021 0.442 -0.073
-0.213 -0.179 -0.313 -0.107
-0.340 -0.076 0.059 -0.017
0.225 0.144 0.404 0.119
-0.154 0.086 0.056 -0.107
-0.118 0.376 0.501 0.064
0.152 -0.094 -0.202 0.202
-0.474 0.312 -0.315 -0.004
-0.307 0.409 -0.079 -0.063
-0.137 -0.333 0.115 -0.119
-0.203 -0.263 0.056 0.112
-0.420 -0.1$9 0.227 -0.007
0.332 0.021 -0.355 -0.247
0.015 -0.334 0.342 0.402
-0.296 -0.318 -0.073 0.046
0.067 -0.551 -0.193 -0.079
-0.419 0.142 -0.588 0.025
0.312 -0.040 4.244 0.413


6 7 8 9 10
-0.080 -O.016 -0.025 0.29'7 0.0O0
0.315 0.203 -0.393 -0.133 -0.157
-0.087 0.026 -0.189 -0.009 -0.145
0.175 -0.006 -0.177 0.107 -0.043
0.159 0.172 -0.149 0.055 0.073
0.012 0.093 -0.105 0.037 -0.016
0.216 0.121 -0.062 -0.000 0.015
-0.178 0.081 -0.171 -0.115 -0.034
-0.097 0.012 0.112 -0.347 -0.023
-0.145 -0.229 -0.191 0.058 0.280
0.051 -0.036 -0.090 0.105 -0.112
-0.284 0.321 0.240 -0.155 0.026
-0.191 0.042 0.301 -0.090 -0.140
-0.300 0.171 -0.036 -0.227 0.257
-0.106 0.003 -0.070 0.100 -0.363
-0.046 0.186 0.193 0.046 0.014
-0.164 0.273 -0.263 -0.119 0.028
0.012 0.208 0.033 -0.107 -0.134
0.045 0.119 0.055 -0.174 0. 05
-0.093 -0.190 -0.003 0.234 -0.465
0.385 -0.358 -0.23 -0.192 0.120
-0.083 -0.019 -0.059 -".022 -0.330
-0.053 0.150 -0.115 0.196 0.031
-0.357 0.082 -0.012 0.004 0.040
0.020 0.117 -0.091 -0.166 0.010
0.088 -0.466 -0.161 0.087 0.314
-0.227 -0.103 0.292 0.059 0.008
-0.116 0.087 0.179 0.212 -0.Q25
0.010 -0.428 -0.074 -0.031 -0.000
-0.216 0.431 -0.253 v.030 0.11i;
0.518 0.241 0.140 O.Ii6 0.184
-0.210 0.053 0.108 0.023 0.042
-0.250 -0.133 0.019 -0.191 -0.013
-0.117 -0.171 0.254 -4.358 -0.207
0.421 -0.057 0.095 -0.259 0.190
0.456 -0.171 0.091 0.105 -0.177
0.111 0.271 0.00. -0.125 0.005
0.237 -0.269 -0.074 0.156 0.025
0.000 0.032 0.161 -0.105 0.299
-0.270 0.010 0.143 0.027 0.086
0.091 -0.198 -0.045 -0.M60 0.104
-0.084 0.037 -0.065 -n.228 0.254
0.013 -0.267 0.04 0..142 -0.105
0.279 0.523 0.159 0.040 0.080
-0.050 0.052 0.265 -).063 -0.091
-0.076 -0.178 0.061 0.210 0.231
0.087 0.073 0.249 -0.121 -0.185
0.184 -0.000 0.175 -;.131 0.222
0.236 0.240 -0.085 0.213 -8.024
-0.172 -0.121 -0.153 0.086 -0.017
0.041 0.077 0.202 -0.045 0.142
-0.095 -0.063 -0.091 .011t -0.198
0.583 0.038 0.134 -0.013 -0.131
-0.096 -0.130 O.C47 -0.171 -0.037
0.443 0.240 0.!62 0.033 0.131
0.084 -0.123 -0.330 0.081 -0.153
0.064 -0.320 0.161 -0.255 0.098
-0.256 0.188 0.248 0.166 0.059
0.011 -0.090 -0.059 0.121 -0.0il
0.095 0.228 -0.116 0.145 0.022








TABLE 10
URRUTrTED PRINCIPAL COPONENTS FOR EDJUCTION STUDENTS
CONSTRUCT 1 2 3 4 5 6 7 8 9 10


0.384 ..140 0.005 -0.021 -0.084
0.244 -0.114 0.224 0.420 -0.363
0.254 -.2389 0.035 -0.040 0.144
-0.106 -0.05C 0.438 -0.329 -0.152
0.359 -0.017 -0.291 -0.041 -0.018
-0.322 -0.250 -0.184 -0.155 0.060
0.230 0.205 0.011 0.142 -0.037
0.237 -0.078 -0.013 -0.152 -0.204
0.164 0.391 -0.076 0.114 0.137
0.230 -0.393 0.044 0.122 -0.163
0.121 -0.103 0.074 0.082 0.065
0.100 -0.281 -0.170 -0.226 -0.125
0.315 0.208 -0.031 0.171 0.14'1
0.524 0.111 -0.327 -0.123 -0.076
-0.194 0.093 -0.337 -0.041 -0.044
-0.084 -0.095 -0.109 -0.452 -0.062
0.410 -0.048 -0.165 0.028 -0.082


0.2`2 -o.145 0.226
0.31' .0093 0.054
-0.271 O.050 0.037
0.332 0.295 0.141
-0.026 0.061 -0.994
0.18 -0.226 -0.179
0.035 5.220 0.023
-0.193 0.329 -0.096
0.070 0.265 -0.194
-0.089 -0.287 -0.027
-0.615 -0.41* 0.049
-0.035 0.209 0.379
0.155 0.375 -0.013
-0.0CS -I.116 -0.150
-0.020 -..22 0.166
-0.149 .113 0.243
0.174 -.152 -0.349


0.154 0.029
0.422 -0.076
0.356 0.081
0.410 0.375
0.662 0.347
0.728 0.205
0.595 -0.512
0.563 -0.230
0.254 -0.585
0.513 -0.315
0.514 -0.083
0.526 0.088
0.407 -0.507
0.451 0.505
0.246 0.553
0.203 -0.216
0.351 0.625









TIfELE' II
UNROTATED PRINCTRFL COMtPOXNETS FOR COMMUNITY CROUP
CONSTRUCT 1 2 3 4 5 6 7 8 9 10
C002 0.572 -0.286 0.196 0.013 0.002 0.370 -0.221 -0.177 0.149 0.027
C007 0.442 -0.095 0.412 -0.410 0.205 -0.106 0.107 0.492 0.175 0.003
C004 -0.041 0.764 -0.124 -0.062 0.413 0.077 -0.131 0.016 0.162 -0.089
C010 -0.154 -0.071 0.079 -0.163 -0.047 -0.074 0.123 0.271 0.056 -0.292
COl0 0.633 -0.325 -0.136 0.308 -0.155 0.208 -0.266 0.217 0.155 0.008
C012 0.036 0.746 -0.121 -0.068 0.424 0.085 -0.160 0.068 0.178 -0.067
C013 0.672 0.339 -0.109 -0.244 -0.435 -0.167 -0.075 -0.205 0.131 -0.122
C015 0.342 -0.169 0.374 -0.263 -0.181 0.310 0.430 0.218 0.119 0.035
Col0 0.044 0.259 0.457 0.096 0.365 0.134 0.535 -0.011 -0.203 0.013
C017 0.194 0.268 0.310 0.218 -0.610 0.404 -0.137 -0.094 0.204 -0.186
C018 0.147 0.299 0.577 0.080 -0.041 0.399 -0.172 -0.337 0.010 0.333
C021 -0.193 -0.080 0.176 -0.252 -0.008 -0.187 -0.060 0.218 0.374 0.395
C022 0.672 0.339 -0.109 -0.244 -0.435 -0.167 -0.075 -0.205 0.131 -0.122
C023 0.273 0.217 0.676 0.344 -0.061 0.106 -0.049 0.160 -0.193 -0.202
C025 0.668 0.398 0.260 -0.164 -0.157 0.040 0.181 0.161 0.118 -0.026
C02? 0.479 -0.012 0.244 -0.208 -0.396 -0.116 -0.091 -0.227 -0.396 -0.178
C029 0.053 0.136 -0.079 -0.262 -0.280 -0.326 0.162 0.131 0.501 0.267
C030 0.387 0.006 0.483 -0.296 0.357 -0.034 0.039 0.387 0.053 0.100
C031 0.247 0.130 0.530 0.502 0.009 -0.047 -0.342 0.237 0.041 -0.296
C033 0.121 -0.193 -0.123 0.356 0.323 0.132 0.474 -0.337 0.266 -0.094
C034 0.680 -0.411 -0.135 -0.261 0.123 0.079 0.127 -0.025 -0.057 -0.146
C035 0.844 -0.077 0.004 0.058 0.211 -0.206 0.098 -0.003 -0.015 0.030
C037 0.365 0.604 -0.156 0.120 -0.032 -0.068 0.439 -0.094 -0.229 0.157
C040 0.095 -0.144 0.068 0.490 -0.044 -0.241 -0.371 0.080 -0.175 0.125
C045 0.500 -0.170 0.151 -0.088 0.252 -0.311 -0.630 -0.073 0.041 0.132
CO48 0.619 -0.010 -0.007 0.540 -0.008 -0.086 0.259 0.011 -0.185 0.300
CO'5 -0.145 -0.023 0.092 0.026 -0.056 0.047 -0.036 -0.283 -0.080 0.501
CC60 0.612 -0.320 0.1(3 -0.070 0.126 0.336 -0.036 -0.235 0.131 0.263
C065 0.620 -0.132 -0.340 0.486 -0.121 0.023 0.009 0.180 0.184 0.066
C066 -0.061 0.057 -0.516 -0.007 0.006 0.554 -0.075 0.424 -0.274 0.075
C047 0.428 0.348 -0.246 -0.079 -0.203 0.507 0.079 0.307 -0.247 0.120
C075 0.471 0.176 -0.288 -0.408 0.177 0.016 -0.271 0.033 -0.174 -0.100
C080 0.688 -0.378 0.114 -0.041 -0.163 0.282 -0.066 -0.029 0.001 -0.033
C081 -0.012 0.827 0.055 0.175 0.216 0.243 0.057 -0.167 0.173 -0.080
C082 0.731 -0.177 -0.362 0.186 0.162 -0.167 -0.051 0.069 0.084 -0.048
C083 0.677 0.058 0.156 0.203 0.261 -0.355 0.062 -0.028 -0.259 0.117
C088 0.603 0.184 -0.061 -0.467 -0.151 -0.304 0.207 -0.165 -0.243 -0.012
C092 0.599 0.531 -0.207 0.151 -0.240 -0.324 0.028 -0.028 0.123 0.089
C008 0.680 -0.411 -0.135 -0.261 0.123 0.079 0.127 -0.025 -0.051 -0.146
C099 0.197 0.328 -0.206 -0.359 0.403 0.291 -0.489 -0.119 -0.030 0.028
C10 0.532 -0.221 -0.053 -0.260 0.337 0.312 0.003 -0.256 -0.066 0.047
C115 0.551 0.109 -0.557 0.361 -0.071 0.174 0.070 0.287 0.117 0.097
C118 0.602 0.055 0.071 0.350 0.428 -0.347 0.005 -0.054 0.076 -0.149
C19 0.018 -0.330 -0.154 0.110 0.177 0.148 0.235 -0.309 0.383 -0.326









from the factor score. The square of each coefficient then

estimates the proportion of the construct score variance

accounted for by the factor score. Note that these tables

have been adjusted for communalities; i.e., the variance

accounted for is common to the entire grid structure, not

unique to a particular variable. Thus the factor structure

matrix displayed in the tables is a structural depiction of

the innercorrelated variance. Further, variables having a

relatively large factor score coefficient are relatively

important in determining the meaning of the factor.

To maximize the possibility of distinguishing important

factor dimensionalities, Varimax rotation was performed on

the factors. This transformation forces the coefficients

to approach zero or one, making it easier to distinguish

heavy from light loadings. The Varimax rotated factors are

displayed in Tables 12, 13, and 14. The most significant

variables were extracted by ranking the coefficients in

decreasing order of magnitude and retaining those which

accounted for the first 50 percent of the factor eigenvalue.

Note that for the first factors, ten or more variables were

found as identified by circled coefficient values in the

tables. For subsequent variables, only a few significant

variables are required to account for 50 percent of the

eigenvalue (thus 50 percent of the variance) of the factor.

This is consistent with the literature which reports that

first factors tend to be quite general, loading heavily on

many variables, while subsequent factors tend to be bipolar,









TABLE 12
UARTIMX RUTTED FACTOR MATRIX FUR ENGIHEERIRG STUDENT CROUP
CUCSTRUCT 1 2 3 4 5 6 7 8 9 10
C002 0.019 0.112 0.605 0.407 0.178 -0.010 0.008 0.184 -0.146 -0.048
C004 0.184 0.102 -0.010 0.298 0.469 -0.060 -0.054 0.089 0.247 0.425
C00l -0.093 0.042 0.118 0.332 0.769 0.060 -0.051 0.125 -0.004 -0.039
C006 0.072 -0.098 0.090 0.499 0.075 0.024 0.054 0.434 0.005 -0.144
C007 0.048 0.295 0.205 0.732 0.239 -0.008 0.016 -0.027 -0.030 0.005
C008 0.169 0.833 0.040 0.242 0.108 0.040 -0.047 0.049 -0.163 0.034
CO09 0.044 0.279 0.480 0.414 0.372 0.356 -0.149 -0.150 -0.015 0.051
C010 0.638 0.341 -0.005 0.317 0.296 0.094 -0.150 -0.191 0.257 0.236
Co01 0.240 -0.113 0.025 0.089 0.149 0.011 0.695 0.236 -0.184 0.067
C012 0.751 0.031 0.418 0.113 0.135 0.200 0.080 0.214 0.149 -0.051
C013 0.032 0.148 0.410 -0.037 0.313 0.135 0.140 0.570 -0.118 0.042
C014 0.198 0.775 0.165 -0.044 0.009 0.195 -0.004 0.032 -0.085 -0.001
C015 -0.092 0.197 0.306 0.392 0.011 -0.046 -0.174 -0.215 -0.149 0.253
C017 0.008 0.064 0.839 0.146 0.140 -0.035 -0.097 0.132 0.101 -0.042
C018 0.144 0.511 0.422 0.179 -0.092 0.122 0.319 0.248 0.237 -0.002
C019 -0.047 0.015 0.816 -0.126 -0.140 0.018 -0.074 0.010 -0.161 0.143
C020 0.315 -0.156 -0.038 -0.075 -0.070 -0.007 -0.052 0.131 -0.069 -0.062
C021 -0.001 0.169 0.171 0.255 0.079 0.050 -0.302 -0.051 0.062 0.393
C022 -0.144 0.396 -0.11U 0.045 -0.046 -0.259 0.230 0.211 0.007 0.273
C023 0.147 0.060 0.140 0.327 -0.137 0.256 0.371 0.185 0.253 -0.046
C024 0.041 0.323 -0.071 -0.045 -0.032 -0.106 0.644 -0.045 0.005 -0.141
C025 -0.073 0.549 0.317 0.161 -0.237 0.029 0.084 -0.014 -0.005 -0.005
C026 -0.222 0.284 -0.048 -0.206 0.014 -0.001 0.032 0.111 0.017 0.084
C027 0.092 -0.135 -0.081 0.054 0.040 0.192 0.75 0.014 -0.062 0.165
C029 0.256 0.047 0.094 0.819 0.042 0.807 -0.080 0.126 0.048 0.222
C030 0.180 0.770 0.171 .266 0.126 0.280 -0.154 -0.093 0.023 0.025
C031 0.187 0.050 0.163 0.685 0.370 -0.004 0.127 -0.175 -0.019 0.400
C035 -0.010 0.033 0.063 0.173 0.434 0.348 0.210 0.187 -0.280 -0.145
C040 -0.079 0.059 0.629 0.083 0.272 0.338 0.042 0.007 -0.235 -0.283
C041 0.134 0.033 0.040 0.303 0.839 0.065 -0.143 -0.057 0.265 0.119
C042 0.637 0.568 -0.003 -0.111 -0.032 0.286 0.085 0.012 0.030 0.055
C043 0.188 -0.007 -0.043 0.146 -0.043 0.751 0.025 0.370 -0.050 0.039
C044 0.289 0.350 0.158 0.274 0.178 0.420 0.034 -0.019 -0.004 0.292
C045 0.234 0.218 0.233 0.198 0.098 0.606 0.198 0.307 -0.106 0.058
C046 0.850 -0.028 -0.000 0.087 -0.011 0.207 -0.245 0.059 0.068 -0.025
C047 0.057 -0.092 0.148 -0.153 0.218 0.683 0.113 -0.117 0.012 0.217
C048 0.137 -0.059 0.267 0.213 0.241 0.574 0.500 0.093 0.216 0.090
C049 0.062 0.285 0.151 0.150 -0.013 0.194 -0.044 -0.083 0.003 0.571
C051 -4.107 0.616 -0.107 -0.136 -0.167 -0.183 0.288 0.169 0.253 0.046
C060 0.108 0.810 0.096 0.055 0.177 -0.022 0.065 -0.014 0.209 0.022
C0o3 -0.052 0.580 -0.071 0.216 0.106 0.103 -0.190 -0.129 -0.025 0.277
C064 0.227 -0.143 -0.064 0.002 -0.071 -0.060 -0.125 0.039 0.632 0.034
C065 0.212 -0.017 -0.043 0.070 0.760 0.018 0.028 -0.064 0.155 0.055
C066 0.605 0.154 0.257 0.140 -0.021 0.108 0.492 0.030 0.016 0.373
C069 0.512 0.374 -0.169 -0.208 -0.053 0.179 0.302 0.056 0.352 0.055
C070 -4.013 -0.108 -0.057 -0.101 0.018 -0.155 0.167 -0.094 -0.158 0.320
C071 0.937 0.052 -0.090 -0.011 0.004 0.101 0.073 -0.004 0.083 -0.053
C072 0.261 -0.064 -0.098 -0.003 0.157 -0.018 -0.173 -0.220 0.559 0.151
C073 -0.110 0.019 0.837 -0.013 0.045 0.074 -0.122 0.155 0.006 0.113
C074 0.237 0.498 0.157 0.196 0.077 -0.098 0.150 0.324 0.270 0.393
C075 0.620 0.240 0.234 0.353 0.198 0.241 0.148 0.064 -0.044 0.186
C076 0.884 0.020 -0.047 0.046 0.086 0.063 0.207 -0.008 0.159 -0.136
C077 0.181 0.141 -0.069 -0.085 -0.034 0.569 -0.074 -0.137 -0.112 -0.257
C080 -0.053 0.089 -0.229 -0.227 0.212 -0.223 0.101 0.103 -0.204 0.178
C081 0.169 0.401 0.103 0.337 0.004 0.192 0.022 -0.441 0.211 0.318
C082 -0.030 0.450 0.55 0.190 -0.185 -0.137 0.024 -0.057 -0.055 -0.015
C084 0.025 0.230 0.079 0.794 0.058 0.097 0.028 0.319 0.021 0.114
C087 0.351 0.229 -0.040 0.404 0.369 0.341 0.157 0.171 -0.016 0.057
CO1q 0.023 0.408 -0.0O4 0.528 0.194 -0.211 -0.076 0.031 -0.164 0.077
C102 0.090 0.493 0.275 0.434 -0.101 -0.126 -0.049 0.073 0.217 -0.010






85


TABLE 12
UARINTAX ROTATED FACTOR MATRIX FOR EXCINEERIN STUDENT GROUP
CONSTRUCT 1 2 3 4 5 4 7 8 9 10

C093 0.140 0.304 0.004 -0.035 -0.173 0.039 0.244 0.4q9 0.483 0.079
CO4 0.021 0.306 -0.138 -0.023 0.066 0.064 0.409 0.064 0.429 -0.007
C095 0.378 -0.362 0.054 0.147 0.125 0.181 -0.030 0.438 u.303 0.229
C097 0.089 -0.128 0.238 -0.164 0.617 0.005 0.155 0.178 -0.204 0.259
C098 -0.098 0.046 -0.088 0.166 0.229 0.090 0.040 0.194 0.055 0.435
C101 0.715 0.141 -0.130 0.074 0.376 -0.043 0.131 -0.033 0.164 0.198
C102 0.598 -0.033 0.189 0.163 0.317 -0.057 0.527 0.174 -0.127 0.066
C103 0.218 -0.154 0.087 0.144 0.601 0.313 0.213 0.251 -0.189 -0.020
C104 0.102 0.098 0.221 0.176 0.302 0.005 0.037 0.426 0.032 0.039
C106 0.206 -0.038 0.029 0.207 0.381 0.235 -0.059 0.540 0.413 0.231
C107 -0.015 0.271 0.048 -0,081 0.074 -0.069 0.070 0.055 0.708 -0.134
ClO8 -0.040 0.332 0.009 -0.042 -0.239 0.273 0.112 0.434 0.025 0.489
C109 -0.030 0.098 -0.160 -0.027 -0.135 0.393 0.358 -0.239 0.036 -0.117
C111 -0.102 0.322 0.175 -0.162 0.614 -0.013 0.113 0.437 -0.147 -0.031
C112 -0.050 0.161 0.619 0.219 0.259 0.194 0.442 -0.044 0.012 -0.022
C13 0.094 0.056 0.510 0.214 0.273 0.528 0.371 -0.169 0.102 0.024
Cll4 0.094 -0.116 0.513 0.311 0.067 0.241 0.094 0.194 0.302 0.345
C115 0.551 0.187 -0.081 0.645 0.048 0.165 0.140 0.077 0.049 0.046
C116 -0.031 -0.187 0.111 0.721 0.086 0.178 0.091 0.063 -0.005 0.190
C117 0.947 0.089 -0.087 0.009 0.029 0.132 0.114 -0.025 0.102 -0.028








TALE 13
UVRIMAX ROTATED FACTOR 0TRIX FOR EDUCATION STUDENT GROUP
CONSTRUCT 1 2 3 4 5 6 7 19 10
CoC1 0.219 0.445 -0.019 0.559 0.007 0.496 0.123 -0.063 0.14? -0.058
C002 -0.060 0.226 0.161 0.098 -0.099 0.156 0.317 0.037 0.672 -0.130
C004 0.787 0.217 0.049 0.273 0.220 -0.022 0.036 0.119 0.199 -0.066
C005 -0.100 0.315 0.263 0.071 0.473 0.516 0.220 0.118 0.375 -0.206
C007 0.559 0.373 0.105 0.172 0.105 0.233 0.471 0.103 0.283 0.014
C008 0.095 0.508 0.116 0.261 -0.030 0.359 0.230 -0.045 0.333 -0.070
C009 0.254 0.476 0.045 0.262 0.289 0.308 0.376 0.212 0.343 -0.065
C010 0.487 0.420 0.048 0.459 -0.004 0.131 -0.030 -0.034 0.334 0.101
Co11 0.272 0.670 0.297 0.178 0.236 0.153 0.051 0.237 0.234 0.187
C013 0.239 0.331 -0.024 0.169 0.188 0.635 -0.067 0.075 v.204 0.073
C014 0.043 -0.049 0.804 -0.048 0.043 0.207 0.171 -0.084 0.062 0.007
C015 0.238 0.699 -0.231 6.010 -0.023 -0.121 0.171 -0.134 -0.005 0.103
C014 0.047 0.608 0.352 -0.032 0.163 0.088 0.101 0.021 -0.079 -0.022
C017 0.237 -0.050 0.127 -0.059 0.459 -0.084 -0.186 -0.111 0.003 0.465
C018 0.418 0.369 0.420 0.052 0.446 0.116 0.083 0.058 0.146 -0.317
C019 -0.067 0.487 0.062 0.022 0.647 0.230 0.196 0.040 0.044 -0.063
C21 0.046 0.128 0.109 0.524 0.209 0.085 -0.128 -0.242 0.439 0.239
C022 0.079 0.020 0.510 0.229 0.716 -0.071 0.008 0.049 0.140 0.142
C023 0.068 0.038 0.13, -0.038 0.773 -0.050 -0.002 0.204 0.081 0.129
C024 0.020 0.050 0.052 0.033 0.168 -0.016 -0.195 0.020 -0.071 -0.549
C025 0.175 -0.183 -0.047 -0.102 0.306 0.252 -0.025 0.593 0.292 -0.011
C026 0.145 -0.015 0.916 0.074 0.164 -0.016 -0.036 -4.077 0.077 -0.011
C027 0.830 0.072 -0.028 0.169 0.263 -0.014 0.245 -0.002 0.000 -0.010
C028 0.145 0.703 0.104 -0.062 0.254 0.395 0.051 -0.164 0.165 -0.034
C029 -0.152 0.442 0.005 -0.193 -0.037 0.197 0.181 -0.002 0.317 -0.035
C030 0.046 0.180 0.141 0.030 0.028 0.782 0.007 0.268 0.134 0.024
C031 0.591 0.574 0.112 0.347 0.186 0.191 0.060 0.149 -0.192 0.035
C032 0.173 -0.032 0.053 -0.049 0.405 0.000 -0.011 0.041 -0.254 -0.097
C033 0.511 0.280 0.189 0.048 -0.040 0.358 -0.017 0.372 0.051 -0.084
C034 0.390 0.063 0.023 0.169 0.333 0.026 0.134 -0.3,6 0.240 0.211
C035 0.016 0.068 0.272 0.091 0.062 0.184 0.748 0.148 0.033 0.098
C036 0.496 0.403 0.294 0.319 -0.046 0.135 0.120 -0.075 -0.034 0.175
C038 0.608 0.272 0.086 0.270 -0.082 -0.041 -0.184 0.119 0.002 0.183
C040 -0.077 0.644 0.011 0.159 0.145 0.037 -0.221 0.324 0.100 -0.059
C042 0.219 0.089 -0.147 -0.049 0.066 -0.026 0.310 0.539 0.077 0.173
C043 0.088 0.197 0.213 0.410 -0.067 0.222 0.328 0.400 0.121 -0.233
C044 -0.044 0.037 0.206 0.807 0.040 -0.050 0.024 -0.07.5 0.284 0.320
C045 0.228 0.094 0.671 0.041 0.032 0.498 0.302 0.227 0.072 -0.045
C046 0.107 0.256 0.381 0.285 0.035 0.280 0.173 0.045 0.005 0.436
C041 0.030 0.416 0.452 0.002 0.179 0.328 0.047 -0.149 -0.057 0.140
C050 0.293 0.319 0.441 -0.100 0.084 0.441 0.223 0.243 0.172 0.057
C051 0.159 0.228 0.307 0.407 0.227 0.287 -0.054 0.043 0.261 0.451
C0o0 0.142 0.130 -0.090 0.741 -0.126 0.194 -0.193 0.152 -0.025 -0.118
C062 0.204 0.250 -0.164 0.080 -0.030 -0.217 0.661 -0.034 0.055 0.087
C064 0.241 0.401 -0.041 0.399 0.014 -0.074 0.053 0.117 -0.062 0.028
C066 -0.050 0.019 0.227 0.310 0.337 0.494 -0.082 -0.006 -0.092 0.125
C067 0.343 0.474 0.339 0.489 0.117 -0.036 0.200 0.234 0.0O6 0.061
C069 -0.079 -0.107 0.139 0.573 0.222 0.066 -0.034 0.202 0.002 0.405
C073 0.365 0.257 0.167 0.092 0.429 0.218 0.532 0.079 0.207 -0.152
C074 0.904 0.086 0.100 0.279 0.023 0.051 -0.024 0.072 0.001 0.031
C075 0.573 0.156 0.242 0.462 0.139 -0.021 0.202 0.16d -0.103 0.339
C076 0.306 0.566 0.117 0.343 -0.014 0.280 0.040 0.056 0.285 -0.200
C077 -0.025 0.162 -0.144 0.321 0.050 -0.014 0.419 0.461 0.168 -0.198
C078 0.339 0.555 -0.041 0.547 0.005 0.143 -0.115 0.240 0.167 0.029
C079 0.147 0.152 0.390 -0.115 0.035 0.096 0.768 0.162 -0.007 0.106
C080 0.034 0.196 -0.120 0.254 0.388 0.383 -0.093 0.152 0.447 -0.337
CC81 0.247 0.449 -0.179 0.140 -0.099 0.188 -0.039 0.477 0.023 0.066
C082 0.077 0.658 -0.130 -0.101 0.004 0.201 0.271 -0.215 -0.136 -0.105
CC83 0.839 0.087 -0.014 0.032 -0.043 -0.050 0.196 0.181 -0.086 -0.113
C084 0.030 0.180 0.152 0.203 0.683 0.298 0.225 -0.018 0.275 -0.059






87

TABLE 13
VARIMAX ROTATED FACTOR MATRIX FOR EDUCATION STUDENT 1ODUP
CONSTRUCT 1 2 3 4 5 6 7 8 9 10

C087 0.012 -0.032 0.220 0.192 0.137 -0.006 -0.087 0.160 -..234 0.412
C088 0.220 -0.012 0.161 0.237 0.219 0.095 0.184 0.672 -4.310 0.029
COC9 0.512 -0.060 0.221 -0.168 0.181 0.016 0.194 -0.085 0.147 0.058
C091 0.079 0.389 0.100 -0.129 0.588 0.378 -0.058 0.015 -0.175 -0.042
CO93 0.293 0.143 0.753 0.160 0.114 0.206 0.148 -0.038 0.061 0.080
C094 0.263 0.803 0.352 -0.066 0.032 0.087 0.197 0.107 0.119 -0.026
CC09 0.458 0.009 -0.004 0.719 0.077 0.133 0.119 0.134 -0.006 0.018
C097 0.632 0.001 0.202 0.314 0.075 0.266 -0.060 -0.047 -0.019 -0.183
C098 0.175 -0.055 -0.105 0.807 -0.055 -0.113 0.057 -0.076 0.003 -0.129
C099 0.667 0.098 0.117 0.056 -0.001 -0.108 0.042 0.429 0.133 0.131
C101 0.485 0.031 0.123 -.01Ol 0.129 0.042 -0.032 0.183 0.728 0.235
C102 0.517 0.138 0.218 -0.074 0.014 0.476 0.144 -0.114 -).123 0.219
C103 0.345 -0.111 -0.018 0.748 0.071 -0.018 0.243 -0.032 -.153 -0.014
C104 0.087 0.038 0.910 0.C87 0.104 0.091 -0.080 -0.050 0.055 0.181
C105 -0.195 0.171 0.390 -4.116 -0.064 0.568 0.214 -0.193 -0.027 -0.037
C106 0.397 0.190 -0.073 0.022 -0.122 0.179 -0.212 -0.290 0.013 -0.023
C110 -0.004 0.096 0.911 -0.112 0.202 -0.102 0.029 0.143 -0.058 -0.018










TALE 14
0URI2 X RCETATED FiCTOR MATRIX FOR CDOMnNITY GROUP


COHSTRUCT
C002
C007
C0019
C010
coil
C012
C013
C015
cold
C017
C018
C021
C022
C023
C025
C027
C029
C030
C031
CO3l
C033
C034
C035
C037
C040
C045
C048
C050
C060
C066
C067
C075
1080

C083
0c0a
0o92
1098

Clod
Co199
C115
C115
C118
c119


1
0.705
0.467
-0.198
-0.082
0.473
-0.134
0.215
0.440
0.004
0.003
0.209
-0.056
0.215
0.032
0.280
0.290
-0.156
0.435
-0.034
0.135
0.742
0.503
-0.178
-0.131
0.460
0.120
-0.041
0.775
0.197
0.068
0.220
0.417
0.715
-0.261
0.404
0.252
0.298
-0.121
0.742
0.383
0.793
0.125
0.204
0.217


2 3 4
0.063 0.069 -0.048
0.051 0.196 0.022
0.069 0.025 0.883
-0.047 -0.136 -0.090
0.048 0.260 -0.188
0.072 0.063 0.887
0.908 0.069 0.105
0.130 -0.235 -0.260
-0.123 0.129 0.151
0.358 -0.324 -0.026
0.049 -0.106 0.233
-0.132 -0.070 -0.0ll3
0.908 0.069 0.105
0.025 0.258 -0.008
0.570 0.120 0.185
0.582 0.080 -0.273
0.371 -0.070 -0.036
-0.073 0.265 0.150
-0.083 0.33 0.063
-0.136 0.169 -0.013
0.218 0.180 -0.169
0.337 0.622 -0.002
0.503 0.199 0.242
-0.149 0.477 -0.204
0.077 0.567 0.118
0.175 0.543 -0.220
-0.074 -0.063 -0.096
0.050 0.117 -0.065
0.167 0.409 -0.136
-0.269 -0.350 0.169
0.270 -0.231 0.222
0.319 0.101 0.375
0.203 0.082 -0.271
0.155 -0.052 0.758
0.232 0.560 0.001
0.252 0.766 -0.003
0.758 0.171 -0.020
0.726 0.390 0.217
0.218 0.180 -0.169
0.021 -0.067 0.717
0.077 0.098 0.138
0.185 0.229 0.103
0.135 0.805 0.184
-0.103 -0.042 -0.044


5 6 7 8
0.095 0.341 -0.117 -0.105
-0.087 0.096 0.309 0.253
0.020 0.003 0.086 0.007
-0.097 0.010 0.060 0.025
0.561 0.305 -0.299 -0.035
0.066 0.021 0.075 0.038
0.113 0.115 -0.131 0.028
0.086 0.212 0.488 -0.019
-0.165 0.128 0.807 -0.091
0.172 0.778 -0.050 -0.075
-0.174 0.566 0.206 0.052
-0.142 -0.069 -0.071 0.131
0.113 0.115 -0.131 0.028
-0.065 0.747 0.350 0.183
0.174 0.314 0.374 0.109
-0.190 0.191 0.036 0.292
0.058 -0.142 -0.055 -0.067
-0.137 0.139 0.373 0.295
-0.008 0.791 -0.041 0.101
0.096 -0.089 0.223 -0.764
0.134 -0.199 0.04' -0.091
0.186 -0.025 0.153 -0.061
0.304 -0.116 0.524 -0.012
0.116 0.231 -0.275 0.185
-0.198 0.082 -0.430 0.275
0.467 0.114 0.315 -0.085
-0.100 -0.070 0.031 0.073
0.092 0.123 0.057 -0.134
0.704 0.129 -0.150 -0.224
0.657 -0.191 -0.011 0.290
0.667 0.055 0.269 0.319
0.100 -0.245 -0.138 0.302
0.218 0.245 -0.032 -0.014
0.060 0.275 0.274 -0.188
0.425 -0.131 -0.174 -0.158
0.046 0.017 0.259 0.10q
-0.083 -0.280 0.223 0.197
0.354 0.075 0.008 -0.005
0.134 -0.199 0.044 -0.091
-0.013 -0.128 -0.221 0.244
0.030 -0.157 0.096 -0.090
0.876 -0.034 -0.055 -0.130
0.037 0.105 0.073 -0.233
-0.045 -0.059 -0.096 -0.706


9

-0.006
0.519
0.011
0.086
0.029
0.050
0.051
0.290
-0.00 2
-0.059
0.083
0.664
0.051
-0.151
0.229
-0.297
0.669
0.397
-0.061
-0.102
-0.079
0.044
-0.142
0.144
-i.083
0.077
0.104
0.034
-0.246
-0.151
-0.138
-0.047
-0.08?
-0.023
-0.073
-0.031
0.119
-0.079
-0.122
-0.137
-0.010
-0.042
-0.076


10
0.171
-0.419
-0.035
-0.435
-0.038
-0.050
-0.014
-0.151
0.033
0.109
0.599
0.073
-0.014
-0.075
-0.121
0.005
-0.017
-0.227
-0.187
0.084
-0.218
-0.051
0.195
O.161
0.104
0.288
0.572
0.322
0.001
-0.077
0.025
-0.150
0.005
0.158
-0.130
0.080
-0.058
0.090
-0.218
0.113
0.123
-0.028
-0.110
-0.138









loading heavily on only a few variables. These combinations

of variables are, of course, more easily interpreted than

the more general first factors (Nie et al., 1970).

The first ten factors for each group are listed in

Tables 15, 16, and 17 along the significant constructs deter-

mining their dimensionality. Note that similar dimension-

alities occur across groups with the appearance of a resource

dimension, a politico-technological dimension, and a politico-

economic dimension most prevalent. It is possible to locate

individual variables (constructs) with regard to these axes

of dimensionality by use of the factor loading coefficients

from Tables 12, 13, and 14. Figure 1 displays the location

of the most frequently cited construct across groups (con-

struct 18 "conventional vs. alternate technology") for each

of the three groups on the resource X politico-technological

plane. Figure 2 displays the location of this variable on

the resource X politico-economic plane, and Figure 3 locates

construct 18 on the politico-technological X politico-economic

plane for each group.

What is clear from the figures is that using similar

dimensionalities and constructs, the different groups per-

ceive the energy problem (at least in this regard) differently.

Strong loadings on a particular dimensionality can provide

important clues to the value placed on a particular construct.

The implications of this interpretation for curriculum devel-

opment are explored in Chapter V.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs