Group Title: quasi-experimental study to assess the effect of an undergraduate social foundations of education course
Title: A quasi-experimental study to assess the effect of an undergraduate social foundations of education course
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098302/00001
 Material Information
Title: A quasi-experimental study to assess the effect of an undergraduate social foundations of education course on selected student attitudes
Physical Description: xi, 133 leaves : ; 28cm.
Language: English
Creator: Bridges, Winston T
Publication Date: 1975
Copyright Date: 1975
 Subjects
Subject: Educational sociology   ( lcsh )
Students -- Attitudes -- Florida   ( lcsh )
Foundations of Education thesis Ph. D   ( lcsh )
Dissertations, Academic -- Foundations of Education -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 129-131.
Statement of Responsibility: by Winston T. Bridges, Jr.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00098302
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 - 000162662
oclc - 02715208
notis - AAS9011

Downloads

This item has the following downloads:

quasiexperimenta00brid ( PDF )


Full Text

















A QUASI-EXPERIMENTAL STUDY TO ASSESS THE EFFECT
OF AN UNDERGRADUATE SOCIAL FOUNDATIONS OF EDUCATION COURSE
ON SELECTED STUDENT ATTITUDES















By

WINSTON T. BRIDGES, JR.


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


1975













ACKNOWLEDGEMENTS


The writer is indebted to the chairman of his super-

visory committee, Dr. Vynce A. Hines, for encouragement,

patience, and counsel. Without his help this project could

not have been completed. Dr. Hal G. Lewis, Dr. Robert R.

Sherman, and Dr. Richard K. Scher served on the supervisory

committee. Their time, understanding, and ideas are deeply

appreciated.

Other faculty members who have assisted the writer

in this project include Dr. William B. Ware and Dr. Robert

L. Curran. Dr. Ware's advice on statistical methodology

was invaluable. Dr. Curran's interest and encouragement

are appreciated.

An expression of gratitude is due Dr. Melvin C.

Baker, Assistant Dean of the College of Education. Dr.

Baker's initial encouragement to the writer as a beginning

graduate student and his continued interest provided much

helpful support.

The writer wishes to acknowledge the special debt

owed to the late Bozidar Muntyan. His example as a concerned

intellectual, teacher, and friend has been of lasting in-

spiration to all who knew him.

Finally, the writer is grateful for the encourage-

ment and assistance from his wife, Patti, and his parents.

















TABLE OF CONTENTS


Page

ACKNOWLEDGEMENTS ..................................... ii

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

ABSTRACT .......................................... viii

CHAPTER

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

Statement of the Problem ............... 1
Significance of the Study ............... 4
Hypotheses ................................ 4
Limitations ............................... 6
Definition of Terms ..................... 7

II REVIEW OF THE LITERATURE ................... 10

The Critics ............................. .. 10
The Nature of Social Foundations
of Education ............................ 22
The Problem of Assessment ............... 28
What Should Social Foundations Study
Accomplish? ........................... 31
Conclusion ................................ 33

III METHOD ....................................... 34

Design ................................... 34
Collection of Data .... ................. 37
Instrumentation ............................ 46

The Scale of Democratic Attitudes
(SDA) .............................. 46
The Value Ranking Test (VRT) ........ 48
The Personal Beliefs Inventory (PBI). 50
The Teacher Practices Inventory (TPI). 50
The Personal Opinion Questionnaire
(POO) .............................. 50


Statistical Methodology .................











Page


IV RESULTS ..................................... .. 54

The Scale of Democratic Attitudes ........ 54
The Subscale of Procedural Agreements
of Democracy ........................... 58
The Subscale of General Statements on
Freedom of Speech .... ............ ... 58
The Subscale of Specific Application of
Freedom of Speech and Procedural
Rights .................................. 60
The Subscale of Political Equality ....... 60
The Subscale of Social and Ethnic
Equality .............................. .. 64
The Subscale of Economic Equality ........ 67
The Dogmatism Scale ...................... 74
The Personal Beliefs Inventory ........... 74
The Teacher Practices Inventory .......... 77
The Change in Relationship Between
General Statements on Freedom of
Speech and the Specific Application
of Freedom of Speech and Procedural
Rights .............................. 80
The Change in Relationship Between
Rankings of Freedom and Equality ..... 87

V SUMMARY, CONCLUSIONS AND IMPLICATIONS ...... 93

Summary ................................. .. 93

The Scale of Democratic Attitudes
and Subscales ........... .......... 93
The Dogmatism Scale ....... ........ .. 93
The Personal Beliefs Inventory ....... 94
The Teacher Practices Inventory ...... 94
The Relationship Between General
Statements on Freedom of Speech
and the Specific Application of
Freedom of Speech and Procedural
Rights .............................. 96
The Relationship Between Rankings of
Freedom and Equality .............. 96

Conclusions ............................. 97
Implications ............................ 103

APPENDICES ......................................... 110

REFERENCES ......................................... 129

BIOGRAPHICAL SKETCH ................................. 132

















LIST OF TABLES


Table Page


1 Characteristics of Students Taking the
Personal Opinion Questionnaire: Complete
Data 42

2 Characteristics of Students Taking the
Personal Beliefs Inventory and Teacher
Practices Inventory: Complete Data 43

3 Characteristics of Students Taking the
Personal Opinion Questionnaire 44

4 Characteristics of Students Taking the
Personal Beliefs Inventory and Teacher
Practices Inventory 45

5 Stepwise Regression Analysis for the
Scale of Democratic Attitudes Using
Complete Data 56

6 Stepwise Regression Analysis for the
Scale of Democratic Attitudes Using
Adjusted Data 57

7 Stepwise Regression Analysis for the
Subscale of Procedural Agreements of
Democracy Using Adjusted Data 59

8 Stepwise Regression Analysis for the
Subscale of Specific Application of
Freedom of Speech and Procedural Rights
Using Complete Data 61

9 Stepwise Regression Analysis for the
Subscale of Specific Application of
Freedom of Speech and Procedural Rights
Using Adjusted Data 62

10 Stepwise Regression Analysis for the
Subscale of Political Equality Using
Complete Data 63











Table Page


11 Stepwise Regression Analysis for the
Subscale of Political Equality Using
Adjusted Data 65

12 Stepwise Regression Analysis for the
Subscale of Social and Ethnic Equality
Using Complete Data 66

13 Stepwise Regression Analysis for the
Subscale of Social and Ethnic Equality
Using Adjusted Data 68

14 Stepwise Regression Analysis for the
Subscale of Economic Equality Using
Complete Data 69

15 Stepwise Regression Analysis for the
Subscale of Economic Equality Using
Adjusted Data 71

16 Coefficients of Determination and Group
Means Using Complete Data 72

17 Coefficients of Determination and Group
Means Using Adjusted Data 73

18 Stepwise Regression Analysis for the
Dogmatism Scale Using Complete Data 75

19 Stepwise Regression Analysis for the
Dogmatism Scale Using Adjusted Data 76

20 Stepwise Regression Analysis for the
Personal Beliefs Inventory Using Adjusted
Data 78

21 Stepwise Regression Analysis for the
Teacher Practices Inventory Using Complete
Data 79

22 Mean Scores on the Teacher Practices
Inventory Using Complete Data 80

23 Stepwise Regression Analysis for the
Teacher Practices Inventory Using
Adjusted Data 81










Table Page


24 Correlation Coefficients of Scores on
General Statements on Freedom of Speech
and the Specific Application of Freedom
of Speech and Procedural Rights Using
Complete Data 82

25 Correlation Coefficients of Scores on
General Statements on Freedom of Speech
and the Specific Application of Freedom
of Speech and Procedural Rights Using
Adjusted Data 85

26 Correlation Coefficients of Rankings of
Freedom and Equality Using Complete Data 88

27 Correlation Coefficients of Rankings of
Freedom and Equality Using Adjusted Data 90

28 Summary of Results 95












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



A QUASI-EXPERIMENTAL STUDY TO ASSESS THE EFFECT
OF AN UNDERGRADUATE SOCIAL FOUNDATIONS OF EDUCATION COURSE
ON SELECTED STUDENT ATTITUDES


By

Winston T. Bridges, Jr.

December, 1975


Chairman: Vynce A. Hines
Major Department: Foundations of Education


The problem of this study was to assess the effect

of an undergraduate social foundations of education course

on selected student attitudes, values, and beliefs. The

study sought to determine the contribution of a social

foundations of education course to a student's demonstration

of democratic attitudes as defined by a Likert scale, as

well as more specific attitudes dealing with democratic

procedural agreements, general statements on freedom of

speech, specific applications of free speech, political

equality, social and ethnic equality, and economic equality.

The study examined the contribution of social foun-

dations to the congruence of choices made by students with

regard to general statementsion freedom of speech and more

specific statements through comparison of correlation


viii










coefficients. The study also investigated the course's

contribution to the integration or logical organization

of the student's belief system by comparing changes in

correlations of rankings of freedom and equality.

Finally, the study explored the contribution of

social foundations of education to the student's demonstra-

tion of rational, scientific beliefs, experimentalist edu-

cational practices, and dogmatism. Influences by age, sex,

social class, and university grade point average were ex-

plored.

Pretesting and posttesting of intact experimental

and control groups were done with instruments designed to

measure dependent variables.i The study was designated

quasi-experimental because neither subjects nor treatment

were randomly assigned. The experimental group consisted

of 151 students in five different sections of undergraduate

social foundations of education. The control group con-

sisted of 84 students in sixldifferent sections of under-

graduate human development and learning. Data were col-

lected during the summer quarter of 1974 at the University

of South Florida in Tampa and St. Petersburg, Florida.

Statistical analysis was done using the stepwise

multiple regression technique with group membership and

time of testing dummy coded as independent variables.

Fisher's transformation to Z was used to compare the sig-

nificance of differences between coefficients of correlation.

The level of confidence used was p < .05.










Neither group membership nor time of testing were

significant except in the comparison of rankings of free-

dom and equality. This would suggest that further research

be undertaken to investigate the possible impact of social

foundations of education on attitudes about specific edu-

cational issues, the processes of attitude formation, and

the significance of variation in methods used in such

courses.

The results of this study indicated:

1. That students who have had an undergraduate social foun-

dations of education course appear to have a greater

understanding of the relationship between freedom and

equality than students who have not had such a course.

2. A direct relationship between university grade point

average and democratic attitudes.

3. An inverse relationship between university grade point

average and support for economic equality.

4. An inverse relationship between social class and support

for economic equality.

5. An inverse relationship between age and economic equality.

6. A direct relationship between age and social and ethnic

equality.

7. A direct relationship between university grade point

average and dogmatism.

8. A direct relationship between maleness and dogmatism.

9. A direct relationship between social class and rational,

scientific beliefs.










10. A direct relationship between maleness and rational,

scientific beliefs.

11. A direct relationship between university grade point

average and experimentalist educational practices.



/ Cha

/ Chairman

















CHAPTER I

INTRODUCTION



Statement of the Problem


The general problem addressed by this study is

the assessment of social foundations of education. Other

studies have approached this problem in a variety of ways.

Examples are Conant's impressionistic survey of teacher

education (1963), Martin's historical study (1968) and

Clarke's study of textbook content (1968). While these

studies may or may not provide sound evidence and reasonable

conclusions about teacher education and social foundations

of education, this study attempts to do so through examina-

tion of the impact of a social foundations of education course

on its students. More specifically, this is done by develop-

ing empirical evidence from actual course operation. There-

fore, the purpose of this research is to assess the effect

of an undergraduate social foundations of education course

on selected student attitudes, values and beliefs.

The study uses instruments which measure changes in

attitudes, values and beliefs through protesting and post-

testing students in various sections of social foundations











courses. A control group of similar students is used to

assess the significance of changes demonstrated by social

foundations students. The study could be described as a

pretest posttest experimental control group design. Since

intact groups were used and neither subjects nor treatment

were randomly assigned, the study has been designated quasi-

experimental to indicate that it does not constitute a true

experiment as defined by Campbell and Stanley (1963).

The social foundations of education course under

study is regularly taught at the University of South Florida.

It is required of all students receiving an undergraduate

degree from the College of Education and may be taken by

those students who wish to be certified as public school

teachers by the state of Florida.

The study will determine the contribution of the social

foundations of education course to a student's demonstration

of democratic attitudes as defined by a general scale of such

attitudes. This scale is subdivided into several components

which allow the examination of the student's acceptance of

the procedural agreements of democracy, general statements

on freedom of speech, specific applications of free speech

and procedural rights, political equality, social and ethnic

equality, and economic equality. Investigation of social

foundations' influence on democratic attitudes is thought to

be legitimate since it is said that such study ought to










provide professional educators with a sense of purpose or

vision (Sherman, 1974). Also, social foundations is said

to have an emphasis on educational decision making (Derr,

1965) and policy study (Raywid', 1972). Such an emphasis

necessarily involves the question of what should or should

not be which, in turn, is affected by knowledge of and

response to such major cultural universals as the democratic

ideal.

The study examines the contribution of the course

to the congruence of choices made by students with regard

to general statements on freedom of speech and more specific

statements. This is done by comparing the correlation of

scores on the pertinent subscales noted above. Also, the

study compares the change in correlation of rankings of free-

dom and equality as a result of having had social foundations

study. The purpose of this is to investigate the course's

contribution to the integration or logical organization of

the student's value system. Stanley (1968), Raywid (1972),

and Broudy (1972) have argued that foundations study may

make such a contribution.

In addition to the above, the study examines the

contribution of social foundations of education to the

student's acceptance of experimentalist beliefs and educa-

tional practices. It is assumed that this provides a valid

indication of a critical, rational scientific predisposition

in the student. Finally, the study explores the contribution











of the course to the extent that a student's belief system

may be open to change.

Influence by age, sex,; social class and university

grade point average on the abo e matters is also explored.

i

Significance ofi the Study


The results of this study should provide empirical

evidence regarding the extent to which an undergraduate

social foundations of education course accomplishes a portion

of its purposes. This study examines the influence of actual

course operation on enrolled students. It is hoped that

such evidence will provide a basis for evaluating social

foundations in addition to pas studies which have relied

on text materials, course descriptions, historical data or

normative essay.

This evidence should prove helpful in judging claims

made by various critics regarding the desirability of such

courses.

Finally, this study should be helpful to those who

are responsible for teaching and improving social foundations

of education courses.



Hypotheses


1. Neither group membership nor time of testing are

significant predictors of scores on the Scale of Democratic

Attitudes.











2. Neither group membership nor time of testing

are significant predictors of scores on the subscale of

Procedural Agreements of Democracy.

3. Neither group membership nor time of testing

are significant predictors of scores on the subscale of

General Statements on Freedom of Speech.

4. Neither group membership nor time of testing

are significant predictors of scores:on the subscale of

Specific Application of Freedom of Speech and Procedural

Rights.


5. Neither group membership

are significant predictors of scores

Political Equality.

6. Neither group membership

are significant predictors of scores

Social and Ethnic Equality.

7. Neither group membership

are significant predictors of scores

Economic Equality.

8. Neither group membership

are significant predictors of scores

Scale.

9. Neither group membership

are significant predictors of scores

Inventory.


nor time of testing

on the subscale of



nor time of testing

on the subscale of



nor time of testing

on the subscale of



nor time of testing

on the Dogmatism



nor time of testing

on the Personal Beliefs











10. Neither group membership nor time of testing

are significant predictors of scores on the Teacher Practices

Inventory.

11. There is no significant difference in the

change in the correlation of scores on the subscale of

General Statements on Freedom of Speech and the subscale

of Specific Application of Freedom of Speech and Procedural

Rights between the control group and the experimental group.

12. There is no significant difference in the

change in the correlation of the ranking of Freedom and

the ranking of Equality on the, Value Ranking Test between

the control group and the experimental group.

13. Hypotheses one through ten will be tested

to determine the effect of age, sex, grade point average

and social class.



Limitations


Data for this study were gathered during the fourth

academic quarter of 1974 at the University of South Florida

in Tampa and St. Petersburg, Florida. The study was limited

to undergraduate sections of social foundations of educa-

tion. Generalizations resulting from the study will be

limited by these factors.











Definition of Terms


1. Social Foundations, of Education Course: An

upper level undergraduate course for four quarter hours

credit regularly offered at the University of South

Florida. This course is defined by the university as the

study of the "social, economic, and political context

within which schools function and the values which pro-

vide direction for our schools; the culture as a motivat-

ing influence in instruction" (University of South Florida,

1973, p. 151). Further, the word course is defined to in-

clude the effect of a particular instructor and curriculum

materials.

2. Attitude: "a relatively enduring organization

of interrelated beliefs that describe, evaluate, and ad-

vocate action with respect to an object or situation, with

each belief having cognitive, affective, and behavioral

components" (Rokeach, 1968, p. 132).

3. Belief: "a predisposition that, when suitably

activated, results in some preferential response toward

the attitude object or situation, or toward others who

take a position with respect to the attitude object or

situation, or toward the maintenance or preservation of

the attitude itself" (Rokeach, 1968, p. 132).

4. Value: "a type of belief, centrally located

within one's total belief system, about how one ought or











ought not to behave, or about some end-state of existence

worth or not worth attaining. Valuesare thus abstract

ideals, positive or negative, not tied to any specific

attitude object or situation, representing a person's be-

liefs about ideal modes of conduct and ideal terminal goals"

(Rokeach, 1968, p. 124).

5. Democratic Attitudes: Defined by agreement or

disagreement to specific statements on a Likert-type atti-

tude scale (Oppenheim, 1966, pp. 133-42).

6. Procedural Agreements of Democracy: A subscale

of a Likert-type scale on democratic attitudes.

7. General Statements on Freedom of Speech: A sub-

scale of a Likert-type scale on democratic attitudes.

8. Specific Application of Free Speech and Proce-

dural Rights: A subscale of a Likert-type scale on demo-

cratic attitudes.

9. Political Equality: A subscale of a Likert-

type scale on democratic attitudes.

10. Social and Ethnic Equality: A subscale of a

Likert-type scale on democratic attitudes.

11. Economic Equality: A subscale of a Likert-type

scale on democratic attitudes.

12. Experimental Beliefs: Defined by Bob Burton

Brown's Personal Beliefs Inventory (1968, p. 82).





9




13. Experimentalist Educational Practices: Defined

by Bob Burton Brown's Teacher Practices Inventory (1968,

p. 87).

14. Openness to Change of Belief System: Defined

by Milton Rokeach's Dogmatism Scale (1960).

















CHAPTER II

REVIEW OF THE LITERATURE



During the past decade there has been considerable

discussion concerning the nature and functioning of the

field of study known as social foundations of education.

Indeed, this might be expected in a field which has so

recently developed. Historical development of the field

has been described by Quillen (1965), Martin (1968), and

Clarke (1968). This discussion has not only included pro-

fessionals working in the field, but prestigious outsiders

such as James Bryant Conant (1963) as well. Although R.

Freeman Butts has recently judged the discussion to be "a

heartening sign of vitality and pertinence" (1973, p. 27),

much of it has addressed alleged fatal weaknesses.



The Critics


Some of the harshest criticism of foundations

courses in general can be found in Conant's impressionistic

survey of teacher preparation institutions done in the

early 1960's. Conant criticizes not only the courses,

but the professors as well:










Those in charge of the foundations courses
often attempt to patch together scraps of
history, philosophy, political theory, soci-
ology, and pedagogical ideology. The pro-
fessors are frequently not well trained in
any one of the parent disciplines; certainly
very few have such mastery of all the disci-
plines as to be able to talk about them except
at the most superficial level . . In
general, however, I would advise the elimina-
tion of such eclectic courses, for not only
are they usually worthless, but they give
education departments a bad name. I have
rarely talked with students or school teachers
who had good words to say for an eclectic
foundations course. Perhaps the kindest word
used to describe most of these courses was
"pathetic." (p. 127)
i
Conant's sweeping condemnation urges the elimination of

these worthless and pathetic eclectic scraps which are

taught by superficial and inadequately trained people whose

students dislike such courses.

In appreciating the validity of Dr. Conant's opin-

ions, consideration should be given to the procedures used

in forming those opinions asi well as the author's background

in the traditional academic discipline of chemistry. The

point to be raised here and following is to the method and

criteria employed in making judgments. While there may be

foundations courses and professors who are guilty as charged,

it does not seem that the claims advanced by Conant are

amply proven. He explicitly admitted that his study "relied

heavily on the opinions of classroom teachers" (p. vii).

There is little attempt to assess the possible biases of

those opinions nor is there evidence of careful empirical

study to assess the nature of these courses and their










professors. Additionally, it is possible that Conant's

traditional academic background in chemistry may have

made him susceptible to a dislike or, at least, a lack of

understanding of courses which attempt to be interdisci-

plinary and have goals which are professional as well as

scholarly.

Conant's opinions regarding foundations courses were

developed and widely circulated in the early 1960's. They

were parrotted in the early '70's by the journalist Charles

Silberman in his widely read,book, Crisis in the Classroom

(1970). These hastily drawnconclusions were probably

detrimental to desirable teacher education programs inas-

much as they contributed to pressure for reduction or

elimination of foundations study. Without serious study of

well-developed foundations courses, teacher education may

come dangerously close to mechanical training resembling

a trade school approach.

In 1964 Charles J. Brauner, a professional educator,

concluded that the foundations of education constituted a

moralistic doctrine. Brauner's analysis focused on the

early foundations programs at Columbia University and the

University of Illinois with much attention given to the

work of Harold Rugg whose influential book The Foundations

of American Education was published in 1947. Rugg's book

describes the development of the early foundations programs

and provides the bulk of Brauner's evidence. Brauner notes:










From Dewey through Brameld, the
tendency has been to build an emphasis on
subjective belief into an instrument for
achieving specific personal commitment
among practitioners . . By adding the
creative expression of the men of force to
the construction of a theory of American
society as a whole, viewed in the frame-
work of the struggle for power, Rugg added
two ingredients: (1) emotional content and
(2) such broad speculation that careful
scholarship had to be forsaken for conver-
sational talk which took objective informa-
tion for granted. In addition, he directed
attention to the specific content of the
welfare state and the principles of democ-
racy. The dogmatic nature of the belief to
be achieved can be judged from the following
assertion by Rugg: i'. democracy is the
only proper government for family, school,
community, nation, or world; the people to-
gether distill judgment and decisions out
of their collective experience." (pp. 218-
219)

Brauner further explains in a tone that he frequently in-

vokes:

By emotional and intellectual means,
the foundations of education would cultivate
belief in a particular social doctrine which
prescribed how a man should relate to his
fellow man. Gentle and general though the
prescription might be, restrained and re-
stricted as the emotional and intellectual
persuasion might be, the foundations-of-
education approach to teacher training had
become oratory for a moralistic doctrine.
Social reconstructionism became its most
virulent form. Journalism emerged as "edu-
cational theory."

The academic world tends to pride
itself upon maximizing clear, explicit, in-
tellectual understanding, all the while
analyzing overgeneralizations and dissecting
hollow abstractions.i By celebrating the
poetic and ritualistic side of general
ideals, the foundations programs, however,
tended to maximize the emotional and unschol-
arly side of instruction. Such direct pursuit





14





of commitment and belief made the founda-
tional courses resemble a minister's sermon
rather than a professor's lectures (p. 220)
. .. .In sum, the foundation movement
strove to establish education as a literary
discipline operatingibetween the exacting
scholarly and logical demands of philosophy
and the careful, detailed attention to lit-
eral description prerequisite to empirical
science. (p. 223)

Brauner's obvious complaint then was that the founda-

tions of education, particularly social foundations, attempted

to create unanimity of belief and in doing so departed from

the standard of dispassionate objectivity. However valid

these claims may be of those whom Brauner has chosen to

analyze, it would seem equally obvious that his work was

narrowly focused. He chose to analyze particular text

materials written by particular individuals at a particular

time. It would appear that it remains to be seen that such

courses were actually taught in a moralistic fashion. Re-

liance on particular text materials as evidence would not

be sufficient to determine whether ministers' sermons or

professors' lectures were actually occurring in social

foundations classrooms. The point is similar to the one

made above. One may well wonder whether Brauner's conclu-

sions exceed the evidence and method used in his study.

George L. Newsome (1971), another harsh critic of

social foundations, takes Brauner to task:

Brauner's charge that social founda-
tions of education is a moralistic doctrine
rests largely upon his analysis of two works,
Harold Rugg's Foundations for American Edu-
cation and Theoretical Foundations of











Education authored by the staff in social
foundations at the University of Illinois.
In each case, it seems to be more the con-
ception of social foundations than it is
the content of any courses that is con-
sidered. Brauner made no attempt, apart
from the works mentioned, to determine
whether the materials and instructional
procedures used in actual courses were
moralistic or doctrinaire. Granting the
importance of conception of a subject in
the shaping of the subject and the teach-
ing of it, it must be remembered that Brauner
considered primarilylone conception and two
rationales for it. The conception studied
was, perhaps, the most common one, but it
is not the only one.1 (p. 23)

In contrast to Brauner's singular indictment of

social foundations courses as moralistic, Dr. Newsome finds

that the term has many and "confused" uses. He lists

seven:

[1] . Divisions of departments offering
a wide array of courses in history of
educational, philosophy of education,
educational psychology, and the like
often go by the name Social Foundations
of Education.

[2] . Introductory courses in education
are sometimes called "foundations."

[3] . Interdisciplinary courses . .
are often called "social foundations."

[4] . Some courses that deal exclusively
with sociology or anthropology in rela-
tion to education are called "social
foundations."

[5] . Titles for some book and occasion-
ally course titles seem to indicate that
there can be "foundations" . of many
special fields in education .. curricu-
lum, administration, urban education, and
other subjects.











[6] The word "foundation" is often found in
titles of books of readings in education.

[7] . Finally, there are books, and no
doubt courses as well, that are plainly
partisan, ideological and prescriptive.
(pp. 23-24)

On the basis of these observations Newsome states that:

With the exception of special courses in his-
tory of education, philosophy of education,
educational psychology, and the like, most
social foundations courses tend to be a hodge
podge of unrelated or poorly related materials
taken from other disciplines. (p. 24.)

He further concludes that:

The nebulous characteristics of social founda-
tions, their frequently moralistic doctrinaire
aspects, their general impotency in teacher
training, their uselessness as introductions
to more specialized courses and their failure
as advanced interdisciplinary studies, leaves
no place for them in teacher education. (p.
26)

So then Newsome recommends the elimination of social

foundations primarily on the basis of varied course descrip-

tions, textbook titles and arrangements of administrative

convenience. It would seem that his conclusions are sus-

ceptible to the same kind of critique that he has made of

Brauner's. That is, Newsomelhas made little, if any, at-

tempt to determine whether the materials and instructional

procedures used in actual courses were confused, hodge podge

or moralistic. It would appear that these conclusions are

not warranted by the evidence of varied "uses." Perhaps

another writer could make the same sevenobservations and

conclude that they represented a variety, and even a










desirable variety, which one could find in any field. Again,

the point is that careful empirical study of actual course

functioning and impact may help us avoid hasty or erroneous

conclusions which are ultimately harmful to teacher educa-

tion.

In 1968 James J. Shields, Jr., a much milder critic,

argued that social foundations should become more relevant,

integrative and analytic (p. 77). By relevant he meant that

such courses should pay more attention to the phenomenon of

metropolitanism and urban education. By integrative and

analytic he meant they should make greater use of the tools

and research findings of contemporary empirical social

science as well as the humanities. According to Shields

these things were not only needed, but were actually occur-

ring:

This trend, which to many represents a "coming
of age" for foundations departments, has been
greatly accelerated by the vigorous attention
social scientists ha e been giving to the
school as an institution and education as a
social process. Along with their professors,
graduate students in all disciplines have been
turning to education for their research stud-
ies. As these students complete their studies,
many of them accept positions in graduate
schools of education and join the ranks of the
"young turks" who are transforming departments
of educational foundations into departments of
philosophy and social sciences. . As a
consequence of this development, and as more
and more of those teaching foundations ask to
be identified with altraditional discipline
rather than education, the entire definition
of educational foundations will change.
(pp. 81-82)











Shields sees no fatal weaknesses nor does he recommend the

elimination of foundations courses. He sees improvement

occurring through the recruitment of specialists outside

the field of education whose primary identification will

be with their respective disciplines.

C.A. Martin's historical study shows this occurring

since World War II (1968). Martin found that:

After World War II a rather substantial shift
occurred in the nature of the social founda-
tions component of the teacher education pro-
grams considered. LAbelled in this study the
Stage of Precision, the 1946-63 time period
witnessed younger instructors with stronger
backgrounds in one of the social sciences or
philosophy making great alterations in the
social foundations programs which developed
during the Stage of Romance. Rejecting the
moralizing approach of the 1930-45 time pe-
riod, the new instructors seemed less con-
vinced that all students needed to take
narrowly prescribed work in the social foun-
dations. Thus they relaxed the requirements
relating to this component of the teacher
education program. ,or did they view the
interdisciplinary, problem-centered courses
as favorably as had heir earlier counter-
parts. Indeed, theylshowed open preference
for the separate subject social foundations
courses such as the history of education and
educational sociology. Such courses, which
drew more heavily upon the content and re-
search methods of a single academic disci-
pline, proved more suited to the subject-
matter oriented social foundations program
which developed after 1945. (pp. 279-280)

Shields argues the necessity of foundations to be-

come more integrative, but acknowledges that the recruit-

ment of academic specialists outside education will create

a fragmentation of study. He views this as a necessary and










inevitable stage in the evolution of foundations. He sees

another stage beyond this:

Another stage must follow in which in-
tensive effort is devoted to facilitating com-
munication among scholars studying education
in all disciplines. iThe first step in achiev-
ing this, of necessity, has to be the creation
of courses and programs that will produce in-
dividuals who are adept enough in the entire
range of disciplines represented in the humani-
ties and the social sciences that they can
integrate the research findings related to edu-
cation into a coherent and systematic body of
knowledge. These individuals would approach
the disciplines in terms of the scholarly needs
of education and not vice versa as is customary
among those in the traditional academic areas.
(p. 84).

Shields' position is that education is not a respect-

able scholarly field nor can it become so through its own

since it must hire outside specialists. These specialists

will fragment the study of education, and they must somehow

be brought together to produce an integrated body of knowl-

edge and discipline of education. Many believe this is

what social foundations attempts to do in the first place.

Shields' evolutionary notionileaves many questions unan-

swered such as who will do the integrating and how. This

may be a major problem considering the acknowledged custom-

ary allegiance of the specialists.

In 1969 Wayne J. Urban considered the Shields

position and found it to be 'naive in the reliance on the

traditional disciplines" (p. 204). Urban analyzes two tra-

ditional social science disciplines, sociology and politi-

cal science, and finds disagreement between the more











traditional specialists and a new breed of generalists

(pp. 201-204). While Urban thinks that while both Conant

and Shields have been "very attuned to the weaknesses and

difficulties in the study of social foundations of educa-

tion," his inclination is:

to search in the directions of a more gen-
eral type of social theory which might
enable one to make intelligent use of the
myriad of educational research and which
also might enable those in social founda-
tions to identify some key problems and
ideas that need to be researched. (p. 204)

Then Urban's position argues the necessity for a

generalist or broad approach to understand and use the

product of specialists, particularlyithe existing product

of educational researchers.

Shields' work (1968, 1969) might be classified as

normative essay based heavily on a review of social founda-

tions textbooks. This approach has merit in assessing the

quality of textbooks, but it has limitations in assessing

the relevance, integration and analysis present in a field

of study. A valuable supplement would be to examine the

nature of professional journals as well as student text-

books. Also helpful would be an examination of all course

materials and actual course operation. Shields has not

done this.

A common thread that seems to run through the ar-

guments of Conant, Brauner, Newsome and Shields is the

study of education has been too broad, ill-defined and











prescriptive; yet it can be improved to become more specific

and analytic with reliance on the tools, knowledge and

orientations of those specialists in the traditional dis-

ciplines. Others besides Urban have argued against this.

In 1964 John Walton maintained the necessity of establish-

ing a separate discipline of education for three reasons:

(1) there are important aspects of education such as teach-

ing and the curriculum whichihave no place in other disci-

plines; (2) other disciplines do not have education as

their central interest; and (3) students of education should

belong to a community of discourse where education is the

major focus (p. 265). Walton saw no improvement through

reliance on outside specialists. In 1961 Elmer Eason

found that educational sociology, as an example of academic

specialty, would be too narrow and restrictive in studying

the social aspects of education. He argued that one:

should not be restricted to the scientific
dimensions of description, prediction and
explanation. Such instruction should also
include the philosophical dimensions of
justification and meaning. Further, in
its scientific dimensions, such instruc-
tion should be alert to relevant resources
existent in all of the social sciences-
not only in sociology of education. In
addition, instruction in the social aspects
of education at the undergraduate and be-
ginning graduate levels should not be
fragmented so that empirical questions are
studied in one course and normative ques-
tions are studied in!another. (p. 142)

The study which did all these things according to Eason was

social foundations of education.










The Nature of Social Foundations of Education


The critical judgments of various writers, particu-

larly Conant and Brauner, appear to have generated much

discussion of the field. While some of the discussion

has been defensive, much of it has been devoted to clarify-

ing the nature of social foundations, its necessity and

improvement.

In 1965 Quillen attempted to distinguish between

social foundations and other similar studies. Quillen

noted that:

"Social foundations" has a broader connota-
tion than "educational sociology." It is
concerned with the contribution of all
social sciences to an understanding of the
interaction of educational institutions and
the larger society. I .. The content of
the social foundations of education draws
from anthropology, economics, political
science, and social psychology. It also
includes content and methods from sociology,
cultural and social history, and social
philosophy. (pp.74, 76)

Therefore, social foundations was considered to be inter-

disciplinary in content and method with scope broadly

focused on the institution of education. While Quillen

maintained that there was little doubt of the value and

growing importance of social foundations for teacher edu-

cation, there was need for a clear definition of its

nature (p. 84).

Also in 1965, Richard L. Derr attempted to define

social foundations of education with an emphasis on its











potential contribution to educational decision making. In

Derr's opinion social foundations referred to the social

factors relevant to schooling and the courses, programs and

literature of education which are brought together to re-

view the social aspects of schooling. Derr states that:

As things now stand, "social foundations
of education" seems to denote two related
things. First, it refers to social fac-
tors which inquiries in the social sci-
ences or in education presumably have
shown to be relevant to the school's
operation.. .. Second, social founda-
tions refers to the courses, programs,
literature in education where the results
of these inquiries usually are brought
together to permit an extensive review of
social aspects of schooling. (pp. 154-55)

In addition to Derr's references to programs of instruction

and the social aspects of schooling, there is emphasis on

the practical application of social foundations in policy

making. He states that, "There is then a need for a branch

of study in education which specializes in identifying,

integrating, and interpreting that knowledge in the social

sciences which pertains to the social dimension of educa-

tional decision-making" (p. 156). This could be best done,

according to Derr, by making social foundations into an

acknowledged discipline with its own corps of specialists.

These specialists would then inform educational decision

makers of relevant social conditions so that policy making

would be more effective. Derr sees this policy orientation

as in keeping with current trends and practices.











In 1968 William O. Stanley developed two rationales

for social foundations of education which included philo-

sophy and history of education. Stanley argued that as a

field of inquiry education can be viewed as a liberal arts

discipline or as a professional study (p. 227). When

viewed as the former, the educational institutions of

society constitute a definite and unique field of study.

In this context Stanley felt that social foundations needed

no defense (p. 228). However, the latter context proved

troublesome. Stanley felt that opposition to the use of

social foundations in the preparation of teachers was due

to a "craft mentality"which views the occupation of public

school teacher as nothing more than a competent classroom

operator (pp. 227-29). Similarly, in 1969 Howell and

Shimahara argued that the vocational bias of a significant

number of education professors resulted in opposition to

foundations study. Stanley judges this conception as

unacceptable for several reasons: (1) it assumes that

all of the basic problems of education are solved or can

be solved without the participation of the classroom

teacher; (2) if we view the teacher as a technician, we

cannot recruit the best brains for the job; (3) the notion

creates and perpetuates a vicious segmentation in the ranks;

and (4) if one set of people determines policies and goals

while another set does the classroom teaching, then the

goals and policies will be skewed and altered in their

application (pp. 229-30).










Hence Stanley saw a major role for social founda-

tions in the professional preparation of classroom teachers.

Foundations study would provide teachers with interpretive

knowledge. Interpretive knowledge is the knowledgeof the

generalist which locates problems within a set of meanings

and provides understanding of the problem (p. 235). Then

interpretive knowledge would rationalize practice and equip

classroom teachers with the ability to participate intelli-

gently in the determination of educational policy. Stanley

concludes:

If the foregoing argument is at all
sound it is clear that without technology
you do not have an effective professional
practitioner, but without theory you have
only a technician. For me the conclusion
is inescapable that we will not have a pro-
fession of teaching nor a corps of teachers
adequate to the demands of education in our
time until the teacher-preparation program
encompasses a good general education, a
thorough grounding in the subject matter
of the courses to be taught, a substantial
background of interpretive knowledge about
the education enterprise as a whole suffi-
cient to produce nascent educational states-
men, and the theory) technology, and prac-
tice in the skills needed by the particular
specialty of the prospective educational
worker. (p. 235)

In 1972 Mary Ann Raywid attempted to bring some

order to the various discussions about the nature of social

foundations. She located seven features which described

the field.

[1] Policy Oriented Study. The clearest
and most prominent feature of the
field is its policy focus. . .










[2] Unifying-Integrative. In order to ac-
complish its policy-decision function
social foundations must draw on dispar-
ate fields. It is interdisciplinary.


[3] Interpretive. The field is interpre-
tive not only to distinguish between
fact and value, but to distinguish
between differing descriptive claims
involved in policy determination. . .

[4] Methodological Sophistication Required.
In order to integrate the findings of
the various disciplines methodological
sophistication is necessary. . .

[5] Contextual Preoccupation. Examination
of the social context of education is
necessary for policy formulation. . .

[6] Ideas and Beliefs are of Concern.
Social foundations assumes that ideas
and beliefs have significant influence
in the determination of educational
policy. . .

[7] Contemporaneous Orientation. Social
foundations is preoccupied with present
circumstances, events and conditions
in order to advise on educational pol-
icy. (pp. 75-77)

To Raywid the major justification for social founda-

tions of education is its role in the professional prepara-

tion of teachers. She argues that social foundations:

Represents a teaching field, not an academic
discipline. It is put together for instruc-
tional not for research purposes, and there
is no reason why it need try to meet or an-
swer to the criteria marking and measuring
academic disciplines, which are primary
fields of inquiry. The particular "package"
constituting Social Foundations is assembled
on the grounds that it unites disparate
material bearing on educational issues,
using the issues themselves as the selecting
and organizing principle. (p. 78)
[cpe p 8










Others such as Stanley and Derr would obviously agree that

social foundations has a major responsibility in the prepa-

ration of teachers. They would also agree that this could

be legitimately accomplished through policy analysis. How-

ever, Raywid's claim that social foundations' policy focus

and interdisciplinary approach preclude it from the status

of an academic discipline may have some difficulty when one

considers the policy focus and integrative approaches urged

in some of the recognized disciplines such as economics and

political science. Walton (1964), Stanley (1968) and, par-

ticularly Laska (1969), have argued the foundations studies

have standing as academic disciplines. For Laska this is

their primary reason for being, while Stanley sees a dual

role of professional and academic responsibility.

Despite these differences there seems to be a broad

area of agreement that social foundations has an important

professional role to play. iRaywid's seven characteristics

taken together would seem to be agreeable to Stanley's no-

tion of interpretive knowledge. Harry Broudy (1972) has

recently made similar references to foundations studies

providing what he calls interpretive theory as opposed to

applicational theory:

The point, therefore, is whether
these studies are useful in some sense
other than as a source of rules to be ap-
plied to practice. !This is an issue in
all general studies land in the role of
foundational studies in any professional
curriculum. The most plausible defense
of them is to point out that they provide











the context of practice rather than the
rules for practice. Thus, an understand-
ing of the sociology of poverty does not
directly give rules for healing the dis-
eases of the poor, but the dietary pre-
scriptions that a physician might give to
the poor will be more enlightened if he
does understand the sociology of their
condition. Knowledge of the social con-
text, therefore, affects the general
strategy of education, of appraising the
teaching situation in many dimensions, and
for making decisions that take account of
these dimensions. (pp. 56-57)

It would seem reasonable to judge social foundations

of education according to the role it seeks to play. To

many it has a major role in providing at least contextual-

interpretive knowledge to those who would be professional

educators.



The Problem of Assessment


If one is willing to grant that social foundations

of education has the potential to make contributions either

as a discipline or as professional education, one might

wonder about progress. Can an assessment of the field be

made? A review of the literature produces a few very nega-

tive opinions. Much of what has been written is essentially

supportive. There has been a great deal of discussion of

the purpose and nature of the field. Recommendations for

reforms have been advanced. However, almost all of the

writing has been in the form of normative essay. And, as

the first part of this chapter sought to demonstrate,










conclusions reached about social foundations as a field

of study may have no firm base of evidence. There is a

distinct lack of hard empirical data to indicate what is

really occurring in the field, at least as it attempts

its professional education function. On this alone, it

may be very difficult to arrive at a defensible assessment

of the field.

Some have been fond of the notion that much of

past educational research has been disparate, unorganized

and without focus. This notion is probably true. However,

opinions and conclusions about social foundations of educa-

tion as a field of study might benefit from the caution

and firm base of empirical evidence that the educational

researcher expects as he engages in the process of evalua-

tion. Empirically based research studies may be helpful in

providing sounder conclusions about how we are doing.

In addition to this evidence problem, some criticism

of social foundations may be guilty of mixing evaluational

criteria. Part of the problem appears to be that many

critics have been legitimately concerned with improving the

field's standing as a discipline. Some of these critics

have applied criteria appropriate to traditional academic

disciplines to the professional education function of social

foundations. In its professional education function, social

foundations is interdisciplinary in order to reach a closer

approximation of reality. To some this interdisciplinary





30




aspect violates traditional academic specialization. Fur-

thermore, social foundations as professional education has

developed an extensive policy study emphasis. In doing

this it must deal with educational prescriptions. This

policy aspect would appear to run counter to the notion of

dispassionate objectivity. Other sources of complaint have

been school administrators, classroom teachers, student

teachers, and some professors of education who have argued

that social foundations is too theoretical, too idealistic

or unrealistic. This group appears to expect specific

classroom recipes which will insure survival or success

in the day-to-day life of the school. This expectation

in itself is unrealistic considering what is or is not

known with certainty about teaching and learning. If

social foundations attempts to become more discipline-like

and attempts to develop more fundamental,basic and special-

ized knowledge about the social aspects of education, then

it may appear to be irrelevant, in an immediate sense, to

school people. The point to be made is that evaluational

criteria ought to be appropriate to the function being

performed by social foundations. A similar point has been

made recently by Robert R. Sherman (1974) who has argued

that most of the criticism and analysis of social founda-

tions has been of little value because it has failed to

evaluate the field in reference to its purpose. Evaluation

of social foundations should be done in reference to its











functions. To evaluate social foundations in its profes-

sional role by invoking traditional academic standards

which have condemned advocacy and the interdisciplinary

approach would seem unfair. To apply professional con-

siderations in judging social foundations when it attempts

its academic function would be similarly unfair.



What Should Social Foundations Study Accomplish?


In light of the above discussion, it would seem

that social foundations has at least two major goals to

accomplish-one professional, the other academic. To many

writers social foundations has a role in providing contex-

tual-interpretive knowledge ito those who would be profes-

sional educators. However, the professional should not

expect specific and definite rules for practice since our

understanding of the educational process in general, as well

as its social aspects, is limited. Inquiry in the social

aspects of education should enhance one's understanding of

the context of practice and suggest ways of improving

practice. In addition to this, the policy orientation of

social foundations is thought to have benefit for school

people. Policy emphasis should lead to a clearer under-

standing of the social aspects of competing claims made

about the proper functioning of schools.

Social foundations is thought to have a role as an

academic discipline. Inasmuch as the institution of










education is a major social institution which deserves to

be regarded as a legitimate field of study, so should the

study of its social aspects be regarded. It is conceivable

that the development of such a discipline will probably

lead to a practical payoff for classroom teachers. However,

it is also conceivable that the study of the social aspects

of education could be a legitimate object of intellectual

curiosity in its own right without yielding an immediate

benefit to classroom teachers. In this sense the disciplines

of economics or political science have been pursued.

Sherman (1974) has defined social foundations as a

field which should provide fundamental knowledge and, most

importantly, vision to professional educators:

The idea of social foundations was-and is-
that education and the culture interact and
that it is essential for teachers to base
their study and practice on fundamental
knowledge relevant for directing the cul-
ture, through education, toward preferred
ends. . As well as drawing together
the resources for describing and analyzing
the culture and the role of education, the
study of social foundations should help to
generate the vision which gives point to
what we do and might: do with education.
. Specifically, I believe that in the
social foundations course itself we need to
be as concerned with questions about the
possible and intended uses of education-
in the sense of those to be chosen deliber-
ately-as we are with describing the cul-
tural setting. (pp. 5-6)

For Sherman that vision will be found in the liberal

democratic tradition. He concludes:











We could, through the social foundations,
revitalize the liberal democratic tradi-
tion and base our education on and judge
it as it contributes to that tradition
through experience, the interdisciplinary
approach, and the test of results in ac-
tion. This is still the work of the social
foundations in the 70's. (p. 11)



Conclusion


There is a need for more effective assessment of

social foundations not only to avoid erroneous and harmful

conclusions, but to improve the possible contributions of

the field. Furthermore, this assessment should be done in

reference to the functions or purposes of the field using

the tools and methods of modern empirical social science

as well as philosophy and history. More specifically, this

study attempts to explore through the use of experimental

design and statistical technique the actual operation and

impact of social foundations on the valuational-attitudinal

structure of its students. Such a study should provide a

firm base of evidence through which social foundations can

be assessed and improved.

















CHAPTER III

METHOD



Design


This study is designated a quasi-experimental pre-

test-posttest control group design. The experimental group

consisted of students enrolled in all sections of the under-

graduate social foundations of education course. The con-

trol group consisted of stud nts enrolled in all sections

of the undergraduate human development and learning course

during the same academic term. Concurrent registration for

this course and the undergraduate social foundations of

education course is discouraged as a matter of university

policy. Students normally complete human development and

learning before taking social foundations. Students pre-

viously or concurrently exposed to social foundations were

eliminated from the control group.

Pretests were administered during the first week of

the academic term to both experimental and control groups.

Posttests were administered during the last week of the

academic term. In order to minimize time consumed for test-

ing, each group was randomly divided in half with one half











being pretested with the Scale of Democratic Attitudes and

the Personal Opinion Questionnaire while the other half was

pretested with the Personal Beliefs Inventory, Teacher

Practices Inventory and Value Ranking Test. On posttesting

the administration was reversed.

This study is designated quasi-experimental after

terminology popularized by Campbell and Stanley (1963)

The design does not qualify as a true experiment because

subjects are not randomly assigned to experimental and con-

trol groups nor is the treatment randomly administered.

Therefore, the design is termed quasi-experimental.

Randomization is done to promote the equivalency of

groups to eliminate threats to validity. The degree of con-

trol necessary to accomplish this sort of randomization is

not possible in this study because of university policy.

However, practical equivalence of the experimental and con-

trol groups can be defined by noting a number of similari-

ties. Both groups are composed of students enrolled in

upper level (300) undergraduate courses worth four quarter

hours required by the university of education majors and

those desiring certification as public school teachers.

Both courses are normally taken within one academic term

of each other. Both courses normally enroll students major-

ing in a variety of educational specialties. It would ap-

pear that the internal validity of the study would not be

threatened and that the intervening effects of history,











maturation, testing, instrumentation, regression, selection,

mortality, and selection-maturation interaction would not

be present unless substantial reason could be given to

suspect otherwise (Campbell and Stanley, 1963, pp. 5-6).

The intent of the study was to assess the impact of

the undergraduate social foundations of education course as

it was normally taught to a majority of students enrolled

during a particular term. Variation in instructors and

curriculum materials were considered to be expected, accept-

able, and an inseparable aspect of course offerings. There-

fore, rival hypotheses supported by these variables were

considered irrelevant given the intent of the study. Fur-

ther, the dependent variables used in the study could be

applied appropriately in all various sections of the course.

The external validity of the study could be quali-

fied. The social foundations of education course under

study was taught at one institution and the findings of this

study would not necessarily be applicable to courses taught

at other institutions. Therefore, it is conceivable that

the results obtained from the experimentally accessible

population may not always hold true for the general popula-

tion (Bracht and Glass, 1968, p. 438). However, the course

was not unusual in its conception and was very similar to

social foundations of education courses taught elsewhere.

To the extent that this can be illustrated by comparing

course descriptions of social foundations of education,











the following examples should be noted:

[1] [The study of the] social, economic and
political context within which schools
function and the values which provide
direction for our schools; the culture
as a motivating influence in instruc-
tion. (University of South Florida,
1973)

[2] A study of the educative effects of
our social structures, the social
values and issues involved in apprais-
ing these effects, and the resulting
social demands upon the schools.
(University of Florida, 1972)

[3] A study of the historical and contem-
porary relations of education and
society, and of schools and colleges
as social systems, from the perspec-
tives of the social sciences. (Univer-
sity of California, Berkeley, 1972)

Strictly speaking, the above factors should be con-

sidered in interpreting the results of the study. The

steps necessary to control for possible interference would

be arduous and the practical outcome might be inconsequential.

It is thought that the need and benefit from a study of

this design outweighs its potential and remote qualifica-

tions.



Collection of Data


Data were collected by the author during the summer

quarter between June 17, 1974 and August 23, 1974 at the

Tampa and St. Petersburg campuses of the University of

South Florida.










Pretesting in the experimental group involved 151

students in five different sections of undergraduate social

foundations of education. Posttesting involved 140 students

in the same number of sections. Three students in the ex-

perimental group refused to cooperate. Eleven students

were not available for postt sting, due to their failure

to complete the course or class absence.

Pretesting in the control group involved 84 students

in six different sections ofi human development and learning.

Posttesting involved 54 students in the same number of sec-

tions. One student in the control group refused to cooper-

ate. The difference of 30 students between protesting and

posttesting was due to failure to complete the course and

class absence. It is suspected that class absence was the

primary factor because two sections of the control group

course had a so-called self-pacing feature which did not

require or encourage regular class attendance. Students

absent from the posttesting session could have contacted

individually for testing, but the test environment would

have differed substantially from the usual setting. The

decision was made to maintain the same test situation for

all subjects rather than vary it for the sake of increasing

control group size.

Experimental and control groups of similar size were

anticipated. The smaller control group was due to the fact

that many students had concurrent enrollment both in social





39




foundations of education and human development and learning.

These students were screened from the control group. Con-

current enrollment is discouraged by college policy (Uni-

versity of South Florida, 19,73), but apparently was not

implemented during this academic term.

Arrangements were made with instructors of the var-

ious sections of the courses comprising the experimental

and control groups prior to the academic term in which

testing occurred. Instructors of all available sections

were cooperative.

Protesting was done at the second class meeting by

the author. The Personal Op'inion Questionnaire which con-

tains the Scale of Democratic Attitudes and the Dogmatism

Scale was distributed alternately with the Personal Beliefs

Inventory, Teacher Practices Inventory and Value Ranking

Test. These latter instruments were color coded with a

gold instruction sheet. See: Appendices A, B and C for these

instruments. Alternate and, therefore, random distribution

of these two packets of instruments to every other student

in the classroom allowed a maximum amount of testing in a

single class period.

Instructions were read to the students. Before

allowing the students to complete the IBM 555 answer sheet,

a verbal reminder was given that the activity was to be

done anonymously, results were to be held in confidence

and had no impact on course grades. Students were urged

to participate honestly and sincerely.










Posttesting was .done at the next to last class

meeting by the author. Students who completed one packet

of instruments on protesting were given the other packet

for posttesting. This was dpne by using the last four

digits of the student's social security number and birth-

date supplied on pretest answer sheets. This information

was transferred to the desired test packet and distributed

for posttesting to the appropriate student.

The study investigated the influence of age, sex,

social class and university grade point average. The

answer sheets required students to enter the last four digits

of their social security number, birthday, sex and occupa-

tion of adult head of household. Age to the nearest year

was obtained from birthdate. A measure of social class

(SES) was obtained from occupation of adult head of house-

hold using Duncan's Socioeconomic Index for all Occupations

(Reiss, 1961). Duncan's Index was selected because of its

essentially continuous measures with scores ranging from 0

to 96. The index was appealing in a practical sense in

that it was suspected that subjects would be more willing

and able to give accurate information on occupation rather

than on income or education. The major elements of the

index were reported to be correlated to the extent of .91

with prestige ratings of education and income. It was

further reported that the index had "a high order of temporal

stability" (Blau and Duncan, 1967, pp. 120-21).










Acquiring university grade point average (GPA)

was indirect and more involved. Direct solicitation of

this information from students was not done because of

possible inaccuracy. Class rolls were obtained from the

university registrar. A current listing of university

grade point average for all students was obtained similarly

and was matched with the names on the class rolls. This

information was then entered on the answer sheets by match-

ing the last four digits of the social security number with

the student's complete social security number on the class

roll.

Specific information regarding the characteristics

of the students who participated in the study is found in

Tables 1 through 4. Tables 1 and 2 contain information

about students on whom complete data were available. Mean

values on age, social class, university grade point average

and sex are given. A mean value for sex is possible because

males were dummy coded 1 and females were dummy coded 0.

The number of males and females are also given as is con-

ventional.

Tables 3 and 4 contain information on all students

who participated in the study to include those on whom

incomplete data were available. On the pretest 46 students

or 30.46 percent of the experimental group did not provide

complete data. This compares to 33 students or 39.29 per-

cent in the control group wh failed to provide complete























0 ro CN (n, ,,
4 m N C 00 m
H H cy ( co
4- l m Nq -IT 00 r-
C II * .
o mU N LnN
0 u ( Ln ,--I



0
H
,---I I r

o '.J CJLn(
) HC:
e 0 U n
0 a) Q)
H 4 (d 4J I -i LD ON LD r- LrA
S.0 0 0 M -

S O4 0 (4 i



S 4.) 4 U
ca

C 4



Y 0 r-
m 0 U 0 H
F 1 4 cC I I
(d slH 4)u C4 'T
Q) 4) HU

W rO -
wL 4Pl i rd 44
S01 a) (4J
D W (UM (u o m 01
00 ELn o 0 -HH o 0 (N
A4 -II 1N i-q -V in o
4m ) C m (U
(N as 4 01

ci 14 ci
4) t 0 :


0 ci



(H )



4) t) X: (u w4
U c U) 0 l



u
























f-A
0
4J
C II










aoo













4-1 C
4C
U

a


















II
0
U










-H II

a
o


( I
%.0 |n f-.


Cd







0
4J
4-1 (
--I Q4
-1-1





m
di







o








0 0





cO
Cd
e 0











-4
4J





uo
. >











UC C
1B









10 4-P
4 C
i0dt
u 54
H C


U

*H


(d





U
SE

10l


H
N


CN
Ln


m

0












o


N N

r 0 0 0

mro N
Ln











--l
0d

00) C

C) U


o

*- -






44
















om o w-
S li


04 0 U: N -4
4-J 0 N N O' CO
C: II "
4J Or I n o- M aN
On u NM On H-1
a) i r
4-J '
o-4 CA

o 0 a
r. P4' I
04i (
-0 4 D
OH C *
Q) a)a (n Im )H (U0o o In ( i (D
",d I I O- n 4

4j In ) 4

-4 ,"- E a)

EA r. LF


0) '() U11
H C






U 4 O )




U) 0 (C ) 4-) 4-Il O 0
S4 -i) 4J
o a U
4-) -rH -H M


L a 4r C
-nH 0 N 4
0 0 4J1 4- U (
a) a4) 1. W
4-) DU .0 UM
10 4-' c o o In o 4iJ U
44 0 N N,-CM n in 4 u'

4-' M N NU [U

u a)


U




*U (z 0 U

-H
U U [) [l 0



U
























1 CO CO 0
m o o0o o r-
r- m f) co m

N H in
|C N L


0l 0
c<'w 0n
14LI d


Mo (-N r-
iH 0

ICC


-4

r-l



a)
CQ >i


C 0


I.


aU







0 (E
H Cd
,-4 -





0

) 4-J


m>
a) a)







04-
d C
I--I C
i >






Cd

u


: a)

r-
--IE

a) x rd) m
Ct a ( F W P4
9 ml ml 0


Ho W 0


im m
In


a)
rl











S 4J
,




O E
-,-

O





r f
a)
Cd

4 H














O a,
.l (d











0 4
4J 0










U 0
u a)
0 0








cJ








0d
a m
rl

d (U


Cd 4J

0 H
o s)












00
d(



Cdrt


-U)
0)
4-1


ftA


O

C II

U





(-i

CI
0
0 F











a
Crl
0)








U
4-'
-H


a)
in-



41
u
Cd

.1










information. On the posttest 37 students or 26.43 percent

of the experimental group was incomplete compared to

17 students or 31.48 percent of the control group. Mean

values were substituted for missing data. This procedure

is recognized by Travers (1969, p. 358) as a commonly sug-

gested method for dealing with the problems of missing data.

These values were obtained from students in the same group

on whom complete data were available.

Student responses to the test instruments on the

IBM 555 answer sheets were machine scored by the University

of South Florida's Office of Evaluation Services. These

scores and information on age, social class and university

grade point average were entered on coding sheets. Also

entered after dummy coding were sex, experimental or control

group membership, pretest or posttest and an experimental

pretest interaction variable. Coding sheets were then

given to a professional card punch operator who punched and

verified the data deck.



Instrumentation


The Scale of Democratic Attitudes (SDA)


Items used by Herbert McClosky in 1964 to measure

the degree of consensus on democratic values exhibited by

political influentials and the general electorate were the

basis of the SDA. The items were grouped in the following










ways: items expressing "rules of the game," items express-

ing support for general statements of free speech and opinion,

items expressing support for specific applications of free

speech and procedural rights, items expressing belief in

political equality, social and ethnic equality, and economic

equality. The final form of the SDA was constructed by

taking all items together. See Appendix A for the items

used by McClosky. The six subscales of the SDA correspond

to above groupings using those items supplied by McClosky.

They are, respectively: The subscale of Procedural Agree-

ments of Democracy (PAD), th subscale of General Statements

on Freedom of Speech (GSFS), the subscale of Specific Ap-

plication of Freedom of Speech and Procedural Rights (SAF),

the subscale of Political Eqpality (PE), the subscale of

Social and Ethnic Equality (SEE), and the subscale of Eco-

nomic Equality (EE). See Appendix A for particular items.

The Likert technique was used in scaling with six

levels of response: disagree very much, disagree on the

whole, disagree a little, agree a little, agree on the

whole, agree very much.

Validity of the McClosky items is indicated by the

following:

While each of these items can stand alone
and be regarded in its own right as an
indicator of a person's opinions or atti-
tudes, each of them is simultaneously an
integral element of one of the 47 "scales"
that was expressly fashioned to afford a
more refined and reliable assessment of
the attitude and personality predispositions










of every respondent. Each of the scales
. has been independently validated
either by empirical validation procedures
employing appropriate criterion groups, or
by a modified Guttmah reproducibility pro-
cedure (supplemented, in some instances, by
a "face validity" procedure utilizing item
ratings by experts). (p. 364)

McClosky does not explicitly identify these 47 "scales" ex-

cept to note their earlier use in 1957-58.

McClosky does not discuss reliability of his items

directly. However, analysis using the Coefficient Alpha

which is an internal consistency estimate of reliability

yields the following: SDA = .9145, PAD = .7406, GSFS =

.7974, SAF = .5812, PE = .7030, SEE = .8484, and EE = .7934

(Cronbach, 1951).

The items of the various scales in the final form

of the SDA were randomly distributed and combined with the

Rokeach Dogmatism Scale.


The Value Ranking Test (VRT)


The VRT is based on experimental work done by

Rokeach on dissonance, behavior change and value change

(1969, 1971). Rokeach asked subjects to rank 18 values ac-

cording to their perceived importance. See Appendix B.

After the subjects had completed their rankings, they were

shown average rankings from another group of subjects. The

difference between the ranking of freedom and equality was

noted and discussed with the notion emphasized that those

who ranked freedom and equality far apart were interested

I





49




only in freedom for themselves. This discussion was de-

signed to produce dissonance or self-dissatisfaction with

the subject's ranking. Additionally, subjects were asked

to indicate their sympathy for the civil rights movement.

Then they were shown data indicating a positive relationship

between civil rights sympathy and a close ranking of free-

dom and equality. This also was done to induce dissonance.

Subsequent measurement showed that experimental subjects

produced closer rankings of freedom and equality; and more

favorable attitudes toward civil rights than the control

group. Additionally, experimental subjects were more in-

clined than the control group to respond favorably to

solicitations to join the N.A.A.C.P. up to seventeen months

after the initial dissonance producing sessions. Rokeach

reports test-retest reliabilities for college students from

.65 to .80 (1973).

The technique used by Rokeach to produce dissonance

would seem to have very close similarity to the typical

classroom situation where major cultural values are dis-

cussed, including the relationship between freedom and

equality in a theoretical and problematical context. Con-

sidering Rokeach's findings, one could expect dissonance

in the student and a subsequent reordering of values as

well as behavior change. It is thought that the extent to

which a course is successful in accomplishing these changes

can be assessed by a change in rankings on the VRT.











The Personal Beliefs Inventory (PBI)


The PBI was developed by Brown to determine a sub-

ject's agreement with the general philosophical beliefs of

John Dewey. See Appendix C. Items were based on Dewey's

writings and subjected to review by various judges and item

analysis techniques. After five revisions a final inven-

tory resulted with forms A and B. Brown reports combined

forms reliabilities ranging from .55 to .78 (1968, p. 100).

Similar forms reliability of .58 was reported when forms A

and B were administered separately. Combined forms were

administered in this study in hopes of greater reliability.


The Teacher Practices Inventory (TPI)


The TPI was developed by Brown in the same manner as

the PBI. The TPI is designed to assess agreement with

teacher practices which are compatible with John Dewey's

thinking. See Appendix C. Similar forms reliability of

.69 was reported when forms A and B were administered sep-

arately. Combined forms reliability ranged from .56 to .94

(1968, p. 100). Combined forms were administered in this

study.


The Personal Opinion Questionnaire (POQ)


The POQ includes the Rokeach Dogmatism Scale. See

Appendix A. Rokeach reportsi reliabilities of .68 to .93











(1960, p. 90). The well-known Dogmatism Scale measures

the intensity with which one holds beliefs. Items on the

Dogmatism Scale were randomly distributed with SDA items.



Statistical Methodology


The primary statistical method used in this study

was stepwise multiple regression. This technique involves

the construction of a regression equation that shows which

independent variables are the most powerful predictors of

the dependent variable. Furthermore, the stepwise technique

not only shows the particular variables which are powerful

predictors, but their desired order in the equation since

order affects their predictive power. As each independent

variable is added to the equation, starting with the one

most highly correlated with the dependent variable, F tests

are performed at each subsequent step to show the contribu-

tion of each variable already in the equation if it were to

enter last. Therefore, it is possible to determine the

optimum order of independent variables and to discard those

which were initially, but no longer, good predictors (Ker-

linger and Pedhazur, 1973, p. 290).

Multiple regression procedures are quite often used

to construct nonexperimental predictive systems. However,

multiple regression can be applied to experimental designs

by dummy coding experimental and control group membership










as independent variables. The dummy coded categorical in-

dependent variables in this study were experimental or

control group membership, whether dependent variable

scores were pretest or posttest, whether dependent variable

scores were experimental pretest or otherwise and sex.

Essentially continuous independent variables were age,

social class and university grade point average. Dependent

variables included scores on the Scale of Democratic Atti-

tudes, its subscales, the Dogmatism Scale, the Personal Be-

liefs Inventory and the Teacher Practices Inventory.

Analysis of variance is a technique that is fre-

quently used in an experimental design such as the one em-

ployed in this study. However, multiple regression offers

several advantages. According to Kerlinger and Pedhazur

(1973, p. 114), the technique is superior to analysis of

variance when essentially continuous independent variables

are involved. In analysis of variance such variables are

typically treated as categorical variables when they are

partitioned into high, medium and low groups. In partition-

ing a variable that may have a range of values, one can

lose considerable variance (p. 8). Therefore, results of

the study may be biased. Multiple regression is also said

to be advantageous when cell frequencies are unequal and

disproportionate as in this study. This condition is not

unusual when the experimenter must deal with intact groups

where random assignment of subjects to groups is not possible.










Experimental mortality also causes unequal n's as in the

present study. The actual correlation of independent vari-

ables such as social class and university grade point aver-

age may also cause this to o cur (pp. 7, 187). With unequal

frequencies of treatment combinations the explanation of

variance becomes ambiguous between treatment and interaction.

Analysis of variance must handle unequal cell frequencies

by artificial means while stepwise multiple regression deals

with this problem through the use of semi-partial correla-

tions and the varying of their position in the multiple

regression equation. This allows the separation of treat-

ment and main effects (pp. 187-193).

Stepwise multiple regression analysis was done

using the BMDO2R program on the University of South Florida

computer (Dixon, 1973).

Hypotheses 11 and 12 of this study deal with the

significance of differences between coefficients of corre-

lation. The technique used for these comparisons was

Fisher's transformation to Z as recommended by Guilford

and Fruchter (1973).

The level of confidence used in this study was


p < .05.


















CHAPTER IV

RESULTS



The Scale of Democratic Attitudes


Stepwise multiple regression is used to test the

operational null hypothesis: neither group membership nor

time of testing are significant predictors of scores on the

Scale of Democratic Attitudes (SDA). Scores on the SDA

constitute the dependent variable. Dummy coded independent

variables include: experimental or control group membership

(EX/CON), pretest or posttest (PRE/PO), interaction between

experimental group membership and pretest or otherwise

(INTER), and sex (SEX). Continuous independent variables

are age (AGE), university grade point average (GPA), and

social class (SES).

The only independent; variable entering the regres-

sion equation is GPA, using subjects on whom complete data

are available. The observed F (1,153) = 6.8937 exceeds

the table F (1,150) = 3.91 (Kerlinger and Pedhazur, 1973,

pp. 512-515). The level of confidence used is p < .05.

Therefore, GPA is a significant predictor of SDA. The





55




stepwise regression analysis is shown in Table 5. The

table includes values for the coefficient of multiple cor-

relation (R) and F for the multiple regression equation.

Regression coefficients or beta weights are also given.

By comparing the value of "F to remove" with the table

F one can determine whether or not the contribution of

the relevant variable is significant when it enters the

equation last. If it is not significant, the variable is

removed. Thus an initially good predictor which has lost

its power can be removed. The variable with the highest

"F to enter" among those variables not in the equation is

entered if its contribution is significant. In the current

situation only one variable made a significant contribution

and the analysis is concluded in one step.

The results are not appreciably different when

subjects with missing data are included. As noted in

Chapter III, group means are substituted for missing data.

GPA is the only independent variable in the regression

equation. The observed F (1,213) = 8.1771 exceeds the

table F (1,200) = 3.89. Therefore, GPA is a significant

predictor of SDA. The stepwise regression analysis is

shown in Table 6.

These results do not allow the rejection of the

null hypothesis.



















S r- I- n 0 COD
0 ) m 10 N :T
J4J D 1- H N 10 '0
IC (N o 10 (Nl
1l
N N N N N
C u) m in n
0 U) U) Ul U) U)
0 4 I 1 -I -
w0 0



M ) 01
-4 1 r- (N (N m H


4 -H
cO1) M) M :





o4 .







0-0
--i ra Q)

au m
















4 rIa
H O



4O a0 o E
0 U >

tO
n q- Q3 r-

*- 0 m


>, 5






0 o







H a (N





0- 4>
(D -U






04)4 44
a) o 44


l 0 0











Qd W
fl MU
~,w l>






57











42 N r n m C co
ro i oDa oo
1o H
0 ,- -
*H
43






o HHH H
0
4- -4 *H
0 t4s P 0 r- 0 0
4JH .4 .o N n 0 o


(M4 C -z H 4 m (N

0 i 0 ( H 0 0 01

4-1- H 0>





tn
I 0 I
0 C 0

r a




4-) )
0 z04 C) H 1 (



w.4 M 0 C 0
4-)







a
0) l 00
*L 4J *d

0) 0



KA (0 'N
-

4) 0 a ri


0 r~



0 N





.0(4
(4
r~H 4.)
(40\

)02 (











The Subscale of Procedural Agreements of Democracy


The null hypothesis tested is: neither group member-

ship nor time of testing are significant predictors of scores

on the subscale of Procedural Agreements of Democracy (PAD).

Using subjects on whom complete data are available, no

independent variable entered the regression equation. In-

cluding subjects with means substituted for missing data,

GPA is the only independent variable entered. The observed

F (1,213) = 4.8863 exceeds the table F (1,200) = 3.89. The

regression analysis is shown in Table 7.

The null hypothesis cannot be rejected and is re-

tained.



The Subscale of General Statements
on Freedom of Speech


The null hypothesis tested is: neither group mem-

bership nor time of testing are significant predictors of

scores on the subscale of General Statements on Freedom of

Speech (GSFS). Using subjects on whom complete data are

available no independent variable enters the regression

equation. The same results are obtained including subjects

with means substituted for missing data. The null hypothesis

is retained.

























C,
o
4-1W











.-1i




04













0
1 0H

























40






-4
144








































,-A 4J
U






















0
r*-
a







































IMW


u U m
a) In o)


oa
4.)



OJ

Um)






4-3 -A
MD
t

0
44 >1



O
m m


>i 0



m e



t4-




4 0

a)


O



)
m






0
a0

>-I










The Subscale of Specific Application of
Freedom of Speech jand Procedural Rights


The null hypothesis tested is: neither group mem-

bership nor time of testing are significant predictors of

scores on the subscale of Specific Application of Freedom

of Speech and Procedural Rights (SAF). The only independent

variable entering the regression equation is GPA using sub-

jects on whom complete data are available. The observed

F (1,153) = 7.3105 exceeds the table F (1,150) = 3.91.

GPA is a significant predictor of SAF. The regression analy-

sis- is shown in Table 8.

The results are not appreciably different when sub-

jects with missing data are included. GPA is the only in-

dependent variable in the equation. The observed F (1,213)

= 7.5053 exceeds the table F (1,200) = 3.89. The regression

analysis is shown in Table 9.

The null hypothesis is retained.



The Subscale ofiPolitical Equality


The null hypothesis tested is: neither group mem-

bership nor time of testing are significant predictors of

scores on the subscale of Political Equality (PE). GPA

entered the regression equation as a significant predictor

using subjects with complete data. The observed F (1,153) =

12.4545 exceeds the table F (1,150) = 3.91. The regression

analysis is shown in Table 10.















r- o 0
CO f -V
m m 0
imO





N N N
Hn Hi H)
^-i 1 1 i-


o !








z o
0 H1



o IO

0


\D T 0
0 o






S 0













ix
ON O,
< u u


n 0 o
N to


0





O m
















--
O rd





24 (
U)



.H 0


Om
uu





UD
ac e













H <
00












l--I
m)
10 r-










u Z
S-I











tc O





r 0
E U
a4
U]




*d





m
.14








CA0


c
0
(d 4)
irl d
H 0



U




H


-r
0

rd




a












13









41
CI











0



4c1
4J








mB















CO C l






cq N ON









o m

Im
o a 0









ED M
(9 1D En
r' cn E


4-1

0
0

r-1

0 m



a'a
0 1
0


4-1 *r-
.Q4
4 ro







0-,

TIEn
u H
tien






O








,-I
ril



4J 0





.r-l





04
ua













00l
a)
oo











ED4
4-l
M0
Se,
tji
d) T


0
0




*H




o
'a






4r








rl
41
0
rl
4


































14



*rf
9)
H3
.0


o
r-l -d
CI
'a 4J
4-J j


0














o
u





r-


























a
>



























H
o



a)
J






E ;I










c~l


4J




0
4-I








HO

-0 4-






63







0 4 0, CO) 0 CO OD CO
4J u, ID o ) o 0 N
41 co m o 4



H ;


,1 4" 441 U) Ln In Ln In 0

r .- 4



,0 .,O
Id -PN N. NN Nm

41 H 'n n rH o HH


-H




4J 0
0-1 ( N




o o 0).

0 "0
H. U \




03 >
-H

















o ( 0

0 H4 Z 0
SP 004
(U +J N. CL













r-I Eg i ,
Ii >Ul 0 H (0






1 a l
ci )




O ,N










SH 4-1














U)
c i 4Jr 0















n W


a4 > 0)
a





04
0) I0
01r H- 0
(,cia U]4]
Ii) -H COI
1- ~o
Hd U]
E-r l0
0
0)6 4-
Hu
rl rl ~ 1 0
Vi(44
-H H






w I u0
(.0r I a
(Clu]v




m 4.)





64




Similar results are obtained when subjects are

included on whom incomplete data are available. GPA is

the only significant predictor in the equation. The ob-

served F (1,213) = 12.5797 exceeds the table F (1,200) =

3.89. The regression analysis is shown in Table 11.

The null hypothesis is retained.



The Subscale of Social and Ethnic Equality


The null hypothesis tested is: neither group mem-

bership nor time of testing are significant predictors of

scores on the subscale of Social and Ethnic Equality (SEE).

GPA and AGE entered and remained in the multiple regression

equation as significant predictors using subjects with com-

plete data. The observed F (2.152) = 6.632 exceeds the

table F (2.150) = 3.06. The regression analysis is shown

in Table 12. In the first step GPA enters the equation

with an observed F (1,153) = 7.5570 which exceeds the

table F (1,150) = 3.91. The variable with the highest

significant F and highest partial correlation among those

variables not in the equation enters next. AGE enters

in the second step with an observed F (1,152) = 5.485

which exceeds the table F (1,150) = 3.91. Both GPA and

AGE are entered last and an F test is performed to deter-

mine whether or not their contribution remains significant.

The "F's to remove" for GPA and AGE are, respectively,

8.2971 and 5.4854. These F's exceed the table F (1,150) =
















.n n mo
S- r-
N





N N N
N, N N
u3.
, ^- '
r- p f -


0




41









H

4


r-
0
O





N
H
N






Ln

0

0


4
0
t41












-4 .,- I
W4J

h CD


0













o
41 I
r-4







0>
o















































u
40
m



































OJ
rd
,-
CII







4J


U
dC


id
C44









a1


o N




0 0
m N



z 0




0 N

4 "-


O



UI
0




U4.1




4-Q


o


044
CC
IH











-r4 *



m H

0 -I4'




m4



aC
+1


w X M
SU) 'A






66







0 4 cN 0C) H 4 m co T
-4W ) Go Nl 'T m) ,-4 m LO (N 0
4 D -l o 0o N CO 0 N m r-m


1 H (( N a r aN r 0 rn r( r(1 (


0

*(4-I V r N N () m H H H GH


al 'o H H
4J oi m m a i m L i m







0 0 0

41 4 r4 0 a) P) 0 D '(
(0 -A a u 0 H



o a
4-1 >- a a I
4 4J 0 -n )) c
a) 4 i Ln N It X N)





o O0 0 a
o0 O N


4J 0 0 N
o m

0 H H H


H41 (N1 H Hn M Un


H -H
U) > (

1 ,(1) N:( 0
H aa)





a) 4-U r- 1

-H ( n .
(B 0 C r-U U








4c
41 -A m Ln i
ol >1 \
1)4 a rCo



























4.1
U)





4-1~ a
Mr r :C










3.91. No other variables could make significant contri-

butions. The analysis is concluded in two steps. The re-

gression coefficient for AGE is negative, indicating an

inverse relationship with SEE.

Similar results are obtained using subjects on whom

incomplete data are available. GPA and AGE are shown to

be significant predictors of SEE. The regression analysis

is shown in Table 13.

The null hypothesis is retained.



The Subscale of Economic Equality


The null hypothesis tested is: neither group mem-

bership nor time of testing are significant predictors of

scores on the subscale of Economic Equality (EE). GPA and

SES enter and remain in the equation as significant pre-

dictors using subjects with complete data. The observed

F (2,152) = 5.537 exceeds the table F (2,150) = 3.06. The

regression of coefficients for both GPA and SES are negative

which shows an inverse relationship between these variables

and scores on EE. The regression analysis is shown in

Table 14.

Using subjects with missing data GPA does not

enter the equation. SES and AGE are shown in be signifi-

cant predictors where the observed F (2,212) = 6.005 ex-

ceeds the table F (2,200) = 3.04. The regression coeffi-

cients for both SES and AGE are negative indicating an















4 a) m c o .0 c0 D It c
Sr m m r o 0 0 ccn Ln Ln rN
C 0 o o 0 N o o o c o



0 0 N N N 0 N N NH H H
O H H H H H H H H H H H H
SN. N N N. . . .




4 ~1 0 ,m m 0 m m
0 I r-, 0) o
J 4 Q Ln ID m T m U Q 1 Ln i



4 >

0 -H E H 0
O I I I I


x z w X 2




0 > 0 ,-r- c
0 "
H


-P F3 0


1 1 00
U) a
0 >, 0 0





rd -q O c
HmII




















oc
0- 0 Na I
























00 Oc 4 0
S N N c



9 0 N










4-) 4J ) P4
H &n1 v rn u
I3 0 C N cc





Hl -H




o 44 44 *
















rC-






69







0 N 0 0 -1 N 0 CO CO Lr
4j o 3 4 r- 0 H H N Ln Ln
4I 0 0 co o N I 0 H ( 4n
(N 4 (N



4 H H H H H H H H H
0
*il N. N NN NN N N N l r l rN l


Srl l
| m a



,- 1

r- H D 4 O CO CN r,
4 m cc m m m Ln 0 Ln Ln m 0 N
0 l (i N 4D O O D O N n cO CO ;T O


0 a
E rl< u
o 0H
o 3
0 o


a 0 4 Pi 0 04 4
Ha H u






u o O ,4
d x x a w
42H> w (n H V) U) f14 H U)




3a



SIN
4ul a)
() 4-4

41 4-I C 0


0 0 42i


in inn m .1o




. r-


- c z

4'1 a (


0) H r0 O H


PU I I I

(M ( L

() 10


1) > r a) 04 w
I .d- 4H U WO
M4 r-










inverse relationship. The regression analysis is shown

in Table 15.

The null hypothesis is retained.

While experimental group membership and posttest

are not shown to be significant predictors of SDA or any of

its subscales, other variables are shown to be so. Multiple

regression analysis provides for significance testing and

a measure of meaningfulness through the coefficient of de-

termination. The coefficient of determination (R2) measures

the amount of variance in the dependent variable explained

by the independent variable. Tables 16 and 17 report R2's

and mean scores on the SDA and its subscales for experi-

mental and control groups. These tables provide a summary

of results obtained thus far. The variable most often sig-

nificant is GPA. Its R2 values show that the proportion

of variance explained by GPA range from 2.24 percent to

7.53 percent. While GPA is often significant, its R2

values do not point to any particularly meaningful conclu-

sions. GPA, AGE and SES are not variables that are normally

manipulatable.

Visual inspection of mean scores for experimental

and control groups on protests and posttests show their

similarity. This indicates their lack of significance as

predictor variables.
















~o Io
o Hn
N- H
Ni


rl 0 'D 00

N N 0

m H


(N(N H N H H
H (N H H
(N(N H N(N (
rlN l- Nl l


o ( m H

SH-4 (N Hz1
o o 0 I


SIn nm


N 0 0


SD








H-I oH -
(N (N (N







I m
CO
0 .I .l


S0 PX ZW
wH H H X H M H w 0


r
0

H H

H







0


rl














EI


(N Hr

m CO
o o








4 14
U M


-i (N


0




(N (N
r( (N

(N H


Ln

N


tI








0
u

4-1


U


0



Li

H 4


(




















O N
En C,









m mH
m 0

401 4


H ON O
H Cl
ON N C


m r-
m 4
0 O


COm

S41

0






E0 m
r4 o
0 4
C4-)











4J 41
C 01

Cm

So







41 4J
C







(11w
54

cw4

aC)


>!


C))


ft EC oC rt < C
04 04 04 24 24 04
CD C (3> 04 U 04


c4



0
0





dM
0 r


ED 00)









ro
H -A-

E- U





0)

S t
u
H Hk
04




44


Q)


0
u
C-


H Co 0 C
N 0 4 N
4 mmON C 4
0 00 0


U)
<: Q h44
Q < U} ii
un c14 [0 u) &


o m
n n

C C


o Jn m Gn eY

En H- Ct









0O "- 'r --l ^D
En CN En Cl O
m H 0 ON m
1) 4 H r4
Ln IT


-T 4J






> rd
OH
C4J
U44 V *d







00




44.4
CC



OH








04
oQ
4J
CC )
04 H

> o
04 H
0U0*
0>i-


















H1-
On 0H r in CO
WN Co H ro 0N H N( C









S4- 4J

-41 (-


-4. + i in (N C. CO C.
a) 00 ) Ln o C H N H
Im s r a a a

uiI








44 .




0 ON N
C C
H cH








0 o 0 H 0 H 0 C O C




4 -, N in CO n
co -P d 4-i ]| .1
S140 ol l C (N 0 0 CO C CO
1) 0 CO in -C m H (N -1








04
U4 x

a



ra H













'0
5C a C < < < < t
41 1 0 >




4d c d *



Vc ul I 0 ODcl m o .
I e n m -4 N 1)
Ea Wr4u


U4U En m (1m1n m n
44 (C 4 3
0001 'C < m < a' aC CO Co



mW cC c


4-1
4-1 14-1 C
a) 00


S 0 C 0 CO CO CO CO
1I U0Q m ( U m CO n CO m in

0-142




U tU










The Dogmatism Scale


The same statistical procedures are used to ana-

lyze scores on the Dogmatism Scale (DOG) as on the SDA.

The null hypothesis tested is: neither group membership

nor time of testing are significant predictors of scores

on the Dogmatism .Scale. GPA and SEX are shown to be sig-

nificant predictors using subjects with complete data. The

observed F (2,152) = 10J451 exceeds the table F (2,150) =

3.06. R2 is .1209 with .0845 attributable to GPA. The

regression analysis is shown in Table 18.

GPA is the only variable to enter the equation

using subjects with missing data. The observed F (1,213) =

14.8681 exceeds the table F (1,200) = 3.89. The coefficient

of determination is .0652. The regression analysis is

presented in Table 19.

The null hypothesis is retained.



The Personal Beliefs Inventory


Stepwise multiple regression is used to analyze

scores on the Personal Beliefs Inventory (PBI). The null

hypothesis tested is: neither group membership nor time

of. testing are significant predictors of scores on the

Personal Beliefs Inventory. No independent variables are

significant using subjects with complete data. Using sub-

jects with missing data, SES and SEX enter the equation.














O o4 H H c1 co cc m -1 -
S4J H 0 t- -H H C N m
4J c CO O a c c cc cN c o c




CN ro o ID m 4DI ID 1 INI
-H ( -d r l 0 H l rIl 0 D D r- ID



0
4-a H z ;I 0
r4 m c n o u D cc rH c









rH 1 0 w
0, 1 n x 1
O i O 4 H m w 4
4 00 I 0
N*H





4 r0i z o 4 o
OO H 0 c
Q 0










QC 441 0 ) *
o aO o S
















0)
Qo Ic co























0.Iw H H






U) H X 4 x
Hm ( 4

























,P 0 U 0 U
'( H H H





0 N kN N N




Q) cci
UI










oJ
H m
(U c cc C' cc

44



Cc
Sc











































U [ [


.H
4J


-,I






0)

Cd
an










OGid








-,
Q)












a)
a)
l-1










4-)











m
(d
4C



0 )
*d
M,


w:





0)


r,






ca
0




rd [







-4
0












rE
-41



































4Q4
U)
C-I








cr-





0.
a,4



o


C-)










The observed F (2,211) = 6.849 exceeds the table F (2,200)

= 3.04. R2 is .0610. The regression analysis is shown

in Table 20.

The null hypothesis is retained.



The Teacher Practices Inventory


Stepwise multiple regression is used to analyze

scores on the Teacher Practices Inventory (TPI). The null

hypothesis tested is: neither group membership nor time

of testing are significant predictors of scores on the

Teacher Practices Inventory. GPA, EX/CON and INTER enter

the regression equation using subjects with complete data.

The observed F (3,137) = 8.241 exceeds the table F (3,125) =

2.68. The coefficient of determination is .1529. The re-

gression analysis is shown in Table 21.

Inspection of mean scores on the TPI shown in Table

22 might lead to the conclusion the experimental group

scores are significantly greater than the control group

scores. Caution would seem to be indicated by INTER's

significant contribution. INTER shows a significant inverse

relationship between experimental group membership and pre-

test scores. Additionally, PRE/PO did not enter the equa-

tion. This would appear to be the result of control group

scores. Finally, the small number of subjects involved may

distort these results.















Ou m 0 O CH
0 4 H '; (0 H

C Ln .Ln


H H H H- r-I O O OH
H H H H H H H H H-
\ '\ '\ '\ \ \ '\ \


m (N
o C
I-


z 0 U 0

0\ P C 0 P
X C U N D x
r 4 H P4 t En CD w 4


N N
^0 H-
r- ri
N m




o o
H~ H
N N
rl rl


C w
U 0
4 U


>1

0
4-)

r(


H


a)

4J4
-r






<0
a







Q)
Qn ra




4-)






On
H 3l





C4

>rl



0
-t





0


u)


Q,
0


HO
H4





U




H


S)
H






0
IU
U)


41


U
*H
U
Ut

a)
o
u





HO







Lc
4J
[)


U) c) UC
(0 0 9






79






0 (o 1-4 L (N Ln r'- N -C (N o(N 1-1 "N cO Ln
0 N (n m (a m -4 (N ( r- mN O O c o o
m m r- co o i n m o Ln Ln
4 ao oo 0 l cc (N r- r-H r- r- r- H ( m H on





.1 ( (N m m ( n m m m m I m m
4j a l i- i-I -I -I -I -I -I -A -I -I -I -I

0



SO n o o m N Ln o H o H H H H H
> 4J 4J '-N 'o N o C N I T r' 'N o 'N 'N 'N 'N

a H O H H H H HO O O O O O O

0U U 1 l I I 1I I I




u M l O 4 (C 'p, p( P
(d > (N w N ; N H (
Id i x u)M w i p In l M X u)
P4i4 di Hi
1-1 >O
o m H r0- ( C(
Ln m IT ( r~




Cd 'N 4H O d< C n H U' C


4- aQ U)

(N ( o
d 4-4 0 (N (N ( (





ra o D
m (N Ho H







Sm m cq
H lo (N H o Hm c







441 01
z E






4 (N (NN N0 0N N
a) I N L 'N N'






C) I










Table 22

Mean Scores on the Teacher Practices Inventory
Using Complete Data


Pretest Posttest



160.85 171.88
Experimental n=53 n=48


158.05 156.16
Control n=21 n=19


GPA and PRE/PO enter the equation including sub-

jects with means substituted for missing data. The ob-

served F (2,211) = 8.241 exceeds the table F (2,200) =

3.04. The coefficient of determination is .0725. The re-

gression analysis is shown in Table 23.

The null hypothesis is retained.



The Change in Relationship Between General
Statements on Freedom of Speech and the
Specific Application of Freedom of Speech
and Procedural Rights


The change in correlation of scores on GSFS and

SAF between experimental and control groups on protests

and posttests is explored. The study seeks to determine

whether or not experimental group membership contributed

to a significantly greater correlation between general






81








o m ) co r r7 r' m T o0 r
O, mN (N U) ( U 0 O NO 0
-41 m o c o U -- N m 0
m 4-1 - (NN r N




r- r- r- r-H (

O I

0H l H4I
> -p r4 r- Ln o -T 'T r- m Ln N





0C 4 w H C (
CU) >
-I H

U n z o z
0) 0 H H 0 0o4o
E-1 I Q
U )



S0 m
U) a CN cr03 N
S O- 0 ,








U0 0 4J0
.1 .1 r U 1< 2 in










10 IoU H N U
a) a) U0
Q4 a)

m o
EN 141 'U)



H ) (.4 H (NC24
0 H
4- 1
N (N



o o c4

U) 3 U4 D In o
U) U) -dI rU ) T


U o 0 r0
PU H-
0) U)


)4* aH Uo o o '
0) I -. a) Na oN
4) : 1 a01





4-1











attitudes and more specific attitudes on the same subject.

If so, such results might indicate that experimental group

experience contributes to the integration of the student's

valuational-attitudinal system. The null hypothesis tested

is: there is no significant different in the change in the

correlation of scores on the subscale of General Statements

on Freedom of Speech and the subscale of Specific Applica-

tion of Freedom of Speech and Procedural Rights between the

control group and the experimental group. Statistical pro-

cedures involve the calculation of correlation coefficients

(r) of scores on GSFS and SAF. The data deck is divided

into experimental group pretest, experimental group post-

test, control group pretest and control group posttest.

These groups are then subjected to stepwise multiple re-

gression which produces the desired.correlation matrix.

Table 24 shows the results of this procedure using subjects

with complete data.



Table 24

Correlation Coefficients of Scores on General Statements
on Freedom of Speech and the Specific Application
of Freedom of Speech and Procedural Rights
Using Complete Data


Pretest Posttest


Experimental r=.387 r=.527
n=52 n=55
Control r=.580 r=.405
n=30 n=18





83




Fisher's transformation to Z is used to compare

differences between r's (Guilford and Fruchter, 1973, pp.

166-67). This procedure involves the computation of the

standard error of a difference between Fisher's Z's using

the formula:




ad n 1-3 n -3
z 1 -2



when n and n equal the number of subjects. Values of

Fisher's Z's for particular r's are available in Guilford

and Fruchter's Table H (p. 524). These Z coefficients are

then entered in the formula:



y -1 2
Z -
od
z



Since the sampling distribution of Fisher's Z is normal,

the sampling distribution of Z -Z is also normal. Z may

be interpreted as a standard score. If the difference in

Z's deviates from a difference of .0 to the extent of 1.960,

then the difference is significant at the .05 level.

Experimental group pretest r is compared to control

group pretest r.



a = + .239
d 52-3 30-3
z


S .662-.408 1.063
239063
.239





84




Since the value of 1.063 is less than 1.96o, a null hypoth-

esis of no difference between r's is retained.

Control group pretest r is compared to control

group posttest r.



= .322
z 30-3 8 18-3


.662-.430 .7
.322


Since the value of .720 is less than 1.96o, a null hypoth-

esis of no difference between r's is.retained.

Experimental pretest r is compared to experimental

posttest r.



S + 1 .197
d 52-3 55-3
z


= .586-.408 = .904
.197904



Since the value .904 is less than 1.96o, a null hypothesis

of no difference between r's is retained.

Experimental posttest r is compared to control post-

test r.


1 1
d = -3 + 18-3 =293


.586-.430 .532
.293











Since the value .532 is less than 1.960, a null hypothesis

of no difference between r's is retained.

Comparisons are also done using subjects with means

substituted for missing data Table 25 shows the coeffi-

cients of correlation for these subjects.



Table 25

Correlation Coefficients of Scores on General Statements
on Freedom of Speech and the Specific Application of
Freedom of Speech and Procedural Rights
Using Adjusted Data



Pretest Posttest



Experimental r=.364 r=.507
n=74 n=70

Control r .580 r=.354
n=43 n=28


Experimental group

group pretest r.


pretest r is compared to control


d = 1 + .197
d 743 433
z


.752-.382
.197


= 1.878


Since the value of 1.878 is less than 1.96o, a null hypothe-

sis of no difference between r's is retained.





86





Control group pretest r is compared to control

group posttest r.



1 1 255
d i3-3 + 28-3 .255
z


z .752-.370 1.498
.255


Since the value of 1.498 is less than 1.96o, a null hypoth-

esis of no difference between r's is retained.
I -
Experimental group pretest r is compared to exper-

imental group posttest r.



d 74-3 70- = 170


S= .559-.382 =1.041
.170



Since the value 1.041 is less than 1.96o, a null hypothesis

of no difference between r's is retained.

Finally, experimental group posttest r is compared

to control group posttest r.



d 0-= 3 + 28-3 235
z


.559-.370 .804
.235804










Since the value .804 is less than 1.96 a null hypothesis

of no difference between r's is retained.

The results of the above comparisons using both

complete and adjusted data d6 not allow the rejection of

the original null hypothesis



The Change in Relationship Between
Rankings of Freedom and Equality


The Value Ranking Test (VRT) is used to explore the

correlation between rankings of freedom equality. The study

seeks to determine whether or not experimental group member-

ship contributed to a closer correlation between these two

values. Such results would indicate that experimental group

membership promoted the understanding of the relationship

between major cultural values and the integration of the

student's value system. The null hypothesis tested is:

there is no significant difference in the change in the cor-

relation of the ranking of freedom and the ranking of

equality on the Value Ranking Test between the control

group and the experimental group. The procedures used to

explore the relationship between GSFS and SAP are used in

this.case also. Table 26 shows the correlation coefficients

for subjects with complete data.

Fisher's transformation to Z is used to compare

experimental group pretest r with control group pretest r.










Table 26

Correlation Coefficients of Rankings of Freedom
and Equality Using Complete Data



Pretest Posttest



Experimental r=.306 r=.518
n=53 n=48

Control r=-.131 r=-.160
n=21 n=19





o -1 1 .276
d 53 21-3 276



Z- .316-(-.132) = 1.623
Z.276 1.623
.276



Since the value 1.623 is less than 1.96o, a null hypothesis

of no difference between r's is retained.

Control group pretest r is compared with control

group posttest r.



od = 1-3 1T + = .345
z


S -.161-(-.132)
Z .345 = .084




Since the value .084 is less than 1.96o, a null hypothesis

of no difference between r's is retained.




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 - Version 2.9.7 - mvs