{4mg 15:6 “20.00

ABSTRACT
SOCIALIZATION, PERSONALITY AND THE
ORIENTATION TOWARD CHANGE

by Joseph M. Allman

This research deals with differences in socialization experiences
and psychological attributes related to an individual's ability to
adjust to changing social and political conditioner In addition to a
static analysis of the interrelationship of such variables, I have
sought to capture some of the reality of becoming a person by examining
impacts of varying sequential patterns of socialization upon sets of
personality attributes.

The first chapter reviews both theipsychoanalytic perspective
upon human deve10pment with its attention to early learning in the
family and attitudinal stability>and cognitive theoretical approaches
which(emphasize later learning and personality change. The balance of
the literature review identifies socialization experiences and personality
attributes expected to have an impact upon the orientation toward change.

Chapter II outlines a theoretical sketch which interrelates
variables and generates hypotheses guiding the analysis. This sketch
describes(four patterns of socialization in relation to several sets
of predispositions including the orientation toward change;

The study design and instrumentation are discussed and evaluated
in the third chapter. The research population is a non-random group

of young people (N - 241) from towns and cities throughout the state

Joseph M. Allman

of Michigan who has participated in summer training programs for
political activists held on the campus of Michigan State University
between 1963 and 1966.

The remaining three chapters present the data and analysis.
Chapter IV contains a discussion of the interrelationship among the
several psychological attributes observed in the study-jéogmatism,
reliance upon authority, trust, liberalism and efficacy.) A multi-
variate statistical model is used to evaluate the impact of these
attributes upon the orientation toward change.

Chapter V discusses thefsocialization variablesjincluding:
extent of communication with authorities in the immediate family and
later in life outside of the family; degree of consistency between early
and later experiences with authority; extent of crisis during sociali-
zation process; age; and education. Both multi-variate and analysis
of variance models are used to examine the relationships among the
socialization experiences, the psychological attributes and the impact
of four sequential patterns of socialization upon the orientation
toward change.

In the last chapter I have summarized findings and reformulated
the theory to be consistent with these observations. \We find that‘;'
trust, dogmatism, and reliance upon authority are greatly affected by
early experience with authority in the family and are relatively
resistant to change) Liberalism, efficacy and orientation toward
change appear to be results of later experiences with authority

X
outside of the immediate family.‘xCrisis experience in socialization

Joseph M. Allman

evidently has profound impacts upon the personalitynnginally, and

most importantly, we find that the four socialization patterns do

produce significant variations in at least two of the personality

attributes--reliance upon authority and the orientation toward change.
I conclude that the attempt to utilize a dynamic model of

human development has shown the possibility for more realistic

theorizing and research into the complex relationship of human

learning, predispositions and behavior.

SOCIALIZATION, PERSONALITY AND THE

ORIENTATION TOWARD CHANGE

By

Joseph MffiAllman

A THESIS

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

DOCTOR OF PHILOSOPHY
Department of Political Science

1968

ACKNOWLEDGMENTS

The thought guiding this study has evolved from seminars and
conversations with Frank Pinner and Milton Rokeach. Frank Pinner
tried very hard to make this a better dissertation, and I have
appreciated and learned from his effort.

Many have shared in this process and one generated it. Micky
McCleery is of a lost genre--friend and teacher who makes the possible
more probable.

My wife Judith, my sons Jim and Barrett, and my daughter Laura,
each in their own very special way, have assured the dissertation's
completion.

I have made mistakes in thinking about and doing this research.
Some I have realized, some remain, but all, contrary to more customary

refrain, I accept willingly as part of becoming a student.

Joe Allman
Eugene, Oregon, 1968

ii

TABLE OF CONTENTS

Page
LIST OF TABLES . . . . . . . . . . . . .
Chapter
u’I. BACKGROUND FOR STUDY . . . . . . . . . 1

Statement of Problem

Political Socialization and Orientations

Theoretical Approaches to Personality
Development and Change

Primary Socialization and Orientations

Secondary Socialization and Orientations

Personality and Orientations

Summary

II. THEORETICAL CONCERNS IN THE ANALYSIS OF THE
ORIENTATION TOWARD CHANGE . . . . . . . . 35

Personality, Socialization and the Orientation
toward Change.

Theoretical Sketch

Statement of Hypotheses

III. STUDY DESI“ O O O O O O O O O I O 51
Subjects

Instrumentation
Reliability and the Inference of Attitudes

IV. PERSONALITY ATTRIBUTES AND THE ORIENTATION
TOWARD CHAN GE 0 O O O O O O O O O O 7 7

Structural and Substantive Orientations
Correlates of the Orientation toward Change
Summary

111

TABLE OF CONTENTS--Continued

Chapter Page

o'V. SOCIALIZATION EXPERIENCES AND THE ORIENTATION
TOWARD CHANGE O O O O O O O O O O I 9 5

Primary and Secondary Environments and
Personality Attributes

Continuity, Education and Personality

Socialization Patterns, Personality and
the Orientation toward Change

Summary

VI. EVAIaUATION OF THE THEORY e e o o o o e e 123
Review of Expectations and Observations
Discussion and Reformulation
Conclusions

BIBLIOGRAPHY O O O O O I O O O O O O O O 152

iv

10.

11.
12.

13.

14.

15.

16.

LIST OF TABLES

SOCIALIZATION PATTERNS AND RELATED PERSONALITY
ATTRIBUTES O O O C O O O O O O O

COMPARISON OF COEFFICIENTS OF STABILITY AND

EQUIVALENCE . . . . . . . . . .
CORRELATION MATRIX OF PERSONALITY VARIABLES . . .
DOGMATISM AND RELATED PERSONALITY ATTRIBUTES . . .
RELIANCE UPON AUTHORITY BY RELATED PERSONALITY

ATTRIBUTES . . . . . . . . . . .
TRUST BY RELATED PERSONALITY ATTRIBUTES . . . .

LIBERALISM BY RELATED PERSONALITY ATTRIBUTES . . .
POLITICAL CYNICISM BY RELATED PERSONALITY ATTRIBUTES .
REGRESSION COEFFICIENTS, ZERO-ORDER CORRELATIONS,

AND T-VALUES FOR ORIENTATION TOWARD CHANGE

RELATED TO OTHER PERSONALITY ATTRIBUTES . . .

THE ORIENTATION TOWARD CHANGE BY RELATED PERSONALITY
ATTRIBIJTES O O O O O O O O O O O

CORRELATION MATRIX OF SOCIALIZATION VARIABLES . .
PRIMARY ENVIRONMENT BY RELATED PERSONALITY ATTRIBUTES.

SECONDARY ENVIRONMENT BY RELATED PERSONALITY
ATTRIBUTES . . . . . . . . . . .

CONTINUITY BY RELATED PERSONALITY ATTRIBUTES . . .
EDUCATION BY RELATED PERSONALITY ATTRIBUTES . . .
REGRESSION COEFFICIENTS, ZERO-ORDER CORRELATIONS,

AND T-VALUES FOR ORIENTATION TOWARD CHANGE

RELATED TO SOCIALIZATION EXPERIENCES AND
PERSONALITY ATTRIBUTES . . . . . . . .

Page

50

73
78

79

81
82
83

84

87

89
96

97

100
102

105

107

LIST OF TABLES--Continued

Table Page

17. ANALYSIS OF VARIANCE WITHIN PERSONALITY ATTRIBUTES
AND BETWEEN EXPERIMENTAL SOCIALIZATION
PATTERNS O O O O O I I C O O O O l 1 1

18. MATRIX OF T-VALUES RESULTING FROM THE COMPARISON
OF MEANS OF PERSONALITY ATTRIBUTES AMONG
THE EXPERIMENTAL GROUPS . . . . . . . . 112

19. COMPARISON OF PREDICTED AND OBSERVED CLASSIFICATIONS

OF PERSONALITY ATTRIBUTES FOR EACH OF FOUR
EXPERIMENTAL GROUPS . . . . . . . . . 113

vi

CHAPTER I
BACKGROUND FOR STUDY
Statement of Problem

To begin with, I will sketch the problem with broad strokes.

We as producers and products of our environment, are creating such
rapid changes in our world that there exists the distinct possibility
of our failing to adapt our world view and consequent behaviors rapidly
enough to avoid our own extinction. The problem, then, that faces us
is one of individual and social adjustment to change. The question we
must answer is: What are the conditions that facilitate or inhibit
individual ability to accept change?

The problem may be reduced to more manageable proportions. As,
an example I have chosen two quotes from public statements made during
the "revolution" on the Berkeley campus in the fall of 1964. These
statements represent what was said and heard during the most intense
moments of the revolt, and further, they indicate the direction and
degree of division in student opinion. It is as if we had a snapshot
of a single event in the process of adjustment to change. This brief
moment illustrates the problem of this study.

It is a bleak scene, but it is all a lot of us

have to look forward to. Society provides no challenge.

American society in the standard conception it has

of itself is simply no longer exciting. The most

exciting things going on in America today are movements

to change America. America is becoming evermore the

utopia of sterilized, automated contentment. The 'futures'
and 'careers' for which American students now prepare are

1

for the most part intellectual and moral wastelands.
This chrome-plated consumers' paradise would have us
grow up to be well behaved children. But an important
minority of men and women coming to the front today
have shown that they will die rather than be standard-
ized, replaceable, and irrelevant.

Mario Savio

The organization which I represent, University Students
for Law and Order, was organized as an alternative course
of action to the so-called Free Speech Mbvement. We
believe that the FSM has exceeded the limits of protest
acceptable to the majority of students at our university.

There is no need nor is there any excuse for civil
disobedience on this campus. Those students involved
demand protection of their rights while at the same
time they are violating our rights.

As an organization USLO does not pretend to know
solutions as to the varied and complex problems
which presently confront us as students. But one
fact is undeniably clear--that unless we are willing
to express ourselves as a responsible body, we have,
no right to expect the continued financial support
of the community at large. ‘

As a course of action, USLO urges the students
to support the legally constituted administration of
this campus on all issues until such time as the civil
judicial authorities dictate otherwise.

We therefore request that you express your moral
support for'our viewpoint by signing the petition which

is now being circulated and by disassociating yourself
from the FSM.

R. F. Dussault

These statements are almost pure expressions of polar types
on an often noted continuum of orientation toward change.1 Duverger

reviewing some of the sociological conceptions of this "natural

 

1S. M. Lipset and S. S. Wolin (eds.), The Berkeley Student
Revolt (Garden City, N. Y.: Anchor Books, Doubleday and Co., Inc.,
1965), pp. 219, 226.

political dualism" suggests that, "It is a summary and approximate
view but not altogether inaccurate."2 He continues, "It is true that
some find themselves completely at home amongst commonplace ideas,
accepted traditions, and conventional habits, whereas others experience
the compelling need to change everything, to modify everything, and to

3 As a result of these basic orientations

innovate in all domains."
he sees a "natural movement of societies" toward a two-party system.
He describes what may be called a collective or cultural orientation
toward change in the following way. "Whenever public opinion is
squarely faced with great fundamental problems it tends to crystallize
around two opposing poles."4
This commonplace observation of the polarization of political
cultures and sometimes (although not often) of the political system
itself under the stress of accommodation to change is the impetus
for the present study. In conflict situations, people tend to group
themselves into those who resist change and those who are predisposed
to accept it. Indeed, Huntington has suggested that acceptance of or
resistance to change is the fundamental difference between the liberal

and conservative ideologies.5 The present study is an attempt to

understand individual response and predisposition to change.

 

2M. Duverger.‘ Political Parties (New York: John Wiley and
Sons, 1963), p. 215. ‘

3Ib1de’ pp. 215-2160

4Ibid., p. 216.

S. P. Huntington. "Conservatism as an Ideology," in ggmpggggigg
Politics: Notes and Reading, ed. Macridis and Brown (Homewood, 111.:
The Dorsey Press, 1961), pp. 276-284.

The present study builds upon research insights of The
Authoritarian Personality, a study which directs attention to behaviors
and psychological predispositions to act such as personality syndromes,
belief systems, attitudes and orientations.6 A growing, though still
fragmentary, body of empirical research is being done under the rubrics
of pOlitical socialization, political culture and public opinion. This
research inquires into attitudes and behaviors and their macro-variable
counterparts--culture and system. If there need be rationale for inquiry
Ithiel De Sola Pool one exponent of the approach adopted here--argues

persuasively, I think that,

. . . it is common to assume that changes in men's actions
are the really important objectives and that changes in
attitudes are but means toward the desired actions. We
would argue, however, that it is the other way around. It
is, for example, relatively easy to get peasants to plant
a particular kind of seed a foot apart instead of six
inches apart. This action can be induced by money payments,
by terror, by authority, by persuasion, by proving it to
be the will of the gods, and by other means. But the
improvement of one such practice does not mean that the
peasant has been in any way modernized. A far more
significant change would be the development of scientific
attitudes toward the adoption of new practices. It is
only that kind of internal changes in the latent structure
of his attitudes that would produce self sustaining
movement toward modernization.

In summary my research deals with differences in socialization
experiences and psychological attributes related to the individual's

ability to adjust to change.

 

6T. w. Adamo, E. Frankel-Brunswik, D. J. Levinson andR. N.

Sanford, The Authoritarian Personality (New York: Harper and Row,
Publishers, 1950).

7Ithiel De Sola Pool, "The Mass Media and Politics in the

Moderniz tion Process," in Communications and Political Development
ed. Pye Princeton, N. J.: r nce o n vers y ress, , p. 49.

My objectives in choosing this problem and approach are broad
and largely unattainable, but they do serve to provide both context
and direction to the effort.

A central problem for students of human behavior is to under-
stand personality development and change. I have chosen to study
the processes of learning and cognitive adaptation in order to identify
some of the human conditions producing and changing personality.
Ultimately the understanding of this problem can lead to information
and skills necessary to facilitate personal change.

A second objective is to study political orientations as they
relate to both personality and socialization experiences. Students
of political socialization are generally interested in human learning,
personality and behavior. Studies of the effects of socialization
upon political attitudes or the effects of personality upon political
behaviors are relatively numerous. But seldom do these studies attempt
to bring these variables together in a single research effort.

A further objective needs to be included because most of the
research in political socialization focuses upon young children or
adolescents and not upon young adults. I propose to question the
assumption underlying such research--the assumption that individual
socialization is largely completed before the person is exposed to
adult experiences. It seems important to inquire if later life
experiences can or do affect citizen socialization.

A final objective is to understand some of the ingredients of
an accommodative or "modernized" personality. The ability of individuals

and groups to accept new ways of acting and thinking may be the single

most important precondition for the continuing develOpment of our own

country as well as the "new nations."
Political Socialization and Orientations

This literature review begins with the problem of determining
what socialization experiences and psychological orientations may affect
individual adjustment to change. Some clarification of the concepts of
socialization and orientations as they will be used in this study is

required.

Definitions of socialization are varied, as the following

quotations illustrate.

. . . the whole process by which an individual born
with behavior potentialities of enormously wide range,
is led to develop actual behavior which is confined
within a much narrower range--the range of what is
customary and acceptablg for him according to the
standards of his group.

Conformity to rules then, with the exception of
special cases, ensured primarily by the process of
socialization--that is, via the development in each
individual of habits which lead him to make responses
whichgconform to the rules instead of transgressing
them.

1.
. . . his learning of social patterns (regularities of
behavior) corresponding to his societal positians as
mediated through various agencies of society.

 

8I. L. Child, "Socialization," Handbook of Social Psychology
ed. Lindzey (Cambridge, Mass.: Addison and Wesley Co., 1954), p. 655.

9I. Whiting and I. L. Child, Child Training and Personality:
A Cross Cultural Study (New Haven, Conn.: Yale University Press, 1953),

p.‘221.

1OJ. Hyman, Political Socialization (Glencoe, 111.: The Free
Press, 1959), p. 25.

. . . it is useful to simplify the complex process of
political socialization by viewing it as the means
whereby members of a political system acquire the 11
three types of basic orientations already mentioned.

For some socialization refers to learning predispositions to behave
(e.g. habits, attitudes, orientations, or expectations) while others
emphasize learning actual patterns of behavior. For present purposes
socialization refers to experiences affecting the development of
individual orientations.

There is enough ambiguity concerning the concept of orientation
to require discussing its use in the present study. Smith, Bruner
and White conceptualize three "action tendencies" (i.e., approach,
avoidance and hostility) which are activated by objects in the person's
psychological field.12 They refer to these tendencies as the "orien-
tations of the attitude." Easton and Hess posit three types of learned
expectations which people use as cues for interpreting political
situations. The latter study equates orientation and expectations,
"These three kinds of expectations--that is, knowledge, values, and

attitudes--we shall call the basic political orientations."13

 

11D. Easton and R. Hess, "Youth and the Political System,"
Culture and Social Character, ed. Lipset and Lowenthal (Glencoe, 111.:
The Free Press, 1961), p. 229.

12B. Smith, J. Bruner and R. White, Opinionsggnd Personality
(New York: John Wiley and Sons, 1956), p. 37.

13Easton and Hess, 92, cit., p. 228.

Almond and Verba argue,

Orientations refer to the internalized aspects
of objects and relationships. It includes: .(1)
"cognitive orientation," that is, knowledge of and
beliefs about the political system, its roles and
the incumbents of these roles, its inputs, and its
outputs; (2) "affective orientations," or feelings
about the political system, its roles, personnel,
and performance; and (3) "evaluational orientations,"
the judgements and opinions about political objects
that typically involve the combination of value
standards and criteria with information and feelings.

In this definition the orientation becomes "the internalized aspects
of objects and relationships." Orientation as used in the literature
is synonymous with Opinion, attitude, belief, expectations, character,
style, predispositions, tendency, and in some cases, personality.

The concept requires some refinement.

One theme emerges from the variations. In most cases orientation
refers to an internalized predisposition to behave. The term shares
common attributes with the concept of attitude. We may gain clarity by
examining a definition of attitude.

I have chosen to review Rokeach's formulation of attitude which
is drawn from his broader analysis of belief systems.15 Rokeach uses
the belief as the unit of analysis. He defines a belief, " . . . as

any simple proposition, conscious or unconscious, inferred from what

 

146. A. Almond and S. Verbs, The Civic Culture (Princeton, N. J.:

Princeton University Press, 1963), p. 15.

15M. Rokeach, The Open and Closed Mind: Investigations into

the Nature of Belief Systems and Personality Systems (New York: Basic
Books, 1960).

a person says or does, capable of being preceded by the phrase 'I

believe that'. . . ."16

Beliefs describe an object or situation as
true or false, evaluate the situation or object as good or bad, or
advocate a certain course of action as desirable or undesirable.
Given these three types of beliefs, Rokeach argues, that each belief
has three parts. First, the belief has a "cognitive" component
representing a person's knowledge, second, an "affective" component
capable of being aroused in varying degrees of intensity, and finally,
a "behavioral" component directing the belief to action.

With the belief as a unit of analysis Rokeach constructs this
definition of attitude, "An attitude is a relatively enduring
organization of beliefs around an object or situation predisposing

"17

one to respond in some preferential manner. This definition is

representative of others in the literature.18
While an attitude is an organization of beliefs, a belief
system is an organization of attitudes. Rokeach states, "We mean it

(the belief system) to include each and every belief and disbelief of

every sort the person may have built up about the physical and social

 

16M. Rokeach, "The Nature of Attitudes," International
Encyclogedia of Social Sciences (New York: MacMillan, 1966), p. 3.

1711314., p. 12.

18For example see M. L. DeFluer and F. R. Westie, "Attitude
as a Scientific Concept," Social Forces, Oct. 1963, pp. 21-36; H.
Blumer, "Attitudes and the Social Act," Social Problems, 3, 1955,
pp. 59-64; D. T. Campbell, "Social Attitudes and Other Acquired
Behavioral Dispositions," Psychology: A Study of a Science, ed.
Koch (New York: McGraw—Hill, 1963); L. W. Doob, "The Behavior of
Attitudes," Ps cholo ical Review, 1947, pp. 135-156; D. Krech and

R. Crutchfield, Theogy and Problems of Social Psychology (New York:
McGraw-Hill, 1948).

10

"19

universe he lives in. While the belief system is the total framework

man uses to understand his world, the attitude is a subsystem related to
specific objects or situations. There are two types of attitudes:
attitudes-towardesituations include beliefs concerning the event or
ongoing activity in-which the attitude-toward-object is activated, while
attitudes-towardeobjects includes beliefs about peOple, groups, institu—
tions, or issues.20

In analogous fashion, we may wish to define orientations as stable
organizations of beliefs about political objects or situations. This is
what some researchers have done. Consider, the following from an

article by Easton and Dennis:

.Even though the children tested assert a growing aware-
ness of government as an idea and object, are they in fact,
able to distinguish it as a sphere separate from other
areas of social life? If attitudes toward the authorities
as an object are to have relevance for later ties to the
system, we need some evidence indicating that even in
their earliest years children are, in fact, able to
recognize some minimal difference between that which is
governmental and that.which is not. Only under such
conditions could we infer that attitudes towards
government-~to which we shall turn in a moment--refer
to distinctively political bonds.21

These authors.are apparently arguing political beliefs can have
relevance for later political behavior only if the individual dis-

tinguishes between political objects and other social phenomena. An

 

19

M. Rokeach, op. gi£., 1960, p. 35.
20M. Rokeach, 02. 933.,1966, p. 3.
21

D. Easton and J. Dennis, "The Child's Image of Government,"

The Annals (361, Sept. 1965), pp. 40-57 (this quote found on p. 49).

11

alternative to their argument is put forth as a major hypothesis of

The Authoritggipn.PerSonplity. "In short, ideology regarding each social
area must be regarded-as a facet of the total personality and an expres-
sion of more central 'sub-ideological' psychological dispositions."22
In other words there are nor "distinctively" political orientations.

This second position is a working assumption of the present
study and leads to the following formulation of political orientation.
To the extent that, at various times in their lives, objects or situa-
tions repeatedly induce-peop1e to behave in a particular manner it is
appropriate to infer an.orientation predisposing them to behave in this
fashion. Similarly, when "political" objects or situations activate
the same behaviors at different times one may infer political orienta-
tions.

We will now use these.definitions to identify socialization
experiences and psychological attributes which may affect the person's
acceptance or resistance to change. Since such an effort is a part
of a larger discussion of personality develOpment, we will first
examine alternative theoretical approaches to socialization and

personality.

Theoretical Approaches to Personality
Development ppd Change

Both psychoanalytic and cognitive theorists have contributed to

the study of socialization.and personality. According to the former

 

22T. W. Adorno, _e_£. g” _p. cit., p. 207.

12

point of view personality.is formed early and remains relatively stable.
cognitive theorists,-on-the'other hand, emphasize the continual striving
of individuals to adapt-their.beliefs and behaviors to changing conditions.
In other words, changeaain experiences lead to changes in personality.
Greenstein refers to.these two theories by distinguishing between "ego

defensive" and "cognitive" personality development.

Personality formation may be along ego-defensive or
more cognitive lines; the connections between personality
and political belief need to be examined rather than
assumed; both personality and beliefs must be examined
in situations in order to understand behavior; the ways
in which individual predispositions and actions aggregate
and affect the political and social system need to be
explicated. And, to turn the circle, it is the political
and social systems which provide the socializing
environment for 'politically relevant' personal
development and the Situations within which political
action takes pléce.2

A few citations from the literature reflecting each of these two
approaches will help to throw light on this distinction.

The extensive literature flowing from The Authoritariap,Personplity
concentrates on adult-attitudes toward authority as a consequence of
early childhood experiences with authority.24 An inventory of many of
these studies appearsuin an article by I. L. Child entitled "Socializa-
tion" and the entirety of the material cited utilizes psychoanalytic
assumptions about personality.25 A political science textbook by

William Mitchell contains.the following reference to socialization:

 

23F. Greenstein, "Personality and Political Socialization: The
Theories of Authoritarian and Democratic Character," The Annals,
OJ! Cit. ’ p. 95.

24T. W. Adorno, _e_1_:-_. _a_l., _p. cit.

251. L. Child, _p_. cit., p. 655.

l3

" . . . whether a child grows up to emphasize rights and demands on

the government to fulfill his own obligations is surely strongly
influenced, if not actually determined, by early family experinces."26
Froman in an article entitled "Learning Political Attitudes" notes many
psychoanalytically oriented writers (e.g., Erikson, Frenkel—Brunswik,
Christie, and Jahoda) believe critical relationships exist between
child rearing practices and later political behaviors.27 Pye's28
review of Erikson's The YounggMan Luther29 invites political scientists
to examine early experiences as important for understanding later
innovative behaviors. Hagen, in a major study of social change,
concludes that changes result from forces altering family structure,
thereby influencing the child's early develOpment.3O He believes
social change is the summation of individual changes produced by

alteration of family experiences. Change is gradual and requires

the maturation of new generations. Clearly, the influence of

 

26W. Mitchel, The American Polity (Glencoe, Ill.: The Free
Press, 1963), p. 164.

27L. Froman, "Learning Political Attitudes," Western Political
Quarterly (June, 1962), pp. 21-38.

28L. A. Pye, "Personality, Identity, and Political Ideology,"
Political Decision Mpkers, ed. Marvick (New York: The Free Press,
1961).

29E. H. Erikson, Young Man Luther (New York: Norton and Co.,
1958).

30E. Hagen, On the Theory of Social Change (Homewook, Ill.,
Dorsey Press, 1962).

14

psychoanalytic assumptions concerning the early formation and relative
stability of political orientations is pervasive.

Studies emphasizing personality change after childhood have
their roots in the cognitive theories of such theorists as Tolman,
Lewin and Krech. From the standpoint Of these theorists behavior is
purposive and adaptive. Smith, Bruner and White clearly articulate

these assumptions concerning personality development.

The human being, according to this approach, is
not governed by a rational calculus, nor is he a blank
slate on which experience traces its inexorable mark.
Nor yet is man an ingenious machine translating
physical stimuli into bodily responses. Like all
animals, he is an organism, a system Of life processes
that somehow maintains its identity in active interplay
with its environment. An organism is never passive,
but survives and grows through constant striving,
responding selectively to relevant aspects of its
environment, and reaching out to incorporate, modify,
fend off, or attain. Final passivity is death; in
life there is always striving to maintain the delicate
adaptation of needs of the organism to its environment.

The impact of cognitive theory can be found in several studies
of political socialization. Hyman's inventory of political socializa-
tion studies finds parental influences are important determinants of
adult behavior, but there is substantial evidence of "the attenuation
of parental influence" given particular conditions and agencies of
change. He shows, for example, that authoritarianism may develop
independently of parental influences, and notes that certain conditions

such as social and geographic mobility, strict parental control,

 

31B. Smith, gp.Igl., pp. cit., p. 30.

15

adolescent rebellion, aging and generational effects attenuate early
experiences with parents.32

Other studies support this general argument concerning the
effects of later life experiences upon personality. Possibly the
most persuasive among these is that of Almond and Verba. In their

cross cultural study of political orientations they write:

. . . the importance of specific learning of orientations
to politics and of experiences with the political system
has been seriously underdeveloped. Such learning is

not only cognitive in character, but involves political
feelings, expectations, and evaluations that result
largely from political experiences rather than from

the simple projection into political orientations of
basic needs and attitude§3that are the product of
childhood socialization.

Orientations and personality can change, and now we can approach
the problem of identifying prerequisite conditions for change.
Eisenstadt in his study of the effects of age differences on
behavior argues from his data that childhood experiences are likely
to influence adult behavior only in sofar as roles learned in the
family are also useful in achieving social status in later life.
But to the extent that roles acquired in childhood do not serve these
needs, later experiences become important in affecting adult behavior.34
Similarly, Davies theorizes that "political maturation" (i.e., the

change from dependence to autonomy) may be either facilitated or

 

32.]. Hyman, 220 Cite, ppe 69-730

33
G. Almond and S. Verbs, _p, cit., p. 34.

348. N. Eisenstadt, From Generation to Genergtion (Glencoe,
111.: The Free Press, 1956).

16

retarded depending on early family relationships. However, when
established family patterns of need satisfaction are dysfunctional
in more adult roles anxiety results and changes are likely.35

Both of these studies make the point that attitudes and behaviors
which are appropriate in the family situation may not be appropriate
to adult interaction in a more complex environment. Levine finds
the reverse of this phenomenon in non-western culture. His data
show familial experiences are predictive of adult behaviors in
relatively static social situations. But, there is probably less
tendency to generalize family experiences to adult relations in more
differentiated social systems where dissimilarity exists between the
family and community structures.36

All of these studies demonstrate that early family experiences
are most likely to affect adult behaviors when adult behaviors and
expectations are similar to those in the family situation. However,
where early family experiences are inconsistent or less similar to
experiences in the larger community it is less likely that early
learning determines adult behavior. To the extent that later life
experiences are inconsistent with early experiences there is a
greater probability of change in individual orientations and
behaviors. Inconsistency between early and later socialization is a

precondition of individual change of and, conceivably, also of ability

to accept change.

35J. Davies, "The Family's Role in Political Socialization,"
W. op. cit., pp. 10-19.

36M. Levine, "The Role of the Family in Authority Systems,"
w. v. (1960). pp- 20-42.

17

A further group of studies exists which suggest that cultural
crisis or personal trauma are prerequisites of personality change.
Wallace, reviewing work on social movements (e.g. nativistic movements,
cargo cults, etc.), defines "revitalization" as " . . . a deliberate,
organized, conscious, effort to construct a more satisfying culture",
and concludes that it is possible to accomplish a basic transformation
of an entire culture in a relatively short time. Given cultural crisis
conditions and the advent of charismatic leadership in the culture
revitalization can be achieved in a single generation.37 Elkins presents
substantial evidence suggesting that basic changes in adult personality
can occur if individuals are forced to pass through sufficiently
stressful circumstances. As evidence he cites changes that took place
in Negro slaves during shipment from Africa, as well as among inmates
of concentration camps.38 Middleton and Putney studying 1440 college
students find that generalized adolescent rebellion relates poorly
with political rebellion (party preference different from that of the
parent). Child-parent relations that are either overly strict or
neglectful tend to increase the probability of "political rebellion,"

especially when the parents are political activists.39

 

37A. Wallace, "Revitalization Movements," American AnthrOpologist,

38$. Elkins, "Slavery and Personality," Studying Personality
Cross Culturall , ed. Kaplan (Evanston, 111.: Row, Peterson and Co.,
1961).

39R. Middleton and S. Putney, "Political Expression of
Adolescent Rebellion," American Journal of Sociolo , 68 (1963),

18

Several major hypotheses flow from this general review of
personality development and change. Personality is affected by early
family experiences and the impacts of such experiences are relatively
stable and resistant to change. Two of these conditions are the degree
of consistency between early and later socialization experiences, and
the degree of trauma or crisis in the socialization processes. Two
further conditions, age and education, are suggested by the studies
of Anderson and Pressy. Anderson, in a sample of high school students,
finds a decrease in dogmatism between the eighth and twelfth grades.4O
Pressy comparing data obtained on students over a span of three decades
finds that societal changes in moral and religious norms alter the
thinking of college students but not younger high school students.41
Young adults with some education may be more likely to change and
therefore accept change than younger adolescents without education.

The foregoing suggests the rudimentary outlines of a model of
personality develOpment.

We can conceptualize personality as a collection of orientations
resulting from past experience. These orientations predispose individ-
uals to respond in some particular fashion to new and changing situations
which arise as the person matures. Some of these orientations are

formed from the experiences in the family situation. If conditions

 

40C. Anderson, "A Developmental Study of Dogmatism During
Adolescence with Reference to Sex Differences," Journal of Abnormal
and Social Ps cholo , 65 (1962), pp. 132-135.

418. Pressy, "Changes from 1923 to 1943 in the Attitudes of
Public School and University Students," Journal of Psycholo , 21
(1946), pp. 173-188.

19

remain similar to those in early family these orientations will persist
throughout the person's life. If, on the other hand, conditions change
in later adult life we expect the individual to make adjustment by
either changing orientations formed earlier or by develOping new
orientations from new experiences. It is possible that crisis, age

and education may work to intensify or ensure this change process

since such conditions appear to make adjustments and adaptations more
likely. We assume that individuals who have made changes in their own
personality are more likely to accept change generally. We can expand
these notions in the theoretical chapter, but for present purposes the
scheme directs attention to the effects of early childhood experiences
(primary socialization), the effects of later experiences outside of
the immediate family (secondary socialization), and the effects of
personality upon specific orientations--in particular the orientation

toward change.

Primary Socialization gnd Orientptions

From this point on we will refer to early childhood socialization
experiences as primary socialization. And the question must now be
raised, what are the critical primary socialization experiences and
what orientations result from particular conditions?

Easton and Hess in a quasi-longitudinal study of more than
700 midwestern school children conclude that young children develOp
positive and uncritical attitudes toward the political community and
political parties based upon affective ties with the family. They

hypothesize that these predispositions are generalizations of

20

relationships with family authorities.42 Greenstein, interviewing
pre-adolescent school children replicates and confirms these findings.
His subjects have extremely positive conceptions Of political leaders
which they can articulate at as early an age as seven years. He
believes that youthful-orientations formed in primary group relations
are most durable.43 An earlier study by Maccoby, Matthews, and Morton
of high school students and their parents observes the same phenomenon.
Student's candidate preference and.party affiliations are related to
family preferences, with highest conformity occuring in families where
there is only moderate interest in politics.44
From these studies we can infer.childhood eXperience with family
authorities greatly affects the orientation to trust and accept as
legitimate political leaders and institutions. It is not entirely
clear what specific relationships in the family produce these orienta-
tions, but several research studies.have pointed out the critical
significance of affective relationships with family authorities

(presumably parents). Orientations resulting from these relationships

appear to be most stable over time. We can hypothesize as a result of

 

42D. Easton and R. Hess, op. cit., p. 227 ff; R. Hess and D.

Easton, "The Child's Changing Image of the President," Public Opinion
Quarterly, 24 (1960), pp. 632-644.

43F. Greenstein, "The Benevolent Leader: Children's Image of
Political Authority," American Political Science Review (December,
1960), pp. 934-943. ,

44E. Maccoby, R. Matthews and A. Morton, "Youth and Political
Change," Public Qpinion Quarterly, 18 (1954), pp. 23-29.

21

these studies that authority relationships in primary socialization
affect trust and belief in the legitimacy of political objects.

Further evidence of the effect of.early authority relationships
upon adult behavior can be found.inwanxaxticle-by Willis who finds,
in a sample of Swedish parents, that-those who have a strict conception
of their relationship to their children.are more likely to have a
similar conception of their own relationship to civil authorities.4S

A recent article by Pinner helps specify an important
characteristic of authority relationships. He notes that parental
overprotection may result in political orientations of distrust,
" . . . where the transition from the intimacy of the home into the
greater world is experienced as a difficult and fear arousing step,
and where there is little experience in the solution of problems
through direct communication."46 Here again trust of politics and
politicians appears to be affected by relationships with parental
authority. But more specifically, as Pinner's study indicates,
communication in problem solving situations may be a critical
attribute of authority relationships.

The following hypothesis seems a plausible inference from this
review of research on the effect of primary socialization upon

political orientations. Political trust results from early authority

 

45R. H. Willis, "Political and Child—Rearing Attributes in

Sweden," opppgl pf Abnopmpl gng Spgigl Ppychology, 53 (1956),
pp. 531-535.

46F. A. Pinner, "Parental Overprotection and Political
Distrust," The Annals, yp. cit., pp. 58-70 (this quote found on p. 46).

22

relationships characterized by more extensive communications between
the child and his authorities.

We now turn to the identification of the effects of secondary
socialization (i.e., socialization occurring later in life and

outside of the immediate family) upon particular orientations.
Secondapy Socialization and Orientations

Perhaps the most extensive empirical study in political
socialization is The Civic Culture by Almond and Verba.47 Their
data are based upon respondent recall by subjects from national
samples in five countries. An extensive summary of their work is
beyond the scOpe of this review; we can describe only their findings
concerning the effects of secondary socialization on political
orientations.

The authors examine the relationship between authority experi-
ences and sense of political efficacy. They present data intended
to show that adult experiences with authorities proximate to political
roles may influence political orientations more than do earlier family
experiences. In their data, positive relationships between individual
participation in family discussions and a sense of political efficacy
exists only among subjects with little education. Possibly family
experiences among the more educated have relatively little affect
upon adult political competence. Almond and Verba summarize this

relationship between family influences and political efficacy as follows:

 

47G. Almond and S. Verba, _p, cit.

23

To sum up the relationship between family partici-
pation and sense of political competence: there is

some connection between democracy in the family and
democratic behavior in politics. Those who had the
opportunity to participate in the family are somewhat
more likely than others to feel politically competent.
But the relationship is a complex one, in which a
number of intervening variables can blunt the impact

of this democratic training on political attitudes.
Family participation is not generalized to political
participation among those with higher education, and

in several countries this generalization is not made

by our younger respondents. The reason may be that
respondents whose political behavior is not affected by
their family training do not need that training as a
means of induction into a feeling of political competence.
Younger respondents and those with higher education are
subjected to numerous pressures leading to a sense of
political competence, and the marginal effect of family
participation is, under these circumstances, not very
great.4

This summary supports the thesis that later life experiences

may have more impact upont the sense of efficacy than early family

influences at least for younger individuals and those with more

education.

The study goes on to consider the impact of participation in

discussion and decision making in school and job situations. Con-

siderable variation exists in the extent of influence of school

experiences on efficacy (which probably indicates that schools in the

five countries vary a great deal). The authors conclude that partici-

pation in school affairs has only marginal importance unless it is

followed by experiences of effectiveness in post-school situations.

When they investigate the effects of participation and discussion in

decisions on the job they find the strongest evidence of the study.

 

481b1d., p. 352.

24

People who report being consulted about decisions on the job alsO
report a high sense of political and personal efficacy, and this
relationship is independent of extent of education. Almond and Verba
conclude that participation in decision making at the work place, in
school, and in the early family are, in that order, important to the
develOpment of a sense of efficacy. Thus, communication with authorities
in situations outside of the immediate family may greatly modify the
early family influences. The authors account for the importance of
these later experiences by noting they are more similar and more
proximate to adult experiences in the political system. This suggests
that generalization of the effects of early family experiences decreases
with the amount of specialization in society. Theorists who believe
early experiences create needs that are generalized or projected onto
adult interactions may be overlooking the learning principle of
discrimination while paying excessive attention to the principle of
generalization. Possibly people find experiences with political
authorities so entirely different from childhood experiences they are
not able to generalize from early childhood to adult situations. As
they accumulate experiences of effectiveness in their adult activities
they may come to discriminate adult relationships with authority from
childhood experiences and change their attitudes accordingly.

For our purposes the Almond and Verba study directs attention
to the importance of secondary socialization for the development of
a sense of political and personal efficacy. The critical variable
appears to be that of authority relationships in school and on the

job. Is it possible to specify this relationship further? In the

25

Almond and Verba study communication with authorities emerges as the
major form of participation in decision making. Just as communication
with early authorities is the critical primary socialization experience
affecting trust, communication with authorities outside of the family
appears to affect efficacy.

Communication with authorities in primary and secondary
socialization is an important independent variable in the development
of political trust and efficacy. We still have not established the
existence of a relationship between either trust or efficacy and the
person's ability to adjust to change. It is possible to speculate
about such a relationship, but leaving the question open for further
investigation seems a more prudent strategy.

The literature on liberal and conservative ideologies leaves
little doubt that acceptance and resistance to change is one of the
chief components of these ideologies. A review of this literature
will help to identify correlates of the orientation toward change.
Most of the research to be considered describes liberal and conserva-
tive orientations in the context of more general investigations of

personality syndromes.

Personality and Orientations

What personality attributes are associated with acceptance or
resistance to change? MUch of the literature indicates an individual's
orientation toward change is a central element of liberal and conserva-

tive thought.

26

The Authoritarian Personality49 is the benchmark from which to
begin a review of research on liberalism-conservatism. This work

describes the "prototypic liberal and conservative" orientations in

the following way.

The prototypic 'liberal' is, according to our guiding
conception, an individual who actively seeks progressive
social Change, who can be militantly critical (though
not necessarily totally rejective) of the present status
quo who Opposes or de-emphasizes numerous conservative
values and beliefs regarding business, success, rugged
individualism, human nature, and the like, and who would
diminish the power of business by increasing the pgger
of labor and the economic functions of government.

.The prototypic 'conservative' in terms of the present
scale, is one who supports the status quo and resists
changes in existing politico-economic power arrangements,
who supports conservative values and traditions, who
believes that labor is properly subordinate to employer
or management, and who wishes to minimize the economic
functions of government in order that individual business-

men can, in free and equal competition provide goods of
maximum quality at minimum cost to the consumer.5

In still another early research effort Eysenck factor analyzes
responses of several political groups and detects two major factors
which account for most of the variance. He identifies these factors
as radicalismrconservatism and tough mindedness-tender mindedness.

As we look more closely at the responses included in the radicalism
factor it is clear that the radical is mere receptive to changes in
law and policy over a wide diversity of issues. Eysenck describes the

poles on his "radicalism" factor in the following way.

 

49T. W. Adorno, 3;. 91., pp. cit.
5°Ib1d., p. 176.
51

Ibid., p. 177.

27

On the one hand, we find a belief that private
property should be abolished, that the death penalty
ought to go, that Sunday observance is old fashioned,
that Jews are valuable citizens, that the divorce laws
ought to be altered, that we should give up part of our
sovereignty, that we should abolish abortion laws, that
we should cure criminals rather than punish them, that
laws favor the rich, that companionate marriages should
be allowed, and that patriotism is a force that works
against peace. On the other hand we have a belief that
nationalization is inefficient, that compulsory religious
education is desirable, that the Japanese are cruel by
nature, that we should go back to religion, that Jews
are too powerful in this country, that flogging should
be retained as a deterrent, that war is inherent in
human nature, that conscientious objectors are traitors,
that birth control should be mgge illegal and that
coloured peoples are inferior.

Eysenck does not isolate any personality characteristics which
might account for radical and conservative attitudes since one of his

major hypotheses is, "

. . . there is in truth only one ideological
factor present in the attitude field, namely that of Radicalism-
Conservatism. The T-factor itself does not constitute an alternative
ideological system but is rather the projection onto the social attitude
field of a set of personality variables."53 Tough mindedness correlates
with extraversion, aggression, dominance, rigidity, intolerance of
ambiguity, narrowdmindedness, and mental correctness.

Both the Adorno,.gp,,pl. and the Eysenck studies indicate that
liberal orientations are associated with acceptance of change while
conservatives tend to resist change. This calls for a more careful
examination of other elements of conservative and liberal views, and

52H. J. Eysenck, The Psychology of Politics (London: Routledge,

Kegan Paul, Ltd., 1954), p. 127.

53Ib1d., p. 170.

28

for an ivestigation of social-psychological correlates of such
views.
McClosky identifies major themes of conservatism as well as

some of the correlates of conservative personalities. He lists the

following value positions as basis of the conservative outlook.5§

Man is a creature of appetite and will, 'governed
more by emotion than by reason' (Kirk), in whome
'wickedness, unreason, and the urge to violence lurk
always behind the curtain of civilized behavior'
(Rossiter). He is a fallen creature, doomed to
imperfection, and inclined to license and anarchy.

Society is ruled by 'divine intent' (Kirk) and
made legitimate by Providence and prescription.
Religion 'is the foundation of civil society' (Huntington)
and is man's ultimate defense against his own evil
impulses.

Society is organic, plural, inordinately complex,
and the product of a long and painful evolution,
embodying the accumulated wisdom of previous historical
ages. There is a presumption in favor of whatever has
survived the ordeal of history, and of any institution
that has been tried and found to work.

Man's traditional inheritance is rich, grand,
endlessly proliferated, and mysterious, deserving of
veneration,.and not to be cast away lightly in favor
of the narrow uniformity preached by sophisters and
calculators' (Burke). Theory is to be distrusted since
reason, which gives rise to theory, is a deceptive,
shallow and limited instrument.

Change must therefore be resisted and the
injunction heeded that, 'Unless it is necessary to
change it is necessary not to change' (Hearnshaw).
Innovation 'is a devouring conflagration more often
than it is a torch of progress' (Kirk).

Men are naturally unequal, and society requires
'orders and classes' for the good of all. All efforts

 

54H. McClosky, "Conservatism and Personality," The American
Political Science Review, 52 (1958) pp. 27-45.

29

at levelling are futile and lead to despair (Kirk and
Rossiter), for they violate the natural hierarchy and
frustrate man's longing for leadership. The superior
classes must be allowed to differentiate themselves
and to have a hand in the direction of the state,
balancing the numerical superiority of the inferior
classes.

Order, authority, and community are the primary
defenses against the impulse to violence and anarchy.
The superiority of duties over rights and the need
to strengthen the stabilizing institutions of society
especially the church, the family, and, above all,
private property.

The resistance to change is clearly a part of the conservative
value system, but another theme emerges--the reliance upon authority,
order, and law over individual reasoning. This dependence upon
authority may well be another important variable associated with
individual acceptance or resistance to change.

The McClosky study also investigates the relationships among
variables such as intelligence, social-psychological attributes, and
clinical-personality variables and conservatism. The conservative is
less educated, less socially aware and less intellectually oriented
than the liberal. The conservative tends to be submissive, anomic,
alienated, pessimistic, less socially responsible, less self confident,
and more guilty. The findings concerning the clinical-personality

variables and conservatism are summarized as follows:

. . . the extreme conservatives are easily the most
hostile and suspicious, the most rigid and compulsive,
the quickest to condemn others for imperfections or
weaknesses, the most intolerant, the most easily moved

 

551b1d., p. 35.

30

to scorn and disappointment in Others, the most
inflexible and unyielding in their perceptions and
judgements of others.5

McClosky's study contains many variables which might prove
useful in this study; certain ones appear to relate to the previous
discussion of socialization experiences and personality. Earlier we
noted that trust is affected by early authority experiences while
efficacy is affected by experiences with authority outside of the
immediate family. Several attributes in the McClosky study (e.g.,
alienation, pessimism, and "scorn and disappointment in others")
represent characteristics which reasonably mdght be associated with
trust, while others (particularly self confidence) might be expected
to be associated with personal efficacy. It may well be the variables
of trust and efficacy are not only related to socialization experiences
but are also directly related to liberalismrconservatism and in turn
may relate to the orientation toward change. Other variables in the
McClosky study invite further investigation.

Such personality variables as hostility, rigidity and intolerance
suggest an underlying attribute such as authoritarianism or dogmatism
may distinguish liberals from conservatives. Since Rokeach's variable
of dogmatism is developed from attempts to find a single dimension
underlying both authoritarianism and intolerance, it appears useful
to investigate this variable in relation to the acceptance or resistance

to change.

 

56Ibid., p. 41.

31

Rokeach argues that the basic characteristic of a person's

"belief system" affecting the extent of dogmatism is

. . . the extent to which the person can receive,
evaluate, and act on relevant information received
from the outside on its own intrinsic merits,
unencumbered by irrelevant factors in the situation 7
arising from within the person or from the outside.

According to Rokeach messages contain two kinds of information:

(1) information about the subject matter and (2) information about
the source of the message. He continues, " . . . we see that there
are two aspects to the communication, and in different people the
dual aspects will be differentiated or fused together according to

cl."58 The less

the degree to which their systems are open or close
dogmatic person (Open belief system) receives, evaluates and acts

on the basis Of the content of the events and messages in a situation
because he is able to distinguish between the content and the source
of the message. To the contrary, the more dogmatic person (closed
belief system) is unable to make such a distinction and is, therefore,
guided in his interpretations by a fusion between subject matter and
the source of the subject matter. Simply stated, dogmatic individuals
are likely to react to the authority rather than to what the authority

says. Rokeach (1960) makes explicit the importance of dogmatism for

the present study when he states,

Reliance on authority, yielding, conformance,
and resistance to acculturation all may have a common
cognitive basis, namely, the ability (or inability) to

 

S7M. Rokeach, pp, cit., 1960, p. 57.

581b1d., p. 60.

32

discriminate substantive information from information
about the source, and to assess the two separately.59

Aside from suggesting the theoretical importance of dogmatism for our
present analysis the above quote also includes a theme appearing in
several studies in this review--the reliance upon authority. This
suggests that a person's dependence upon authority may be a critical
personality attribute in the analysis of the orientation toward change.
The following chapter presents a theoretical schema which includes this
theme of reliance upon authority as a central dimension tying together
the impacts of socialization and personality upon the acceptance or
resistance to change. Before turning to this part of the presentation

I will summarize the variables derived from this literature review.

Summapy

At the outset of the literature reveiw we defined socialization
broadly as experiences affecting the development of individual person-
ality. We discussed the concept of orientation as a predisposition to
particular behaviors inferred when objects or situations activate
peOple to behave in similar ways in different situations and at different
times in their lives. We then examined theoretical approaches to the
general problem of personality development and change. The major
impact of psychoanalytic theory has been to focus attention upon the
effects of early childhood experiences upon later adult behaviors.

Such assumptions have led to many studies emphasizing the early forma-

tion and relative stability of orientations and personality. An

 

59Ib1d., p. 60.

33

alternative approach, cOgnitive theory, has directed attention to the
preconditions of personality change. In this view personality is
adaptive to later life experiences. We attempted to identify several
conditions associated with such changes. One group of studies suggests
inconsistencies between early experiences in the family and later
experiences outside of the immediate family are likely to produce
adjustment in personality. Another series of efforts were introduced
to point out the importance of crisis or personal trauma as precondi-
tions of change. We can summarize these studies with the general
hypothesis that the greater the extent to inconsistency and discontinuity
(more crisis events) in socialization the greater the likelihood of
personality change. I further hypothesize that people who have
experienced changes in their own personality are more likely to accept
changes in the external world. Still other studies found age and level
Of formal education to be related to individual acceptance Of change.
In this study, we have an attempt to identify the effects of primary
socialization, secondary socialization, and personality attributes

upon the orientation toward change.

The review of the effects of primary socialization upon ogientation
OngPHSP (political and general) results from early authority relation—
ships characterized by more extensive communication between the child
and his authorities. The review of studies concerning the effects to
secondary socialization (experiences outside of the immediate family)
found personal efficacy to be a result of communication with authorities
and participation in decisions affecting the individual. Finally we
discussed the literature on personality attributes related to particular

orientations.

:Is

34

Since liberal and conservative political orientations have been
found to be associated with the acceptance or resistance to change,
much of the material reviewed dealt with correlates of conservative
and liberal attitudes and personalities. The orientations of trust
and efficacy were expected to be related to more liberal political
views. Further, the studies suggest the variables of dogmatism and
reliance upon authority are central to our analysis of the orientation

toward change.

CHAPTER II

THEORETICAL CONCERNS IN THE ANALYSIS OF THE
ORIENTATION TOWARD CHANGE

Personality, Socialization and the
Orientation Toward Change

I have conceptualized the orientation toward change as a relatively
enduring organization of beliefs predisposing the person to accept or
resist change. This orientation is a general personality attribute
affecting individual response across situations (e.g., social, political,
economic, public, private, etc.) and through time in similar situations.
I view the orientation toward change, as any orientation, to be a part
of an interrelated set of predispositions composing the individual's
personality. The set of orientations acts to predispose behavior of
peOple as they confront new and changing conditions. With this
conceptualization it is possible to view the orientation toward change
as it is affected by personality attributes and socialization experiences
acting upon the entire personality. This simple schema permits the
classification Of variables identified in the literature as relevant to
the present study. The purpose here is to clarify the schema as a way
of organizing and relating major variables, and to prOpose several
hypotheses to guide the data analysis.

We can begin by organizing variables as either personality

attributes or socialization experiences. We have identified the
35

36

following variables in the literature as socialization experiences:
(1) extent of communication with authorities in the early childhood-
family situation; (2) extent of communication with authorities in
relationships established at the present time in the person's life;
(3) the extent of consistency (i.e., similarity) between the experiences
in the early family and experiences in the broader community of adult
interaction; (4) extent of continuity (i.e., traumatic or crisis events)
in the socialization process; (5) age; and (6) extent of formal education.
The balance of variables critical to the present investigation can
be grouped as personality attributes. These variables appear to result
from socialization and to predispose the individual to particular
responses in varying situations. The personality attributes include
both structural and substantive properties. Rokeach's work on dogmatism
is a specific attempt to understand the structure rather than the content
of personality. He describes his effort in the following way: "The
discrepancy we may note between what is said and the way it is said is
a discrepancy between content and structure. Our theoretical task, then,
is to formulate the formal and structural properties of belief systems
apart from specific content, and in such a way that they can be measured."1
Reliance upon authority is a second attribute which I believe to
be a part of personality structure. As dogmatism describes a person's
ability to differentiate source and message and act on either a fusion

or separation of the two, reliance upon authority describes the extent

to which the individual accepts opinions of others. For theoretical

1M. Rokeach, pp. cit., 1960, p. 15.

37

purposes I am using reliance upon authority in a very broad sense.
Authority can be represented by people, established ways of doing things,
law, social norms and expectations or any one of a variety of sources
external to the situation and providing cues about apprOpriate behaviors
for particular situations. Given this broad definition it is apparent
that all peOple more or less rely upon authority in any situation.
This is the attribute I want to investigate--the extent to which
individuals rely upon external cues in determining their behavior
rather than issues and activities in the situation itself. For the
purposes of this study I have Operationalized reliance upon authority
in much narrower terms. It is defined as the extent to which an
individual agrees with the Opinions of his authorities (people influential
in his life) in the secondary environment. That is, I want to investigate,
the extent to which people agree with the Opinions of their own authori-
ties On a wide variety of issues and matters of mutual concern. Those
subjects who agree more are considered more dependent upon authority.
With this description Of dogmatism and reliance upon authority
clarifying the structural attributes of personality, we can further
examine the attributes Of personality content in relation to the
orientation toward change. Several studies suggest what people believe
about particular social Objects and situations may influence their
beliefs about change. The orientation toward change appears to be a
major criterion distinguishing liberal from conservative views Of

politics.2 Liberals are more likely to accept change in social and

 

2Adorno, _e_p. a_l., pp. cit., 1950; Eysenck, pp. ci

t., 1954;
Huntington, _p, cit., 1961; and McClosky, _p, cit., 1958.

38

political life. If liberalism is related to the acceptance of change
then we might expect to find other correlates of liberalism to affect
the orientation toward change. For instance, McClosky finds the
conservative to be more alienated, less self confident, and more
likely to scorn Others in their interpersonal relations. This suggests
the variables of personal efficacy and trust may be important correlates
of acceptance of change.3 These three variables are part of the
personality content which is to be investigated in the present study.
With the forgoing clarification we can now attempt to bring
together these theoretical notions and specify specific hypotheses

reflecting both the theory and the findings in the literature.
Theoretical Sketch

Before attempting to state hypotheses it may be useful to present
an intuitive theoretical sketch of how these variables interrelate.
Perhaps this sketch can make more realistic the skeletal framework
uncovered to this point.

The first socialization variable, extent of communication with
authorities in the early family situation, draws attention to the
possibility a person may have either Open, communicative, interpersonal
relations with early authorities or a closed, non-communicative relation-
ship. Both types of early experiences surely have profound effects
upon the child. If early experiences are such that the child finds

authorities to be accessible and communicative he learns to respond to

 

3McClosky, pp, cit., 1958.

39

authority with more information and understanding of both the authority
as a person and of what the authority says. Such interpersonal contact
generates less reliance upon the authority, app authority, and more
dependence upon what the authority, as a knowable and therefore more
real person, communicates about himself and the world. The growing
dependence upon what is communicated rather than upon who communicates
contributes to the child's ability to distinguish between authority and
what the authority says (this ability to distinguish message from source
of message is the essential characteristic of the less dogmatic person).
These more open and communicative experiences may lead to early formed
orientation to trust others generally since those closest to the child
have proven trustworthy.

On the other hand, if the early authority experiences involve
but little interaction or communication we can reasonably expect the
child to learn different ways of relating to authority. Without communi-
cative, interpersonal relations the child responds to an Object which he
knows little about but upon which he depends greatly. He is taught to
respond to arbitrary dictates of authority by blind Obedience rather than
by persuasion and information. His identification with and dependence
upon the authorities, in a vacuum of communicated understanding, leads
to a growing dependence upon authority generally and an inability to
differentiate between the authority as an object (rather than an under-
standable person) and what the authority communicates. In other words,
since the child does not have information or understanding of the
authorities, he can respond to authority only with obedience or diso-

bedience. Without communication, the child can not learn to respond

40

to authority as a source of information about people and events in his
world. Such arbitrary and unpredictable experiences may engender distrust
toward other people generally.

As the child moves beyond the immediate family experience into
relationships with other authorities (the second socialization variablefiiflx
extent of communication with authorities at the present time in tfigjfl
subject's life) he may find his early experiences, either communicative
or non-communicative, reinforced. On the other hand, he may find his
early experiences are quite different from those of later relationships.
In the latter case we expect an adjustment in personality to conform to
the requirements arising from the new experiences. Logically this
transition from early to later experiences with authority can develop
in any one of four possible patterns: (1) both early and later
experiences with authority are communicative and open; (2) both early
and later experiences with authority are non-communicative and closed;
(3) early experiences are communicative but later experiences are non-
communicative; and finally (4) early experiences are non-communicative
but later experiences are Open and communicative. Each of these four
possible patterns of socialization can be expected to produce different
personality attributes and different orientations toward change.

At this point we can introduce the third major variable--extent
of consistency between early and later experiences. The greater the
inconsistency or dissimilarity between early and later experiences
the more likely a person will be to make adjustments in his personality.
In the first two patterns, with greater consistency between early and

later experiences (either both communicative or both non-communicative),

41

we expect little change in personality. That is, the early formed
personality is reinforced and supplemented but not changed. In the
first socialization pattern (both early and.1ater experiences with.
authority are communicative) we expect personality to be less dependent
upon authority as an Object, to be able to differentiate between the
authority and what the authority communicates, and to trust other
peOple. Further we can expect these later communicative relations with
authority to give the individual a growing sense of personal efficacy
in his interactions outside the immediate family. Our hypothetical
individual, with consistently communicative authority relations, is
less reliant upon authority, trusting of other people, able to dif-
ferentiate the authority from what the authority says (i.e., less
dogmatic) and likely to have a stronger sense Of personal efficacy.

We can further speculate that such a person would be likely to have-
liberal political views because authority (e.g., strong central
government) is not a thing to be feared. And because the person is
less reliant upon the authority of custom, law and usual ways of doing
things, he is more able to accept change in varying situations.

Now let us construct the second hypothetical socialization pattern
where early and-later experiences with authority are consistently non-
communicative or closed. In the case of consistent, non-communicative
relations we expect orientations which are formed early to remain
relatively constant. But we expect a different set Of attributes to
develop. If the child grows up in non-communicative relations he is
more likely to be dependent upon authority as an Object rather than a

source of information and understanding, less able to differentiate the

42

authority from.what the authority communicates (i.e., more dogmatic),
and, because of early arbitrary experiences with people, less trusting
of others generally. The lack of communication in later experiences
may also produce a sense that he can not control or affect his new
environment of interaction any better than he could in the early
family situation. This leads to a lesser sense of personal efficacy.
The person with consistently closed (non-communicative) experiences

is more reliant upon authority as an Object, more likely to fuse
authority with messages from the authority (more dogmatic), less
trusting of other people, and less personally efficacous. This
person may well adopt more conservative political attitudes since he
has reason to fear strong and potentially arbitrary authority (e.g.,
strong central government). And, because he relies upon the authority
of tradition, law and usual patterns of behavior, he is less able to
accept change in various life situations. There are two further
socialization patterns (3 and 4) which involve greater inconsistency
or dissimilarity between early and later authority relationships.

The individual may move from open, communicative relations in
the family to closed and non-communicative relations in later life,
or to the contrary, from non-communicative relations in early life
to Open, communicative ones in later life. In both cases the incon-
sistency in socialization may produce adjustments and change in
personality. We hypothesize that individuals making such adjustments
in their own personality are more likely to accept change in the
external world as well. If we expect changes we must specify what

change seems likely. In both cases the changes in orientations are

43

likely to be contrary to the ones formed early in life. We can

discuss this more easily by continuing to develop personality profiles
resulting from particular socialization patterns. First, the socializa-
tion pattern involving non-communicative early experiences and
communicative adolescent and young adult relations.

As in the second socialization pattern we expect the early,
non—communicative relations to produce more reliance upon authority,
less ability to differentiate source and message (more dogmatic) and
less trust in other peOple. But because of the adjustments required
by the transition from the closed family situation to Open and come
municative authority relations later in life we expect an attenuation
of early effects. The individual becomes less reliant upon authority,
less dogmatic, more trusting Of others, and in addition, develops a
sense of efficacy from his communicative experiences in later life.

We seem to have constructed an adjusted hypothetical personality
very similar to the one developing from the first socialization
pattern, and we therefore, expect results similar to the ones found
in the first process although they may be less pronounced than the
attributes produced in consistently open relations. This person is
also expected to be more liberal about politics and more likely to
accept change for the reasons discussed previously. Again, these
orientations may be less intense because of the adjustments made in
the personality.

Finally, we consider the fourth socialization pattern--early
authority relations are communicative but later experiences are closed

and non-communicative. The early communicative experiences are expected

44

to produce less reliance upon authority, less dogmatism, and more.
trust of people. In this case the adjustments required by the
transition to a closed environment of young adult relations with
authority diminish the effects of early experiences and lead to the
following changes in personality. The individual becomes more reliant
upon authority, more dogmatic, and less trusting. He_develops.less

of a sense of personal efficacy because of unsuccessful communication
and participation in later life situations. The emerging profile is
similar to the one produced by the second process. And again we
expect a similar set of attributes to ones produced by the second
process but they should be less intense because of the transition.
This person is expected to be conservative and less accepting of
change, and again, for reasons similar to ones argued in the discussion
of the second socialization process.

We have develOped a theoretical sketch of personality attributes
emerging from four different types of socialization patterns. In this
effort we have specified relationships among the major variables with
the exception of age, extent of education, and extent of personally
felt crisis (degree of continuity). We are led to expect discontinuity
to contribute to a more accepting orientation toward change since
discontinuity should relate to personality change which in turn is
expected to relate to acceptance of change in the external world.

Age is important in that it is usually not until late adolescence
that the individual has experiences beyond those of the immediate
family and can, at least potentially, experience inconsistency in

socialization. The last variable, education, may serve to extend the

45

effects of inconsistency between early and later life experiences,
and to facilitate individual abilities to communicate with authority
generally.

This completes the theoretical sketch to be used in the present
study. The sketch has been developed to organize and relate both the
variables and findings identified in the literature review. It would
be unrealistic indeed to expect such a theoretical view to describe
accurately the complex reality of personality development and change.
It is an attempt to capture the process of human development by
constructing a more dynamic model relating socialization and person-
ality. It would be even more unrealistic to expect the present study
to produce data supporting all the details considered here. A more
practical use of the schema is to organize the analysis and to generate
hypotheses. The testing of these hypotheses may in turn produce a
reformulated framework closer to reality and further from this

speculation.

Statement of Hypotheses

Hypothesis l:p/The greater the degree of dogmatism the greater the
degree of resistance to change.

The second structural attribute, reliance upon authority, is
also expected to relate to the acceptance of change. In this case
a person who is more dependent upon authority generally (e.g., custom,
usual ways of doing things, influential others, etc.) might be expected
to be less likely to accept changing situations which require departures

from normal or usual patterns of thought and action. A second hypothesis

is:

46

Hypothesis 2: The greater the reliance upon authority the greater the
resistance to change.

We can now develop some expectations concerning the content
variables (i.e., trust, efficacy, and liberalism) in relation to the
orientation toward change. The variables of trust, efficacy and
liberalism appear to be related as evidenced by McClosky's study (1958)
of conservatism. Basic elements of the conservative ideology are the
resistance to change, dependence upon authority, and rigidity. All
of this suggests a set of orientations likely to be associated with the
orientation toward change in the following manner:

Hypothesis 3:J’Greater personal efficacy, more trust of other people,
and liberal political views are more likely to be associated with the
acceptance of change.

If dogmatism and reliance upon authority relate to the orientation
toward change, and trust, efficacy, and liberalism related to the same
orientation then these variables should relate to each other. To test
this speculation we might generally hypothesize:

Hypothesis 4: High dogmatism, dependence upon authority, low trust,
less efficacy and lcw liberalism relate positively to resistance to
change.

We can now move on to the second question concerning the effects
of socialization experiences upon the orientation toward change. Five
attributes of the socialization which we think are related to the
acceptance of change have been identified: (1) extent of communication
with authorities in the primary environment; (2) extent of communication

with authorities in the secondary environment; (3) the degree of

47

consistency between the communication patterns in the primary and
secondary environments; (4) the extent of discontinuity in socializa-
tion experiences and (5) level Of formal education achieved.

The review of literature can be interpreted as demonstrating two
alternative hypotheses concerning the relationship of socialization
and personality. The first hypothesis, from psychoanalytic assumptions,
posits that the type of individual experiences (e.g., disciplinary or
permissive, communicative or non-communicative, etc.) determines
orientations composing personality. This places primary emphasis upon
experiences occurring early in childhood. A second view of the relation-
ship between socialization and personality places more emphasis upon
experiences in later life (outside the immediate family) and argues that
personality is affected by the joint operation of early and later life
experiences. That is, the greater the dissimilarity between experiences
in early and later life the more likely the effects of early family
influences will be altered or changed.

The present theoretical schema can be used to partially test
these rather broad hypotheses in the context of the analysis of
acceptance of change. With this intent I have constructed the following
hypotheses:
Hypothesis 5: If the type of socialization experience has more
influence upon personality than the variation in early and later
experiences then we should find the greater differences in the
orientations toward change occurring between consistently open and

consistently closed socialization patterns.

48

Hypothesis 6: If instead, dissimilarity between early and later
experiences has more effect upon personality than type of experience
then we expect to find the greater differences in the orientations
toward change occurring between the consistent (i.e., both environmental
open or closed) and the inconsistent (one environment open and the other
closed) socialization patterns.

There is some literature suggesting communicative experiences
with authorities in the family and later associations are important
determinates of the trust of other people and sense of efficacy,
respectively. Two hypotheses seem appropriate.

Hypothesis 7: A more Open primary environment should relate to more
trust of other peOple.

Hypothesis 8: A more Open secondary environment should relate to a
higher sense of personal efficacy.

Since efficacy and trust are expected to be correlates of the
acceptance of change we can further hypothesize:

Hypothesis 9: Groups from more open primary and secondary environments
should be more accepting of change than those from more closed primary
and-secondary environments.

Still other studies suggest crisis events and formal education
relate to personality change and the acceptance of changing social
values, respectively. We can investigate the following hypotheses in
the present analysis.

Hypothesis 10:p/The greater the degree of discontinuity the greater

the degree of acceptance of change.

49

L ,I

Hypothesis ll:x The more education (years of high school and college)
the greater the acceptance of change.

We have tried to consider in the foregoing hypotheses how some
of the personality attributes and socialization experiences might affect
the orientation toward change. The third question concerns how
socialization patterns, personality-attributes and the orientation
toward change may interrelate. For this purpose I have tried to
formalize the above theoretical sketch in a table (Table l) specifying
interrelationships among the major variables.

This completes the presentation of hypotheses to be used in
guiding and organizing the analysis, and we can.now appropriately

discuss the study design for the investigation.

50

 

owemnu mo moon

ownenu on some

omcmno umamom .o lueoooo ououoooz .o lumfimou ououovoz .o omnono unooo<
amHHouonaH amHHouonHH

Emfiamuonwa 30A .m swan oumuooox .m 30H ououoooz .m Emfiamuonaa swam

kamOfimmo housemmo Hmoomuon hoooawmo Hooomuoe hooOmeo

Hmoomuom 304 .q swan unsuspoz .q 30H ououomoz .q Homemuoe swam
umnuu umouu

umnuu 304 .m swan OuOHOpOz .m 30H Ououovoz .m umnuu swam

mufiuonusm %ufiuozunm mo mono hufiuoaunm do mono zufiuoaunm

no unooaoeon .N Iveoeopna oumuoooz .N twosome oumuomoz .N mo monopooeovaH
amfiumawov Beaumamop

Beaumawop swam .H. 30H ha0uouoooz .H swan maououovoz amfiumawov sou

.H

 

Amoooumamdoov AmonoumfimeouaHv

oomoau .u«>nm some .ua>nm

.Oom muoooooom .oom huovnouom
oomoau nauseouu>om
GOHOONfiHofioom %uoaaum

AmonoumfimooooHv
.vomoao .uH>dm ammo .Ha>nm
.oom auoonooom .oom muooeooom
some unoaaoua>om
downwawaofloom huosfium

Ahoaoumfimaoov

 

mononHuuuo hufiaoeomuom wouoaou one monouuom nowuoufiaowoomll.a mqm<a

CHAPTER III

STUDY DESIGN

Subjects

This research was carried out with the cooperation of young
adults who participated in a three-week political training program
held during the summers over a four-year period. Each of the four
programs involved some one hundred high school graduates from the
state of Michigan. The students were selected by their teachers to
attend these training sessions on the operation of state and local
governments. The explicit purpose of the program was to recruit and
encourage capable young people to become active in political affairs.
To be selected the students had to demonstrate their potential for
such involvement by their interests and activities in high school.
These young peOple, judging from my experiences as an advisor to each
Of the sessions, are politically active, articulate, and intensely
aware of their world and its problems.

There were several reasons, aside from their accessibility and
awareness, for using these groups as subjects for research. The
theoretical view adopted in the present study suggests the period of
late adolescence, when the person emerges from the immediate family,
to be a crucial time of personality adjustment--a time for the

individual to entertain new ideas, experience new relationships, and

51

52

change old and long held beliefs. Further, the research in socialization
and political attitudes to date has involved for the most part either

the very young or older adults. The present study provides some insight
into an important and as yet inadequately examined link in the develop-
ment of political orientations.

I directly administered questionnaires to the subjects in the
most recent program and mailed questionnaires to those previously
involved. The returns were unusually high for such voluntary research
groups. Over 70% of the directly administered interviews and nearly
65% of the mailed questionnaires are included in the analysis. As a
result the study is based upon the completed interviews from 241
respondents. Additionally, I gave a post-test questionnaire to the
most recent program participants to measure the stability of instru-
mentation. All of the Likert type items were post-tested as an analysis
of test-retest reliability. The period between tests was approximately
three weeks. Before continuing, it may be useful for the reader to be
aware of some population characteristics in order to make explicit any
biases entrenched in the study.

The subjects were 49% males and 51% females. They ranged in
ages from seventeen to twenty-one years. The distribution by age was:
13% seventeen years; 252 eighteen years; 27% nineteen years; 24% twenty
years; and 11% twenty-one years. These young peOple are, for the mbst
part, from.small towns (82% are from towns under 10,000); of the
protestant faith (26% are Catholics and.6z list no religion); from
large universities (about a fourth of those in college are from schools

of less than 10,000 students); and from affluent families (more than

53

half have family incomes in excess of $7,500). Almost a third of the
total group had not entered college at the time of the study. The
balance is divided in the following way on the variable of years of
college: 26% had one year of college; 24% two years; and 17% three
years.
Though not a cross-section of college students this group of

young people seems a suitable source of data on the research problem.
But since no sampling procedures have been followed the conclusions

are necessarily limited to the present group.

Instrumentation

The questionnaire for this research is composed primarily of
structured instruments supplemented by open ended items. A ten page
questionnaire has been developed and protested upon twenty-five
undergraduates in an introductory political science course. The
final form described here is a result of that protesting.

Each of the personality variables and the dependent variable
(orientation toward change) are measured by Likert type agree-disagree
Opinion items. These items are presented randomly in the questionnaire

and are preceded by the following statement.

THE FOLLOWING ARE STATEMENTS ABOUT A NUMBER OF IMPORTANT SOCIAL AND
PERSONAL QUESTIONS. THE BEST ANSWER TO EACH STATEMENT BELOW IS YOUR
PERSONAL OPINION. WE HAVE TRIED TO COVER MANY DIFFERENT AND OPPOSING
POINTS OF VIEW. YOU MAY FIND YOURSELF AGREEING STRONGLY WITH SOME OF
THE STATEMENTS, DISAGREEING JUST AS STRONGLY WITH OTHERS, AND PERHAPS
UNCERTAIN ABOUT OTHERS. WHETHER YOU AGREE OR DISAGREE WITH ANY STATE-
MENT, YOU CAN BE SURE THAT MANY PEOPLE FEEL THE SAME AS YOU DO.

MARK EACH STATEMENT IN THE LEFT MARGIN ACCORDING TO HOW MUCH YOU AGREE
OR DISAGREE WITH IT. PLEASE MARK EVERY ONE. MARK 1, 2, 3 OR 4, 5, 6
DEPENDING ON HOW YOU FEEL IN EACH CASE.

54

l. I DISAGREE VERY STRONGLY 4. I AGREE A LITTLE
2. I DISAGREE ON THE WHOLE S. I AGREE ON THE WHOLE
3. I DISAGREE A LITTLE 6. I AGREE VERY MUCH
Beginning with the items to measure the orientation toward change,

we can discuss each set of opinion statements designed to indicate the

personality attributes.

Orientation Toward Change

The orientation toward change, as conceptualized in this study,
is a general predisposition to accept or resist change across a wide
range of public and private situations. I have tried to construct
items which direct the respondent's attention to a variety of situations
in which change may take place. Disagreement with these items indicates
a readiness to accept change while agreement indicates a resistance to

change. The questionnaire contains the following eleven items.

1. Legislative apportionment is a complex matter, and the changes
that have been made are probably going to make things worse.

2. It is never wise to introduce changes rapidly in social policy,
government, economics or in most things for that matter.

3. I don't think we should try to politically change the racial
situation in this country any more than we already have.

4. More often than not when you change things they get worse
rather than better.

5. I think there are many things in society that just should not
be changed.

6. I feel that one of the most important aims in life is to settle
down in a community.

7. It is better to stick with the way things are usually done than
to try new ways.

8. Generally politics are so complex that if you attempt to change
some things you are likely to upset the entire system.

55
9. The best job for me would be one which allows me to say in the area
where my parents, sisters and brother, etc. live.

10. Considering everything there are usually more advantages gained by
staying in the same job than by moving around.

11. I don't think we need to change our foreign policy toward Viet Nam
in any way.

One of the groups used in the study was available at the time I
carried out the research, and I was able to retest all of the opinion
items on the same individuals after a lapse of three weeks which should
have reduced recall effects. The eleven items in this set obtain a
test—retest reliability coefficient of .75 indicating substantial
response stability over time. The same set of items has been tested
as to the degree of internal consistency (equivalence rather than
stability) in a single test situation.1 The Kuder-Richardson reliability
coefficient (formula 20) for the eleven items is .482. Since this
moderate level of internal consistency seemed insufficient to argue a
psychologically interpretable orientation, a Likert item analysis (item
score correlated with test score) has been performed, and seven items
have been eliminated as not meeting the desired significance level. The
remaining four items (2, 6, 10, 11) appear to adequately reflect the
general predisposition toward change under a variety of conditions, and
result in an equivalence coefficient of .61. Scores from these four

items are used as measures of acceptance or resistance toward change.

 

1L. J. Cronbach, "Coefficient Alpha and the Internal Structure
of Tests," Ps chometrika, 16, No. 3 (Sept., 1951), pp. 297-334; A. R.
Baggaley, Intermediate Correlational Methods (New York: McGraw-Hill,
1960).

 

56

Some contrasts between coefficients of equivalence and stability
as indicants of reliability are discussed in the last portion of this

chapter.

Dogmatism

I have measured the first structural attribute Of personality
(i.e., the way peOple believe rather than what they believe) with an
abbreviated form of Rokeach's Dogmatism Scale (1960). The essential
characteristic of the dogmatic person is the inability to differentiate
message and source. The less dogmatic person is able to separate
source and message and in turn able to act on the content of the
messages in the environment relatively free of his evaluation of the
source of the message. Troldahl and Powell (1965) have designed a
shorter form of the Rokeach scale appropriate for field studies. They
have ranked the twenty items with highest degree of correlation with
test score; these twenty items adequately reflect the theoretical

dimensions of the original dogmatism scale.

1. The present is all too often full of unhappiness. It is only
the future that counts.

2. I believe that most people just don't know what's good for
them.

3. In a discussion I often find it necessary to repeat myself several
times to make sure I am being understood.

4. It is only when a person devotes himself to an ideal or cause that
life becomes meaningful.

5. I feel that of all the different philosophies which exist in this
world there is probably only one which is correct.

6. I think there are two kinds of people in this world; those who
are for the truth and those who are against the truth.

10.

11.

12.

13.

14.

15.

l6.

17.

18.

19.

20.

57

I believe that man on his own is a helpless and miserable
creature.

I think that even though freedom of speech for all groups is a
worthwhile goal, it is unfortunately necessary to restrict the
freedom of certain political groups.

I believe that while I don't like to admit this even to myself
my secret ambition is to become a great man, like Einstein or
Beethoven, or Shakespeare.

I believe that the main thing in life is for a person to want
to do something important.

I feel that it is often desirable to reserve judgment about
what's going on until one has had a chance to hear the opinions
of those one respects.

I think that the United States and Russia have just about
nothing in common.

I feel that most people just don't give a "damn" for others.

I feel that it is better to be a dead hero than to be a live
coward.

I believe that most of the ideas which get printed nowadays
aren't worth the paper they are printed on.

I believe that in this complicated world of ours the only way
we can know what's going on is to rely on leaders or experts
who can be trusted.

I'd like it if I could find someone who would tell me how to
solve my personal problems.

My blood boils whenever a person stubbornly refuses to admit
he's wrong.

I believe that to compromise with our political Opponents is
dangerous because it usually leads to the betrayal of our own
side.

I think that the highest form of government is a democracy and
the highest form of democracy is a government run by those who
are the most intelligent.

58

The Dogmatism Scale is presented in the literature with reliability
coefficients-in excess of .75.2 In the present analysis the items produce
a stability coefficient Of .65 and a coefficient of equivalence of .56.
Part Of the reason for the lower internal consistency resulting from
the present analysis may be the type of reliability measure. I have
used the Kuder-Richardson coefficient which is the mean of all possible
split halves tests. Rokeach's reliability coefficients are derived.
from split halves tests. The items are believed to have sufficient
stability and equivalence for the purposes of the study, and higher

scores from these items indicate a higher degree of dogmatism.

Reliance Upon Authority

Reliance upon authority has been conceptualized as the extent to
which an individual accepts or relies upon authorities external to a
situation in determining his own Opinions. It seems reasonable to assume
that people who agree more often with authorities about a wide variety of
issues and situations are more reliant, generally, upon external authority
in the determination of their opinions. The respondents were requested to
list the peOple most influential in their lives at the present time,
and then to indicate whether they agreed or disagreed with these individ-
uals about the following subjects: (1) religion, (2) politics, (3) social
problems, (4) choosing a spouse, (5) raising children, (6) choosing a
job, (7) social groups to belong to, (8) foreign affairs, (9) way of

living, and (10) organizations to belong to. To standardize the scoring

 

2M. Rokeach, _p_. cit., 1960, pp. 71-97; U. c. Troldahl and F. A.
Powell, "A Short-Form Dogmatism Scale for Use in Field Studies," Social
P re s, 44 (December, 19 5), pp. 221-214.

59

of the instrument I have divided the number of disagreements by the
number of authorities listed, and this standardized score is used tO
classify the individual as more or less reliant upon authority. Those
subjects with higher scores are classified as being less reliant upon
authority.

The instruments measuring the content attributes (what the
person believes rather than how he believes) of personality can now
be considered. The content attributes are efficacy, trust, and

liberalism.

Personal Efficacy

I have measured this attribute with the personal efficacy scale
developed in the Campbell,.gp..gl., voting study.3 The scale is a four
item Guttman scale. As the items indicate, the scale attempts to
measure the extent to which the individual feels he has control over

his personal environment.

1. I seem to be the kind of person who has more bad luck than good
luck.

2. I have often had the feeling that it is no use to try to get
anywhere in this life.

3. I would rather decide things when they come up rather than always
plan ahead.

4. There's not much use for me to plan ahead because there's usually
something that makes me change my plans.

In the present study these items resulted in a stability

coefficient of .61 and an equivalence coefficient of .42. The item

 

3A. Campbell, P. Converse, W. Miller and D. Stokes, The American
Voter (New York: John Wiley & Sons, 1964).

60

with lowest correlation with test score has been eliminated, and the
three remaining items (1, 2 and 4) have a coefficient of equivalence

of .76. In this scale high scores indicate lower personal efficacy.
Trust

This attribute has been measured by two separate scales. One
scale indicates general trust of other poeple, while the other is
designed to measure trust of politics and politicians. I have used
Rosenberg's MisantrOphy scale to measure the general trust attribute.4

It is described as a five item Guttman scale.

1. Human nature is fundamentally COOperative.
2. Most peOple can be trusted.
3. If you don't watch yourself, people will take advantage of you.

4. Most people are more inclined to help Others rather than think of
themselves first.

5. No one is going to care much what happens to you, when you get
right down to it.

The present analysis finds the scale to have a coefficient of
stability of .58 and an equivalence coefficient of .36. Item analysis
resulted in drOpping two items not meeting the desired significance
level, and the remaining items (2, 3 and 5) have a coefficient of
equivalence of .49. This scale has the lowest level of internal

consistency of any of the instruments. To further reduce items in

 

4M. Rosenberg, "Misanthropy and Political Ideology," American
Sociological Review, 21 (1956), pp. 690-695.

61

order to improve equivalence substantially reduces the possible scale
score variance, so I have retained the scale in its present form.

In addition, a measure of political cynicism has been developed
using items from two different sources.5 The following four items

have coefficients of stability of .58 and equivalence of .55.

1. Money is the most important factor influencing public policy.
2. Politicians never tell us what they really think.

3. It seems to me that politicians lack the education and_knowledge
necessary to make intelligent decisions.

4. To be a really successful politician one has to give up all morality.
Liberalism

The items here have been selected to indicate various liberal
and conservative views, and have been taken from several sources
(Eysenck, 1954; Adorno,.gp..gl., 1950; Rokeach, 1960; and McClosky,
1958). They cover a wide variety of beliefs about political issues,
relationship of individuals and government, human nature, etc.
Disagreement with odd numbered items and agreement with even numbered

items indicates less liberalism.

1. Present laws favor the rich as against the poor.

2. It is right and proper that religious education in schools should
be compulsory.

3. I believe that the death penalty is barbaric, and should be
abolished.

 

5F. A. Pinner,,pp, cit., 1965; Agger, Goldstein and Pearl,

"Political Cynicism: Measurement and Meaning," The Journal of
Politics, XXIII (August), pp. 477-506.

62

4. I believe that war is inherent in human nature.

5. I think that the civil rights organizations should become stronger,
have more influence generally.

6. I doubt very much that government supported medical aid programs
will solve this country's health problems.

7. I believe that our containment of communism foreign policy is
outmoded by present day conditions.

8. I feel that men like Henry Ford or J. P. Morgan, who overcame all
competition on the road to success, are models for all young
people to admire and imitate.

9. I feel that the President has the right to tell big business to
reduce their prices in order to prevent economic problems.

10. America may not be perfect, but the American Way has brought us
about as close as human beings can get to a perfect society.

These ten items have a coefficient of stability of .67, but the degree
of item equivalence is relatively low. The item analysis eliminated
all but three items (3, 4 and 5) which are used in the present analysis.

These three items obtain a coefficient of equivalence of .61.

Primary and Secondary Environments

The following instruments are designed to measure the socializa-
tion variables. The first two instruments measure the extent of
communication with authorities in the primary and secondary environ-
ments. The open socialization environment is conceptualized as
involving more communication with authorities while the more closed
environment involves less communication. It is necessary to identify
the subject's authorities and measure the extent of communication with
those authorities. To measure the number of authorities the subjects

were requested to list "the most influential adults" in their life.

63

For the primary environment the subjects were to recall those adults
influential "at about the time you entered first grade" while for the
secondary environment they list people influential at the present

time in their life. Then the subjects were asked to think about these
relationships and to respond affirmatively or negatively to a list of
possible characterizations of those relationships. Even numbered
characteristics are indicative of a more communicative relationship
while odd numbered items reflect less communicative relations. The

instrument appears in the questionnaire in the following form.

THESE QUESTIONS ARE ABOUT YOUR RELATIONSHIPS WITH OTHERS WHEN YOU WERE
VERY YOUNG -- AGAIN ABOUT THE TIME YOU ENTERED FIRST GRADE OR AS EARLY
AS YOU CAN REMEMBER. AS YOU ANSWER THE QUESTIONS TRY TO RECALL HOW

THINGS WERE AT THAT TIME IN YOUR LIFE AND NO SO MUCH HOW THEY ARE NOW.

1. Who were the most important or influential adults in your life
during this period of your life prior to entering the first grade?
List as many as you really consider to be important to you at
that time in your life.

 

 

2. Now carefully consider your relationship to each of these adults
during this period of your life. For each person listed in the
above questions please indicate by writing in the appropriate
letter (Y for yes and N for no) which of the following items
seem to apply to your relationship to that person. In order not
to become confused about which adult you are answering the ques-
tions for it may be helpful to cOpy the adults listed above in
the blank spaces provided below.

PERSONS LISTED ABOVE

 

 

 

l. Difficult to be close to.

2. Easy for me to talk with.

3. Generally impatient with me.
4. An affectionate person.

5. A person some peOple might
be afraid of.

64

PERSONS LISTED ABOVE

 

6. Seldom got angry with
me even when I gave
reason to be

7. Most conversations were
about my discipline.

8. Usually considerate to
me.

9. Convinced that he (she)
knew what was best for
me.

10. Often around so I could
talk or do things with
him (her).

The instrument for the secondary environment is presented in the
following form in the questionnaire. The major difference between this
and the preceding item battery is that the subject is to list the adults
important in their life at present, and the ten characterizations of
the relationship are changed somewhat. The even numbered items indicating

less communication with the authority and odd numbers indicating more

communication.

NOW I WOULD LIKE YOU TO READ AND ANSWER THESE QUESTIONS ABOUT HOW THINGS
ARE IN YOUR LIFE AT PRESENT.

1. Who are the most important and influential adults in your life at
the present time?

 

2. Carefully consider your relationship with each of these adults at
this time in your life. Again, for each person that you have listed
above please indicate by writing in the appropriate letter (Y for yes
and N for no) which of the following items seem to apply to your
relationship with that person. Once again it may be helpful to copy
the persons listed above in the spaces provided.

65

PERSONS LISTED ABOVE

 

1. Pretty hard for some
people to get to know.

2. A person you can talk to
even about the most dif-
ficult personal problems.

3. Believes most of the same
things as I believe.

4. Considers my opinion to
be important.

5. The source of most of my
ways of thinking about
things.

6. Disagrees with me on many
important matters.

7. Seldom asks my advice on
important matters.

8. A person involved in a
variety of things.

9. A person who knows what is
really good for me.

10. A person to whom I often
go for advice, although I
don't necessarily follow
it.

Each of these ten characterizations is designed to measure the
extent of communication with authorities in the two environments. In
order to measure the extent of internal consistency a split halves
reliability coefficient has been calculated after the items are cor-
rected for direction. The primary environment instrument has a

coefficient of .64 and the secondary environment instrument obtains

a coefficient of .55. The scoring of these instruments presents some

66

problems. Since I am only interested in the extent of communication

it is necessary to standardize each subject's score on both instruments.
In each case I have added the number of negative responses to even
numbered items and the affirmative responses to odd numbered items,

and then divided by the number of authorities listed. Each resulting
score is the average number of communicative responses per authority.

High scores indicate a more Open socialization environment.
Consistency_pnd Continuity

Consistency refers to the extent of similarity between the kind
of communication the subjects have with authority in the primary and
secondary environments. To measure the effects of consistency it is
necessary to combine primary and secondary environments. I have
combined the more open and more closed subjects from each of the
environments into the following four experimental groups: (1) subjects
with high scores in both environments (i.e., Open primary and secondary);
(2) subjects with low scores in both environments (i.e., closed primary
and secondary environment); (3) subjects with high scores in the
primary environment and low scores in the secondary (open primary and
closed secondary); and (4) subjects with low scores in the primary and
high scores in the secondary environment (i.e., closed primary and
Open secondary). It is possible to compare these four groups as to the
differences in acceptance or resistance to change among subjects from
each of the four socialization patterns.

The variable of continuity refers to the extent of individually

perceived trauma or crisis in socialization. I have asked the subjects

67

to "Think about the experiences that you have had from childhood to
the present, and list below the ten most important experiences that
you have had." Most Of the experiences listed by the subjects are
such events as eighth grade graduation, first date, driving the family
car, vacations, etc. However, other experiences were listed which
seem to indicate the type of traumatic or disorientating events I

am seeking to measure. These experiences include such events as

"being lonely, getting in several scraps, parent's divorce,"
"death in the family," etc. Subjects are classified as experiencing
discontinuity if they list one or more of the latter type Of experi-
ences. Sixty-nine subjects are classified as experiencing discontinuity
in socialization.

The last socialization variable, extent of formal education, is
measured in years of college completed by the respondent. More than
two thirds of these subjects have completed some college, while about
a third had not entered college at the time of the study.

This completes the presentation Of instrumentation for the
study. Before going on to the data analysis I have included a

discussion of one of the problems of determining instrument reliability

which has plagued this and most attitudinal research.

Reliability and the Inference of Attitudes

Any research effort based upon measurement rests upon an all-too-
Often-unverified assumption concerning the accuracy or dependability of
the measuring devices. The "reliability coefficient" is a means of

demonstrating the degree of confidence the researcher may have in this

68

assumption. As Cronbach states, "A reliability coefficient demonstrates
whether the test designer was correct in expecting a certain collection
of items to yield interpretable statements about individual differences."6
Cronbach further argues two alternative approaches to the measurement

of reliability.

A retest after an interval using the identical
tests, indicates how stable scores are and therefore
can be called a coefficient of stability. The
correlation between two forms given virtually at the
same time, is a coefficient of eguivalence, showing
how neprly two measures of the same general trait
agree.

Apparently Cronbach's suggestion finds favor generally among
researchers. Baggaley (1964) makes reference to a joint publication
of the American Psychological Association, the American Educational
Research Association and the National Council of Measurements Used
in Education which attempts to standardize terms being used as
estimates of reliability. Their publication suggests the following
procedures are in common use for this purpose: (1) temporal (i.e.,
retest); (2) interobserver agreement; (3) internal consistency which
in turn can be measured by equivalence, split-halves procedures, or
Kuder—Richardson formulas.8

Given these different procedures it becomes of critical
importance to understand which of the methods are most appropriate

for inferring attitudes and what is the relationship among the different

 

6Cronbach,.pp. cit., p. 297.

7Ibid., p. 298.

8Baggaley, pp, cit., p. 61.

69

methods. Cronbach argues retest procedures may be the preferred
method for estimating instrument reliability, but because it is
difficult to recapture subjects for the post test (after a lapse of
time necessary to reduce recall effects), it may be more practical to
use measures of internal consistency as estimates of reliability.9
Baggaley offers another caution about the use of retest procedures

in estimating reliability. He notes that if the lapsed time before
retest is too long there may be genuine changes in the population
interfering with the estimate or reliability. On the other hand if

the retest is made too soon (e.g., less than one week) the subject's
recall of the previous test may influence reliability. He suggests

the period between tests should be specified, and that it should be
not less than one week nor more than six months.10 Even if retest
methods are the preferred way of estimating reliability they are

still the less practical. Given this practical problem many research-
ers have adopted the practice of using measures of internal consistency
to estimate reliability. They assume a positive relationship between
these different procedures. This assumption has been debated, as we
will see, but it also appears amenable to empirical testing which is

the purpose of this portion of the chapter.

 

9Cronbach, _p_. cit., p. 298.

10Baggaley,__p. cit., p. 62.

7O

Cronbach cites both Guttman and Guilford in the debate. Guttman

argues,

. . we have the assurance that if the items are
approximately scalable, then they necessarily have
very substantial test-retest reliability.11

Guilford takes the opposite view that,

There can be very low internal consistency and
yet substantial or high retest reliability. It is
probably not true, however, that there can be high
internal consistency and at the same time low retest
reliability, except after very long time intervals.1

And Cronbach concludes, "The only prOper conclusion is that alpha (his
term for the Kuder-Richardson coefficient measuring internal consistency)
may be either higher or lower than the coefficient of stability over
an interval of time."13
The implication of this debate and its assumptions are clearly
critical for attitudinal research. On the one hand there are those
who claim measures of internal consistency are not only adequate for
assuming stability of items over time, but are the best criterion for
inferring an attitude. An attitude is a hypothetical construct. It
must be inferred from either verbal or overt behaviors. The usual
practice is to infer an attitude when individuals respond to items

on a questionnaire (or in some other manner) in an internally consistent

manner. For example, Guttman argues items must associate closely enough

 

11Cronbach, pp, cit., p. 308.
12Ibid., p. 308.
13

Ibid., p. 309.

71

to meet the arbitrary criterion of a coefficient of reproducibility of
.90 or better. The argument is that if all the items designed to measure
a particular attitude are interrelated or internally consistent to this
degree then it is acceptable to infer that the attitude being measured
actually exists in a given population.

There are problems with this argument from my point-of—view.
First--What level of internal consistency is required to infer an
underlying attitude, and second, do not most definitions of attitude
require stability over time as well as internal consistency? We need
to examine both questions.

Cronbach argues "psychological interpretability" can be based
upon a less severe criterion than scalability. He says, "In view of
the relations detailed above, we find it unnecessary to create homo-
geneous scales such as Guttman, Loevinger, and others have urged."14
The argument here is that the researcher does not have to infer the
person's score on each item from his score on the total test which
seems to be the purpose of attaining unidimensionality. Some lower
level of internal consistency may well be psychologically as inter-
pretable as a "Guttman scale." Cronbach further argues that such
interpretability may be assumed when the score is obtained from,

" . . . a set of items having a substantial alpha and not capable of

division into discrete item clusters which themselves have high alpha."15

 

1"Ibid., p. 330.

lsIbid., p. 330.

\l
F)

If coefficients of equivalence are to be used in the interpre-
tation or inference Of attitudes then scalability may be too severe a
criterion. Lower degrees of internal consistency may be acceptable
for the interpretation of attitudes from questionnaire items. This
conclusion may resolve some practical problems in the measurement of
attitudes, but it does not confront the more difficult problem as to
what criteria are apprOpriate for the inference of attitude.

To examine this question more closely I return for a moment to
the definition of orientation (attitude) as used in this study. "An
attitude is a relatively enduring organization of beliefs around an
object or situation which predisposes one to act in some preferential

16 The definition does not require the "organization of

manner."
beliefs around an object or situation" to be internally consistent

at any particular level, although it does require some degree of
relation to the objects or situations. The definition does require
the organization of beliefs to be "relatively enduring." That is,

the definition seems to require some level of stability over time
(i.e., retest reliability). Evidently both coefficients of stability
and equivalence are required for the inference of an attitude. This
conclusion focuses our attention on the comparability of these two
coefficients. If the two measures produce similar results then either

may be used in the interpretation of attitudes underlying behavioral

responses. If they are not similar in result, then it may be

 

16M. Rokeach, pp. cit., 1966, p. 3.

73

necessary to obtain both measurements of internal consistency and
stability before appropriately inferring an attitude.

Because seventy of the respondents in this study were available
for retesting, it was possible to obtain retest scores after a period
of approximately three weeks. Retest reliability coefficients have
been calculated for the scales measuring the orientation toward change,
liberalism, dogmatism, personal efficacy, political cynicism, and
misanthrOpy (general trust of other people). In order to shed some,
light on the problem of the comparability of coefficients of stability
and equivalence, I have further calculated both Kuder-Richardson and
split-halves reliability coefficients for the same scales. It is
useful to recall that two of these instruments are presented in the
literature as Guttman scales (i.e., personal efficacy and misanthrOpy).

The following table compares these coefficients.

TABLE 2.w—Comparison of coefficients of stability and equivalence

 

 

 

Test- Kuder- Split

Retest Richardson Halves
l. Orientation toward Change .75 .48 .03
2. Liberalism .67 .30 -.22
3. Dogmatism .65 .56 .66
4. Personal Efficacy (Guttman scale) .61 .42 .49
5. Political Cynicism .58 .55 .58

6. MisanthrOpy (Guttman scale) .58 .36 -.58

 

7a:

The measurements of internal consistency are those calculated before item

analysis improved the levels of internal consistency for the scales being
used in the present analysis. These data are meant to compare different
reliability estimates, and not to argue the reliability of instruments
used in the analysis.

The table reveals some disconcerting relationships among the
different coefficients. The comparison of coefficients cast real doubt
on the use of measurements of internal consistency to approximate retest
reliability. I have computed Spearman rank order correlations among
the three modes of reliability estimation. I first ranked the sets of
items from highest to lowest reliability for each of the three methods,
and then correlated these rankings. The two measures of internal
consistency (i.e., Kuder-Richardson and Split—Halves) correlate at the
.60 level. However, when the two measures of internal consistency are?
compared with the retest coefficient the results are clearly inconsistent
with the assumption that one type of reliability estimate can approximate
the other. The correlation between Kuder-Richardson and retest coeffi-
cients is .03, and the correlation between split-halves and retest
coefficients is -.60. Both of these latter correlations are difficult
to accept, and the negative correlation between split-halves procedures
and retest is incredible. Such findings demand replications with other
data. If these findings survive replications, then a great deal of
attitudinal research based upon split—halves reliability estimates is
drawn into serious question.

The explanation of this serious discrepancy between measurements
of internal consistency and retest reliability may lie in the type of

variance each test is designed to measure. Measurements of internal

3213:
EKC‘.

have

75

consistency are designed to reflect the degree to which each item
varies from each other item or from the score for the total items
(i.e., inter-item variance). The retest reliability is based on the
variance in total score from one situation to another (i.e., inter-
score variance). As Cronbach has suggested these are different
measurements and not mathematical approximations of one another,
although they may be empirically associated. As we have seen here
they are neither mathematical or empirical approximations of one
another. The question becomes which of the types of reliability
extimates are most apprOpriate for the inference of attitudes. I
have argued earlier the definition of attitude being used here places
more emphasis upon stability over time than upon equivalence within a
particular situation. That is, we need to measure inter-score variance
more importantly than inter-item variance. The organization of beliefs
which make up the attitude or orientation in a particular situation
may be relatively loose (i.e., lower internal consistency), but this
is a characterization of the attitude not an indication that the
attitude does not exist. If the same relatively loose organization
of beliefs is found to exist on different occasions with the same
people (retest reliability) then we have evidence that the attitude
does exist even in its relatively inconsistent form. The authors of

The Authoritarian Personplity note the same phenomenon.

The reliability and internal consistency of the
PEG scales suggest, on the one hand, that liberalism
and conservatism are relatively organized and measureable
patterns of current politico—economic thought; and, on the
other hand, that within each of these broad patterns there
is considerable sub-pattern inconsistency, and simple

’6

ignorance. To ignore either the relative generality or
the relative inconsistency would, it seems, lead to
serious misunderstanding of the problem 1’

Clearly some level Of internal consistency is required to avoid
the inference of an attitude when in fact the scale measures several
different objects or represents sev'eral dierrent attitudes, but I
am arguing that too severe a criterion for internal consistency may
well rule out the existence of an attitude when in fact it merely
reflects the relatively unorganized state Of the beliefs composing
the attitude. On the basis of these findings it is difficult to
argue the existence of an attitude or orientation without both
retest and equivalence coefficients. The scales in the present
analysis have met both criteria, and therefore, I have considerable
confidence in the reliability of these instruments.

We can move to the data presentation with these considerations

of study design and instrument reliability complete.

 

17T. W. Adorno, g£._§l.,_gp. cit., p. 176.

 

CHAPTER IV

PERSONALITY ATTRIBUTES AND THE ORIENTATION

TOWARD CHANGE

The theoretical framework guiding the organization and preSentation
of data directs attention to the effects of personality attributes and
socialization experiences upon the orientation toward change. In this
chapter I first discuss the relationships among the personality variables
being used, and second, the impact of these attributes upon the accept-

ance of change.
Structural and Substantive Orientations

The general problem is the investigation of the existence and
extent of relationships among the structural and substantive personality
attributes. As a first step, the correlation matrix (product moment
correlations) in Table 1 gives some idea of the degree of relationship
among the variables.

As Table 3 indicates there are significant relationships between
dogmatism.and political cynicism, orientation toward change, personal
efficacy and trust. Liberalism relates to the variables personal
efficacy and trust, and efficacy, in turn, relates to trust.

These correlations only identify linear relationships. To

further probe the data, I have used four by four contingency tables.

77

78

TABLE 3.--Correlation matrix of personality variables

 

 

 

l. 2. 3. 4. 5. 6. 7.

l. Dogmatism - .00 +.35* ~.25* -.33* -.l6* .06

2. Liberalism — - .02 +.15* .09 .05 .08

3. P01. Cynicism - - - .05 -.26* —.26* .04
4. Orientation

toward Change - — — - .OO .08 .10

5. Personal Eff. — — - - — +.13* .01

6. Trust - - — — - — .07

7. Reliance upon
Authority - - - - - - -

 

*Significant beyond the .05 level

This use of variables in cross—breaks permits the identification of
non-linear relationships. Kendall tau correlation coefficients have
been computed in order to indicate the extent of association utilizing
a non-parametric technique requiring assumptions more closely approxi-
mated by the data.

The substantive orientations include trust, sense of efficacy,
and the degree of liberalism in one's political views. Table 4 below
relates dogmatism with the personality attributes. Dogmatism does
not correlate with liberalism.

The strongest evidence is found relating dogmatism with personal
efficacy and political cynicism. Individuals who are less dogmatic
are likely to be personally efficacious and less likely to be cynical

about politics and politicians. The chi-square for the relations of

79

TABLE 4.--Dogmatism and related personality attributes

 

 

 

Dogmatism
Lo. Dog. Mod. Lo. Mod. Hi. Hi. Dog.
(N=43) (Ns8l) (N=74) (N=43)
Personal Efficacy
ZHI 25.6 21.0 13.5 11.6
ZMOD HI 44.2 33.3 32.4 14.0
ZMDD L0 27.9 34.6 35.1 44.2
2L0 2.3 11.1 18.9 30.2
Political Cynicism
ZLO 30.2 16.0 9.5 9.3
ZMDD LO 44.2 38.3 28.4 18.6
ZMDD HI 16.3 30.9 48.6 34.9
ZHI 9.3 18.6 34.9 37.2
Trust .
ZHI - 27.9 18.5 21.6 9.3
ZMDD HI 44.2 32.1 32.4 30.2
ZMOD LO 23.3 34.6 31.1 44.2
ZLO 4.7 14.8 14.9 16.3
Reliance.upon
Authority
ZDEPENDENT 7.0 8.7 13.5 2.3
ZMOD DEP 25.6 18.8 31.1 30.2
ZMOD 30.2 26.3 18.9 34.9
ZMDD INDEP 14.0 38.8 28.4 20.9
ZINDEPENDENT 23.3 7.5 8.1 11.6

 

dogmatism with personal efficacy is significant beyond the .01 level,
and the Kendall's tau for the relationship is .235. Dogmatism and
political cynicism have a relationship significant at the .001 level
with a tau of .262. Trust and dogmatism do not produce a significant
relationship, but the directions evident in the data and the correlation
coefficient suggest further analysis. When the extremely high and low
dogmatic groups are related with two categories of trust (high and law)

rather than four (hi, mod. hi, mod. 10, and lo) we find a relationship

80

which is significant beyond the .01 level with a tau of .134. Both
political trust and trust of others relate positively with low dogmatism.

The cross—classification of dogmatism by reliance upon authority
yields a non—linear relationship which explains the lack of correlation
in the matrix. Upon first glance the relationship appears linear, and
in comparing the more extreme classifications of the two variable I find
a relationship significant at the .10 level. However, when the data are
grouped with high and low dogmatics in one group and moderately high,
moderately low dogmatics in another and then examined in relation to
extreme independence against the other four categories of reliance upon
authority grouped as dependence, we find individuals with moderate
levels of dogmatism (rather than extremely high or low) are more likely
to be independent of authority. This relationship is significant at
the .05 level.

Table 5 shows the second attribute of personality structure with
other related personality variables. Reliance upon authority relates
to all other variables except personal efficacy. Dogmatism, it will be
recalled, relates to all other variables except liberalism; we now find
the second structural attribute relating to all other variables (includ-
ing liberalism) except personal efficacy.

The relationships in this table yield no significant chi-squares
given the four by four classifications, but again, the directions seem
important. We do find a relationship between reliance upon authority
and liberalism after collapsing categories. I have combined the
dependent and moderately dependent subjects into one group and the

other three categories into another. The two groups are then related

81

TABLE 5.--Reliance upon authority by related personality attributes

 

 

Reliance upon Authority
Dependent Mod.Dep. Mod. Mod.Indep. Independent

 

(N=21) (N=62) (N=63) (N=67) (N=27)

Liberalism

Z L0 9.5 9.7 15.9 4.5 18.5

X MOD LO 42.9 53.2 36.5 38.8 29.6

2 MOD HI 38.1 24.2 33.3 40.3 40.7

2 HI 9.5 12.9 14.3 16.4 11.1
Trust

Z HI 14.3 27.4 40.6 14.9 14.8

2 MOD HI 42.9 27.4 36.5 26.9 55.6

Z MOD LO 38.6 38.7 28.6 37.3 22.2

1 L0 14.3 6.4 14.3 20.9 7.4
Political Cynicism

Z LO 14.3 14.5 19.0 11.9 18.5

2 MOD LO 28.6 30.6 34.9 35.8 25.9

2 MOD HI 47.6 38.7 33.3 31.3 25.9

2 HI 9.5 16.1 12.7 20.9 29.6

 

to liberalism (low and moderately low in one group and high and
moderately high in another) at the .05 level. The direction of the
relationship indicates that the more liberal the groups are in their
political views the more likely they are to be independent of their
authorities.

It is also necessary to collapse classifications to find a
relationship between reliance upon authority and the two variables
trust and political cynicism. This time reliance upon authority is
divided into two groups (dependent, moderately dependent and moderate
against moderately independent and independent) and related to trust
(low trust against high) and political cynicism (low, moderately low

and moderately high against high cynicism). These relationship are

82

significant at the .05 level for political cynicism and at the .10
level for general trust of other people. In both cases the direction
of the relationship suggests that the politically cynical and the least
trusting groups are likely to be independent of their authorities.

When we examine the data on personality content variables--trust,
liberalism and political cynicism-~and related attributes, we find the
following.

Trust relates to dogmatism and reliance upon authority, which
has been discussed, to personal efficacy and political cynicism
discussed here, and to the orientation toward change to be discussed
later. Trust does not relate to the extent of liberalism character-
izing one's political beliefs.

In Table 6 we find relationships well beyond the .05 level.

The directions indicate that subjects characterized by trust of other

TABLE 6.--Trust by related personality attributes

 

 

 

Trust
High Mod. Hi. Mod. Lo. Low
(N=47) (N=82) (N=80) (N=32)
Personal Efficacy
Z HI 27.7 12.2 15.0 25.0
Z MOD HI 36.2 36.6 30.0 15.6
Z MOD LO 23.4 39.0 41.2 28.1
Z LO ~ 12.8 12.2 8 31.3
Political Cynicism
Z LO 27.7 12.2 11.2 15.6
Z MOD LO 40.4 37.8 31.3 12.5
Z MOD HI 21.3 34.1 42.5 34.4

Z HI 10.6 15.9 15.0 37.5

 

83

people are less likely to be cynical about political affairs and
personalities. General trust of other people is clearly related to
trust of politics and politicians. The tau for the relationship is
.191. The relationship between trust and personal efficacy is also
clear; there is a tendency for groups who trust others to be personally
efficacious (tau = .110).

The second content variable of personality, liberalism, is
found to relate to only three other attributes--efficacy, reliance
upon authority (discussed above), and orientation toward change

(discussed below).

TABLE 7.-—Liberalism by related personality attributes

 

 

 

Liberalism
Lo. Lib. Mod. Lo. Mod. H1. H1. Lib.
(N=26) (N=100) (N=82) (N=33)
Personal Efficacy
Z HI 23.1 25.0 11.0 9.1
Z MOD HI 30.8 27.0 34.1 39.4
Z MOD LO 30.8 35.0 36.1 36.4
Z LO 15.4 13.0 18.3 15.2

 

As the distribution is presented, there is no relationship in the
data; however, when the extreme groups of the efficacy variable are
related with only two classifications of liberalism (low and moderately
low against moderately high and high) the relationship is significant
at the .05 level (N - 80). Conservatives (less liberal subjects) are

more apt to be personally efficacious than are the liberal subjects.

84

The last variable in this part of the presentation is political
cynicism. Individuals who are cynical about politics, as previously
found, are more likely to be dogmatic, less trusting of others and
independent from their authorities.

In Table 8 cynicism is strongly related to efficacy.

TABLE 8.--Politica1 cynicism by related personality attributes

 

 

Political Cynicism
Lo. Cyn. Mod. Lo. Mod. Hi. Hi. Cyn.

 

(N=37) (N=79) (N=83) (N=42)
Personal Efficacy
Z HI 37.8 13.9 18.1 7.1
Z MOD HI 24.3 39.2 32.5 21.4
Z MOD L0 32.4 36.7 34.9 35.7
Z L0 5.4 10.1 14.9 35.7

 

This relationship is significant beyond the .001 level (tau = .203) and
the direction clearly suggests the groups with less political cynicism
are more personally efficacious.

With this presentation of data on the relationships among the
various personality attributes completed, we can discuss these attributes

in relation to the orientation toward change.

Correlates of Orientation Toward Chang;

What_is the impact of the various personality attributes upon the
individual's acceptance of change? We have developed four hypotheses in
the earlier theoretical chapter. The first three hypotheses led to the

general expectation that high levels of dogmatism positively relate to

85

dependence upon authorities, less trust of peOple and politics, less
personal efficacy, less liberal beliefs, and finally, greater resistance
to change.

This statement of the general hypothesis for this portion of the
analysis requires a multi-variate statistical model. We have expected
the acceptance of change (i.e. the criterion variable) to be affected
by several personality attributes (i.e. the independent variables). I
have used multiple regression and multiple correlation analysis to test
this hypothesis. The following table presents the simple.(zero-order)
correlations between the criterion variable and each of the independent
variables, the regression coefficient (beta weights) of each independent
variable holding others constant, and the computed t-value for each of
the beta weights. The t-value for the beta weights has been computed
by dividing the beta weight by estimated standard error of the partial
regression.

Before presenting the table it is necessary to discuss briefly
the interpretation of both the regression coefficients and the multiple
correlation coefficient. Blalock compares the two coefficients in the

following manner.1

The partial correlation is a measure of the amount
of variation explained by one independent variable after
the others have explained all they could. The beta
weights, on the other hand, indicate how much change in
the dependent variable is produced by the standardized
change in one of the independent variables when the
others are controlled.

 

1H. M. Blalock, Social Statistics (New York: McGraw-Hill, 1960).

21bid., p. 345.

Thea

mt:

Oii

I
".9”

CI:

86

Blalock's reference to "standardized change" points out the
essential characteristic of the beta weights. Since the independent
and dependent variables being compared are often measures of different
kinds of data it is necessary to standardize units so the comparison
is meaningful. To do this each-variable is divided by its standard
deviation, and "We thus measure changes in the dependent variable in
terms of standard-deviation units for each of the other variables, a
fact which assures the same variability in each of these variables.
These adjusted partial slopes are called beta weights."3 In more
intuitive language, the beta weights indicate how many units (or parts
of a unit) of change one can expect in the dependent variable by one
unit of change in the independent variable, holding other independent
variables constant. Blalock notes beta weights and partial correlations
ire different types of measures of association but they do render
comparable results. He concludes,.9 ... . they will not give exactly
the same results although usually they rank variables in the same
order of importance."4

The multiple correlation in the analysis indicates how much of
the total variation in the criterion variable can be explained by all
of the independent variables acting together. We can now examine Table
9 more closely to determine the impact of the various personality

attributes upon the acceptance of change.

 

31bid., p. 345.

4Ibid. ’ p. 345.

87

TABLE 9.--Regression coefficients, zero-order correlations and t-values
for orientation toward change related to other personality attributes

 

Correlation with Regression
Variables orientation toward Coefficient t-values
change (beta weight)
Dogmatism .2524 .0951 4.0159
Liberalism .1417 .1358 2.0602
Political Cynicism .0537 .0276 0.3729
Personal Efficacy .0065 .1074 1.2753
Trust .0582 .0582 0.7471
Reliance upon
Authority .1036 .1553 1.2524

 

The test of significance for the total regression produces an
F-ratio of 4.3297 (df = 6, 234) which is significant beyond the .001
level. The multiple correlation resulting from the regression analysis
is .3161. Computing the coefficient of determination (or alienation)
we find approximately lOZ of the total variance in the orientation
toward change is explained using the six personality attributes in the
table. This is, indeed, a disappointingly small amount of explained
variation. Perhaps it is to be expected given the relatively complex
psychological predisposition being predicted.

The beta weights and associated t-values lead to the conclusion
that dogmatism, liberalism, personal efficacy and reliance upon authority
are the most important contributors to the variance in the acceptance
of change. The fact that some of the independent variables are

confounded affects the results of the regression analysis. As Blalock

88

states, "On the other hand, if the intercorrelations among the independent
variables are quite high in magnitude the multiple R will ordinarily not
be much larger than the largest total correlation with the dependent
variable."5 In this context it is important to recall that dogmatism

and personal efficacy correlate with a coefficient of .33. There

are also significant correlations among trust, personal efficacy and.
political cynicism, but these variables contribute little to the change

in the criterion variable.

With these reservations explicit it is still clear that we have
substantially improved the amount of variance explained by these variables
by using the combination of variables in the regression model.

In order to examine more closely the relationship between the
personality attributes and the acceptance of change, I have also
constructed contingency tables for each of the personality attributes
related to the orientation toward change. This analysis may clarify both
the existing relationships and the lack of relationship among the
variables.

Table 10 shows orientation toward change cross-classified with
four personality attributes. Acceptance of change does not correlate
with personal efficacy or political cynicism (although it does relate
to general trust of others). The fact that the present analysis does
not find a relationship between efficacy and acceptance of change points
out the less powerful nature of the statistical model being used. The
previous.regression model did suggest efficacy as a contributor to
acceptance of change. These relationships are not strong, but it is

important to note they are not consistent with either the theoretical

89

TABLE lO.--The orientation toward change by related personality
attributes

 

 

Orientation Toward Change

 

Accept Mod. Accept Mod. Resist Resist
(N=44) (Ns83) (N=78) (N=36)

Dogmatism

Z LO 29.5 19.3 10.3 16.7

Z MOD LO 38.6 39.8 28.2 25.0

Z MOD HI 20.5 27.7 37.2 36.1

Z HI 11.4 13.3 24.4 22.2
Liberalism

Z LO 13.6 6.0 15.4 8.3

Z MOD LO 22.7 48.2 48.7 33.3

Z MOD HI 47.7 32.5 20.5 50.0

Z_HI 15.9 13.3 15.4 8.3
Trust

Z HI 12.8 25.6 20.0 3.1

Z MOD HI 38.3 31.7 33.7 37.5

Z MOD LO 27.7 30.5 33.7 40.6

Z LO 21.3 12.2 12.5 18.8
Reliance upon

Authority

Z DEPENDENT 6.8 11.0 7.7 8.3

Z MOD DEP. 27.3 19.5 29.5 30.6

Z MOD 18.2 28.0 28.2 27.8

Z MOD INDEP. 29.5 31.7 24.4 25.0

Z INDEPENDENT 18.2 9.8 10.3 8.3

 

expectations or the results of the regression analysis. The following
serves to underscore the caution which must be used in interpreting the
results of the foregoing analysis.

Dogmatism and orientation toward change in Table 10 do not yield
a statistically significant relationship, although both the directions
in the data and a tau of .183 suggest a relationship can be found if

the classifications are re-ordered. In a two by two cross-break (low

90

and moderately low dogmatism against moderately high and high; accept,
moderately accept against moderately resist and resist) there is
substantial relationship, significant beyond the .001 level. The
direction is consistent with the first hypothesis. Groups with subjects
likely to be less dogmatic also contain subjects who accept change. Our
first hypothesis is supported by the present data, so we may turn to
the second hypothesis in this search for problems contributing to the
lack of support for the general hypothesis.

In testing the second hypothesis there is no relationship in
the above table as the data are ordered. When the data are reordered
into the following groups: accept and moderately accept as one group;
resist and moderately resist as another; three classifications of
dependence the third group; and two classifications of independence as
the fourth; we discover a chi-square for the two by two table of 2.238.
A chi-square of 2.7 is required for statistical significance at the .05
level since the direction has been predicted, so the resulting relation-
ship is not statistically significant. I do consider this supportive
of the second hypothesis since the relationship does approach statistical
significance and is in the predicted direction. We conclude that there
is only modest support for the contention that subjects who are more
independent from their authorities are also likely to accept change.

When the relationship between the acceptance of change, liberalism
and trust is examined, we again discover what appears to be curvilinear
relationships. There is in the regression analysis no evidence of
relationship between the acceptance of change and trust, although there

is evidence of linear relation between liberalism and acceptance of

91

change. Since the regression model is sensitive only to linear
relationships it is not surprising that we have found a curvilinear
relation between trust and acceptance of change not reflected in the
previous analysis. However, it is surprising to find strong evidence
of a curvilinear relation between liberalism and acceptance of change
since we have had previous indication of a linear relationship.

The analysis of these two relationships again involves construct-
ing two by two contingency tables from the data in Table 10. I have
regrouped the data with low and moderately low liberalism as one
classification, high and moderately high liberalism as another, groups
accepting and resisting change as a third, and finally those who.
moderately accept or resist as the foufth classification. The chi-
square for this analysis is significant at the .01 level, and the
direction of the relationship suggests that subjects with more liberal
political views are more likely to either accept or resist change while
those with less liberal views are more moderate in their acceptance or
resistance of change. This is clearly inconsistent with the expectation
that liberalism relates to acceptance of change in a linear fashion.

I have also found a curvilinear relation between trust and
acceptance of change. This relationship is obtained by using only the
extreme groups on the orientation toward change variable (i.e. most
accepting and most resisting) and-cross~classifying these groups with
two classifications of the trust variable (e.g. high and low trust
subjects in one group and moderately high and moderately low subjects
in another). This results in a chi—square significant at the .01 level

(N = 80). The direction indicates subjects from the moderate categories

92

on the variable of trust are more likely to accept change than are
either the most trusting or the most distrusting subjects.

Testing the third hypothesis, personal efficacy and political
cynicism are not realted to orientation toward change as expected.
General trust of others and liberalism are related, but not in the
expected manner.

Our general expectation was that subjects who tend to be dogmatic
also tend to be dependent upon authorities, less personally efficacious,
less trusting of people and politics, conservative in political beliefs,
and finally, resistant to change. Less dogmatic subjects are likely
to accept change, and there is moderate support for the belief that
individuals who are independent of their authorities are likely to
accept change. We can only conclude from the previous analyses that
personal efficacy and political cynicism are not related to the
individual's ability to adjust to change. Finally, we have seen in
the contingency analysis that the most liberal subjects are apt to
either accept or resist change, while moderately trusting subjects
are likely to accept change. All of these findings suggest the
acceptance of change is most affected by dogmatism, liberalism,
independence of authority and moderate levels of trust of people

generally.

Summary

The first problem in.this chapter is the examination of inter—
relationships among the various personality attributes. We find

people with less dogmatic belief systems are more likely to be

93

personally efficacious and more trusting of politics and people.

Also, those who are more efficacious are less liberal in political
views.{(lndependence of authority is positively related to more
liberal views and a greater degree of distrust of politics and_peOple;)
[Both low and high dogmatic subjects are likely to be more independent
of their authorities;i Next we have examined the question: Do
personality attributes affect the orientation toward change?

We find, here, liberalism, dogmatism, reliance upon authority
and moderate levels of trust have the most impact upon the acceptance
of change. Personal efficacy and political cynicism apparently have
no effect upon our dependent variable. it appears that individuals
characterized by low dogmatism, moderate trust of people, and
independence from their authorities probably accept change most
easily;ffifvidently liberals are either most likely to accept or most
likely to resist change.

We have confirmed the hypothesis that the greater the degree of
dogmatism the greater the degree of resistance to change. There is
some evidence, although not strong, to support the hypothesis that
the greater the reliance upon authority the greater the resistance to
change. The third hypothesis is not supported by the data. (it was
expected that the greater the personal efficacy, the more trust of
other people, the more liberal the political beliefs, the more likely
the acceptance of change} Inwreality, efficacy does not relate to
acceptance of change, moderate levels of trust rather than high or

low relate to acceptance of change, and liberals are either accepting

or resisting of change.

94

The only conclusion appropriate at this point is that dogmatism,
liberalism, moderate levels of trust, and independence of authority
(at least disagreement with the opinions of authorities) contribute
to the existence of a psychological predisposition to accept change
in various social and personal situations. In recalling the results
of the regression and multiple correlation analysis we are reminded
that only a very small portion of the variance in this complex

predisposition has been explained with these variables.

CHAPTER V

SOCIALIZATION EXPERIENCES AND THE ORIENTATION

TOWARD CHANGE

The foregoing has been an examination of the effects of personality
attributes upon the orientation toward change. The present chapter
examines the impact of various socialization experiences upon the
personality attributes and the predisposition to accept change.

Theoretically we have given closest attention to the authority
relations a person forms both in early family situations and at the
present time in their life. Specifically, we wish to examine the
extent of communication with these authorities. The primary environment
is the set of authority relations formed early in life while the
secondary socialization environment is the collection of authority
relations at the present time in the person's life.

What are the effects of the primary and secondary environments
upon various personality attributes including the acceptance of change?
What are the effects of discontinuity and education upon the personality
attributes? Do the four combined socialization processes produce
differences in the various personality attributes, particularly the
orientation toward change? What is the combined effect of the socializa-

tion variables and the personality attributes upon the orientations toward

9S

96

change? The balance of the theoretical expectations are tested in
the discussion of these questions.

A first exploratory step before investigating the separate
relationships among these variables involves gaining some insight
into the correlations among the variables. Table 11 presents a

matrix of correlations between these variables.

TABLE ll.--Correlation matrix of the socialization variables

 

 

 

 

Pri. Env. Sec. Env. Cont. Educa.
Primary Environment - .18* .04 .09
Secondary Environment - - .07 .17*
Continuity - - — .08

Level of Education - — - _

 

*Significant beyond .05 level

Primary and secondary environments are significantly related as
well as continuity and level of education. These intercorrelations
must be kept in mind as we proceed with the analysis. First, then,
we investigate the impact of the primary and secondary environments
upon the various personality attributes.

Primaryignd Secondary Environments
and Personality Attrihgtes

Two theoretical hypotheses are tested in the following analysis.
First, more open primary socialization should relate positively to a
greater capacity to trust other people. Second, the more open the

secondary environment the higher the sense of personal efficacy.

97

Table 12 relates the primary environment to personality attributes and
enables us to test these hypotheses as well as discuss several other

relationships which appear in the analysis.

TABLE 12.--Primary environment by related personality attributes

 

 

Primary Environment

Closed Mod. Closed Mod. Mod. Open Open

 

(N=18) (N259) (Nn7l) (N=64) (N=28)
Reliance upon
Authority
Z DEPENDENT 5 6 5.1 8.5 7.9 21.4
Z MOD DEP 27.8 22.0 22.5 30.2 32.1
2 MOD 11.1 23.7 31.0 23.8 32.1
Z MOD INDEP 27 8 28.8 31.0 30.2 14.3
Z INDEPENDENT 27.8 20.3 7.0 7.9 0.0
Liberalism
Z L0 22.2 11.9 7.0 6.2 21.4
Z MOD LO 33.3 23.7 56.3 50.0 28.6
Z MOD HI 38.9 42.4 23.9 32.8 42.9
Z HI 5 6 22.0 12.7 10.9 7.1
Political Cynicism
Z L0 11.1 11.9 19.7 10.9 21.4
Z MOD L0 22.2 30.5 33 8 35.9 35.7
Z MOD HI 50.0 32.2 31.0 34.4 39.3
Z HI 16.7 25.4 15.5 18.8 3.6
Dogmatism
Z L0 33.3 20.3 12.7 12.5 25.0
Z MOD L0 16.7 30.5 36.6 35.9 39.3
Z MOD HI 38.9 32.2 35.2 26.6 21.4
Z HI 11.1 16.9 15.5 25.0 14.3

 

There are no significant relationships among the variables in

Table 12. However, upon more careful examination several important

relationships are revealed.

When we collapse the classifications of

both variables (i.e. dependent, moderately dependent and moderate in

98

one group against moderately independent and independent in another;
and closed, moderately closed primary experiences against moderate,
moderately Open and open experiences) we find a relationship between
primary environment and reliance upon authority which is significant
at the .02 level. The direction of the relationship in the two-by-two
cross break indicates that subjects from open primary environments are
likely to be dependent upon authority.

Cross-classifying primary environment and liberalism (closed and
moderately closed primary against three categories of subjects with
more open primary experiences; and high and moderately high liberalism
against moderately low and low liberalism) reveals a relationship which
is significant at the .05 level. In this case, groups with liberal
subjects are likely to be from closed primary environments.

The third cross-classification permits testing the first theo-
retical hypothesis. Open primary environment should relate positively
to trust of others. No such relationship is found. However, when
primary environment (closed, moderately closed and three classifications
of openness; and low, moderately low against high and moderately high
cynicism) and political cynicism are cross-classified we find a
relationship significant at the .05 level and in the predicted direction.
If political trust is substituted for the general variable trust, peOple
from open primary environments are found to be trusting. We must keep
in mind that the general variable of trust does not relate to primary
environment, although political trust does. We have some evidence to

support the first hypothesis.

99

The last relationship in the table involves dogmatism and the,
primary environment. Once again we appear to have a curvilinear
relationship significant beyond the .05 level. In this comparison I
group closed, moderately closed and open subjects in one category and
moderately open and moderate in another and compare these groups with
the least dogmatic subjects against three categories of more dogmatic
subjects. The least dogmatic subjects are more likely to come from
either the more open or the more closed primary socialization environ-
ments .

Table 13 presents data used in the second part of this investiga-
tion concerning the impact of socialization environments upon personality
attributes. Here the extent of communication with authorities in the
secondary environment is related to personality characteristics.

After collapsing categories, secondary socialization does relate
to orientation toward change, liberalism, and personal efficacy. To
find a relationship between acceptance of change and secondary socializa-
tion I combine closed, moderately closed and moderate subjects in one
group and moderately open and open in another and cross-classify these
groups with those accepting of change against three categories of
resistance to change and find a relationship beyond the .01 level of
significance. The direction of the relationship indicates that subjects
from open secondary socialization experiences with authority are likely
to accept change.

Communication with authorities in the secondary environment also
is associated with the degree of liberalism characteristic of the subject's

political views. In this case it is necessary.to dr0p the moderate

100

TABLE 13.--Secondary environment by related personality attributes

 

fl - -
1 f‘fi

Secondary Environment

Closed Mod. Closed Mod. Mod. Open Open

 

(N=17) (N=29) (N=83) (N=102) (N=10)

Orientation toward

Change

Z ACCEPT 5.9 13.8 12.0 26 5 20.0

Z MOD ACCEPT 47.1 41.4 32.5 32 4 30.0

Z MOD RESIST 29.4 34.5 36.1 29 4 30 0

Z RESIST 17.6 10.3 19.3 11 8 20.0
Liberalism

Z L0 5.9 10.3 9.6 13.7 0.0

Z MOD LO 58.8 48.3 44.6 34.3 40.0

Z MOD HI 35.3 37.9 30.1 35.3 40.0

Z HI 0.0 3.4 15.7 16.7 20.0
Personal Efficacy

Z HI 23.5 17.2 16.9 19.6 0.0

Z MOD HI 29.4 34.5 30.1 31.4 40.0

Z MOD LO 35.3 27.6 34.9 36.3 50.0

Z LO 11.8 20.7 18.1 12.7 10.0

 

category of the secondary environment and use only the more open and
the more closed run against the least and most liberal as a group
and the moderate levels (either moderately high or moderately low) of
liberalism. The relationship is significant at the .05 level (N a 158)
using the Yates correction for cases with individual cells having
relatively few observations. Subjects who are either the least or
most liberal seem to come from relatively open communicative relations
with their authorities at the present time in their lives (i.e. open
secondary socialization).

The final relationship in the table permits investigation of the

second hypothesis in this portion of the analysis--open secondary

101

socialization should relate positively to sense of personal efficacy.
Relating variables (using only closed and moderately closed subjects
against only the most Open subjects in comparison with high and.
moderately high efficacy and low, moderately low efficacy) we find a
Yates corrected chi-square of 2.746 (N - 56) which permits the acceptance
of the hypothesis since both the direction is as predicted and the level
of significance (one-tailed test) is beyond the .05 level.

Thus far we have confirmed our first two hypotheses at modest but
acceptable levels of significance, and can now discuss the variables of
continuity and level of education in relation to the personality attri-
butes.

After a presentation of the data on the socialization variables
of continuity and education the findings and examination of the relation-
ship between the four experimental socialization processes and resulting
personality attributes is summarized. The final portion of the chapter
is a discussion of the findings and their significance in terms of the

theoretical considerations.

Continuity, Educgtion and
Personality Attributes
Two hypotheses are tested in the following presentation. First,
individuals experiencing higher levels of discontinuity (more personally
felt crises) are likely to accept change. Second, education is associated
with greater acceptance of change. In addition, several other relation-
ships which are found among these socialization experiences and the

various personality attributes are examined.

102

First, continuity is related to the personality attributes.

TABLE l4.--Continuity by related personality attributes

 

 

 

Discontinuity
High Low
(N=69) (N=l70)

Orientation toward Change

Z ACCEPT 29.0 13.5

Z MOD ACCEPT 29.0 37.1

Z MOD RESIST 30.4 33.5

Z RESIST 11.6 15.9
Dogmatism

Z LO 21.7 16.5

Z MOD LO 40.6 31.2

Z MOD HI 20.3 34.1

Z HI 17.4 18.2
Reliance upon Authority

Z DEPENDENT 2.9 11.2

Z MOD DEP 17.4 29.6

Z MOD 24.6 26.6

Z MOD INDEP 39.1 23.7

Z INDEPENDENT 15.9 8.9
Liberalism

Z L0 7.2 11.8

Z MOD LO 31.9 45.9

Z MOD HI 42.0 30.6

Z HI 18.8 11.8
Personal Efficacy

Z HI 11.6 20.6

Z MOD HI 34.8 29.4

Z MOD LO 30.4 37.6

Z LO 23.2 12.4
Trust

Z HI 13.0 21.2

Z MOD HI 46.4 29.4

Z MOD LO 30.4 34.7

Z LO 10.1 14.7

 

103

We first relate degree of continuity and the predisposition to
accept change. The relationship is significant beyond the .05 level
and consistent with the hypothesis. Clearly, the subjects who respond
in the questionnaire as having some form of personally felt crisis are
more likely to accept change in personal and social situations.

Several other relationships exist between the extent of disconti-
nuity and the personality attributes. Subjects having discontinuity
are apt to be less dogmatic. This relationship is significant at the
.05 level when high, moderately high levels of dogmatism are combined
in one group and low, moderately low dogmatics in another and these
two groups run against high and low discontinuity.

Discontinuity relates to reliance upon authority, with a signifi-
cance beyond the .02 level. Those subjects experiencing some kind of
discontinuity are likely to be independent from their authorities, as
measured by their agreement with those authorities on several different
issues and opinions. The Kendall tau for this relationship is .235.

The three remaining relationships provide further information
on the effects of discontinuity upon personality. When low and
moderately low liberals are grouped and high and moderately high
combined and these groups run agains continuity we find subjects
experiencing discontinuity are likely to be liberal, significant at
the .01 level. There is some tendency for those experiencing disconti-
nuity to be personally less efficacious. As presented in the table,
the relationship is significant at the .10 level, and when those scoring
in the extremely high or low categories of efficacy are compared on

degree of continuity we find the relationship is significant at the

104

.02 level. Personally felt crisis tends to decrease the person's
sense of personal efficacy. When discontinuity and trust are related
it is significant at the .10 level which indicates some linear relation-
ship may be present, however closer analysis suggests another curvilinear
relationship. When the two groups (high, low trust subjects and mod.
high, low) are related with discontinuity, we find subjects with higher
discontinuity more moderate rather than extremely trusting or distrusting.
The last socialization variable we consider in this portion of
the analysis is the level of education in relation to the various
personality attributes. We hypothesize that education should relate
to the acceptance of change. Referring to Table 15, when level of
education is cross-classified with acceptance of change there are
important directions apparent in the data but as ordered they are not
significantly related. Reclassifying the data into two groups (college
and no college) we find a highly significant relationship (.01 level).
Those with college experiences are more accepting of change than those
with no college.
The strongest relationship in the table is between educational
level and sense of personal efficacy (significant at the .001 level),
but it is a rather complex and non-linear relationship. Those having
no college experience are much more likely to score as highly efficacious,
while those having one year of college are less likely to be efficacious.
Subjects with two years of college once again seem more efficacious while
there is no relationship between three years of college and efficacy.

This is difficult to interpret, but unless there is a systematic bias

105

TABLE 15.--Education level by related personality attributes

 

 

Education (Years of College)

 

No College One Yr. Two Yrs. Three Yrs.
(N=74) (N=66) (N=56) (N=43)
Orientation toward Change
Z ACCEPT 12.2 19.7 21.4 23.3
Z MOD ACCEPT 28.4 37.9 37.5 37.2
Z MOD RESIST 37.8 34.8 25.0 30.2
Z RESIST 21.6 7.6 16.1 9.3
Personal Efficacy
Z HI 29.7 9.1 19.6 7.0
Z MOD HI 20.3 25.8 39.3 48.8
Z MOD LO 35.1 51.5 23.2 27.9
Z LO 14.9 13.6 17.9 16.3
Liberalism
Z L0 20.3 10.6 3.6 4.7
Z MOD LO 40.5 40.9 41.1 44.2
Z MOD HI 28.5 34.8 37.5 37.5
Z HI 10.8 13.6 17.9 14.0
Reliance upon Authority
Z DEPENDENT 6.8 9.2 5.4 14.0
Z MOD DEP 31.1 23.1 25.0 23.3
Z MOD 35.1 21.5 30.4 11.6
Z MOD INDEP 18.9 30.8 32.1 34.9
Z INDEP 8.1 15.4 7.1 16.3

 

in these groups of which I am unaware, there is some tendency for

efficacy to decrease in college then to increase with more education

and finally the relationship disappears with still more education.

By comparing the least liberal group with all other and

classifying the level of education between no college, one year of

college and two or more years of college I find a relationship

significant beyond the .01 level (using Yates correction).

The

tendency is for those having more education to be more liberal.

106

Finally, there is a curvilinear relationship between education
and reliance upon authority significant beyond the .01 level. When
the following are compared: subjects having no college and those with
most college; subjects with most and least dependence upon authority;
and those moderately dependent and independent; there results a
tendency for those with three years of college to have greater reliance
upon their authorities. Those with more years of education are either
the most dependent or the most independent of their authorities while
those with no college are more moderate in their reliance upon authorities.
There are two further questions for analysis before proceeding to the
concluding discussion. What is the combined impact of the personality
attributes and the socialization experiences upon the orientation toward
change? What is the effect of consistency in socialization (e.g. the
four socialization patterns) upon the orientation toward change and
other personality attributes?

Sociglization Patterns, Personality and the
Orientation toward Chang;

We find in the previous chapter the combined impact of personality
attributes upon the orientation toward change explains some lOZ of the
variance in this predisposition. Now we can discuss the results of a
regression and multiple correlation analysis of both the personality
attributes and the socialization experiences using the predisposition
to accept change as the dependent or criterion variable. Table 16
presents the total correlations, beta weights and the tests of

significance for the beta weights (t—values).

107

TABLE l6.--Regression coefficients, zero-order correlations, and
t—values for orientation toward change related to socialization
experiences and personality attributes

 

 

Total correlations Regression

 

 

Variable with orientations Coefficients T—Values
toward change (beta weights)

Socialization:

Continuity .1384 .5892 1.3423

Primary Environment .0326 .0401 0.2842

Secondary Environment .1538 .2803 1.8304*
Psychological:

Reliance upon Authority .1036 .1216 0.9209

Liberalism .1471 .1140 1.7248*

Personal Efficacy .0065 .1038 1.2322

Dogmatism .2524 .0917 3.8910**

Trust .0749 .0481 0.6203

Political Cynicism .0537 .0245 0.3320

 

*Significant beyond .10 level

**Significant beyond .05 level

The total regression analysis is significant beyond the .001
level with an F ratio (df 9,231) = 3.5866. The multiple correlation
resulting from the analysis is .3502 which in turn has a coefficient
of determination of 12.3. The addition of the socialization variables
to the personality attributes increases the percent of variance explained
by about 2.5Z. It is clear that the personality attributes are the

more important predictor variables but it is also clear that the

108

socialization variables do add somewhat to our ability to explain this
relatively complex predisposition to accept change.

Considering both the beta weights and their tests of significance
we can conclude that only the variable of dogmatism produces a change
in the criterion which is significant beyond the .05 level. Both
liberalism and secondary environment produce changes significant at
the .10 level. The balance of the variables do produce some changes
but it is difficult to argue these changes could not have occurred by
chance alone. In any case the variables of dogmatism, liberalism,
and secondary environment and continuity appear to be the most important
in the analysis.

The second question to be investigated in this portion of the
chapter is: What are the effects of consistency upon the various
personality attributes, particularly the acceptance of change? There
are several specific hypotheses to be investigated in this portion of
the analysis. First, we hypothesize that if the type of socialization
experience has more influence than the variation in early and later
experiences, then we should find the greater differences in the
orientations toward change occurring between the consistently open
and consistently closed socialization patterns. Second, if instead,
dissimilarity between early and later experiences has more effect
upon personality than type of experiences then we expect to find the
greater differences in the orientations toward change occurring between
the consistent (i.e. both environments open or closed) and the incon-
sistent (i.e. one environment open and the other closed) socialization

patterns. We also have hypothesized that groups from open primary

109

and secondary environments should be more accepting of change than

those from closed primary and secondary environments. The balance of

the expected relationships between socialization patterns and personality
attributes are summarized in Table 1 (Chapter II).

To test these hypotheses I have utilized the more powerful
statistical model of testing differences in means by analysis of
variance rather than the more appropriate non-parametric statistics
previously in use in this analysis. There are reservations which must
be made explicit about meeting the assumptions of this analysis of
variance model as well as the regression model previously used. The
data in this study may not satisfy all assumptions of these models,
but for the present exploratory purposes it seems justifiable to use
these techniques to identify any findings which might otherwise be
lost to the less powerful, if more appropriate, non-parametric modes
of analysis.

I have divided the total population into four experimental
groups, each group representing one of the four socialization patterns.
Group 1 (N = 50) is composed of subjects scoring either in open or
moderately Open classifications on both the primary and secondary
environment variables. Group 4 (N = 21), on the other hand, includes
the subjects scoring either closed or moderately closed on both of
the environment variables. These two groups represent consistent
socialization patterns, but different types of experiences (i.e. group
1 has Open communicative experiences with authority throughout their
life to present, while group 4 has consistently closed and non-

communicative relations with their authorities).

110

Group 2 (N = 42) is made up of subjects scoring open or moderately
Open on the primary environment variable and moderate, moderately
closed or closed on the secondary environment variable. Group 3
(N = 29) is composed Of subjects scoring closed or moderately closed
on the primary relationship variable and Open or moderately open on
the secondary variable. These two groups represent the more inconsist-
ent socialization patterns. The second group represents those who
have moved from open to closed relationships with their authorities
while group 3 represents those who have moved from closed to open
communicative relations with authority. We expect these transitions
from one type of socialization experience to another to produce the
most acceptance of change in the individual's orientation. Those
who have made more adjustments or changes in their personality can
be expected to accept change in personal and social situations more
easily than those whose life path was unbroken.

Table 17 presents the means for the various personality attributes
within each of the four experimental groups. The table also includes
an F—ratio for each of the comparisons and the means for the total
group to give the reader some indication of the personality attributes
most strongly affected by the different types of socialization patterns
and the directions of the differences from the means of the total
group.

Table 17 indicates that the four socialization patterns produce
significant variations in the means for two of the personality attributes.
Reliance upon authority and the orientation toward change result in

differences in means among the four groups which could not have-arisen

111

TABLE 17.--Analysis Of variance within personality attributes and
between experimental socialization patterns

 

 

Means for Experimental Groups

Group 1 Group 2 Group 3 Group 4 Total F—Ratios

 

Variables (N=50) (N=42) (N=29) (N=21) (N=142)(df=3,l38)
Dogmatism 61.82 64.17 61.79 60.33 62.69 0.9
Liberalism 9.34 8.93 9.59 8.10 9.09 1.1
Political

Cynicism 10.86 10.74 11.55 10.52 10.92 0.6
Personal

Efficacy 6.96 7.05 6.52 7.86 7.03 1.1
Trust 8.18 8.55 9.35 8.86 8.63 1.3
Reliance upon

Authority 2.94 2.91 4.24 3.67 3.30 5.1*
Orientation

toward Change 13.68 14.79 13.52 15.34 14.23 2.3**

 

*Significant at the .01 level

**Significant at the .10 level

by chance more than one time in a thousand and one time in ten,
respectively. Before analyzing these findings in relation tO the
theoretical model in chapter two it is useful to show the t—values
for the comparison of means among the four experimental groups.

.The F-ratio indicates some degree of variation among the four groups
while the t-value indicates which means are contributing most to

the total variance measured by the F test.

112

TABLE 18.--Matrix of t—values resulting from the comparison of means

of personality attributes among the experimental groups

 

 

 

Group 1 Group 2 Group 3 Group 4
(0X0) (OXC) (CXO) (CXC)
Group 1
Dogmatism — 1.1762 .0365 .6346
Liberalism - .6695 .3159 1.5856
Political Cyn. - .2433 .9157 .4744
Personal Eff. - .1334 .8333 1.1544
Trust - .7597 1.9075 .9857
Reliance upon
Authority - .2470 3.2073 1.6678
Orientation
toward Change - 1.6369 .2164 1.7562
Group 2
Dogmatism - - .9854 1.5274
Liberalism - - .8742 1.0906
Political Cyn. - - 1.1377 .2742
Personal Eff. - - .9908 1.0681
Trust - - 1.3305 .4586
Reliance upon
Authority - - 3.2441 1.7810
Orientation
toward Change - - 1.9668 .7614
Group 3
Dogmatism - - — .4912
Liberalism - - — 1.6013
Political Cyn. - - - 1.3128
Personal Eff. - - - 1.6325
Trust - - - .5858
Reliance upon
Authority - -- - 1.0189
Orientation
toward Change - - - 2.0300

 

Although only two personality attributes show significant

variation among the four experimental groups, we can discuss each
attribute in turn to examine the extent of variation from the

theoretical model proposed in Chapter II (Table 1). To facilitate

113

the comparison of Observed and expected classifications of personality
attributes for the experimental groups I have constructed Table 19
comparing the predicted classifications of each attribute with that
level actually obtained.

TABLE l9.--Comparison of predicted and observed classifications of
personality attributes for each of the four experimental groups

 

 

 

Group 1 Group 2 Group 3 Group 4

Variables (Pred./0bs.) (Pred./Obs.) (Pred./Obs.) (Pred./Obs.)
Dogmatism (LO/MOD HI) (MOD HI/HI) (MOD LO/MOD LO) (HI/LO)
Liberalism (HI/MOD HI) (MOD LO/MOD LO) (MOD HI/HI) (LO/LO)
Political

Cynicism (LO/MOD HI) (MOD HI/MOD LO) (MOD LO/HI) (HI/LO)
Trust (HI/HI) (MOD LO/MOD HI) (MOD HI/LO) (LO/MOD LO)
Personal

Efficacy (HI/MOD HI) (MOD LO/MOD LO) (MOD HI/HI) (LO/LO)
Reliance

upon

Authority (LO/MOD HI) (MOD HI/HI) (MOD LO/LO) (HI/MOD LO)
Orientation

toward (ACCEPT/ (MOD RES/MOD (MOD ACC/ (RESIST/

Change MOD ACC) RES) ACCEPT) RESIST)

 

A review of the table suggests that the theory holds for the
variables personal efficacy, liberalism, trust and orientation
toward change. Reliance upon authority, dogmatism and political
cynicism appear to depart from our expectations so substantially as
to require basic reformulation of that part of the theoretical

sketch. Further, we must keep in mind the differences we have

114

found in the orientation toward change not only conform relatively
well to the theory, but are significant at the .10 level. We also
find significant differences in the reliance upon authority
(significant beyond the .01 level). This statistical significance
requires a re-examination of the theory in light of the existing
pattern of relationship between the socialization processes and
reliance upon authority, since the variable obviously does relate
but not in a manner consistent with the theory. We will discuss
each personality attribute in turn to gain insight into the useful-
ness of the theory and ways it may be reformulated to more closely
account for these observations.

When we examine dogmatism we find no differences among the means
Of the groups which are not reasonably attributable to chance alone.
But it may be of theoretical importance that lower levels of dogmatism
appear to result from the third and fourth groups which share closed
primary socialization experiences with authority. It also seems
important that group 2 has the most dogmatic subjects and group 4
the least. Political cynicism fares little better than dogmatism.
Here the only point of interest is the relatively large differences
between groups 3 and 4. Group 3 has the most cynical subjects while
group 4 has the most trusting subjects, but again the differences
may be due to chance alone.

Reliance upon authority departs from theoretical expectations,
but the differences in these means requires a closer look at the results.
We have found lower degrees of dogmatism in group 3 and 4 and also

subjects who are independent Of authority in the same groups. We find

115

no difference between group 1 and 2 or between 3 and 4 which could
not have arisen by chance alone, but we do find more significant
differences between groups 1 and 3, l and 4, 2 and 3, and 2 and 4.
The differences suggest a complete revision in the theory. Group 1
produces more dependent subjects than either 3 or 4 and the differences
are significant. Group 2 produces more dependent subjects than 3 or 4
and again the differences are not readily attributable to chance. It
appears the first two socialization patterns (both environments open;
and primary Open and secondary closed) tend toward a greater reliance
upon authority while the latter two processes (closed primary and
Open secondary, and both environments closed) produce more independent
subjects. More specifically, group 3 produces the most independent
while group 2 produces the most dependent subjects. Since both
processes involve inconsistency we can only deduce inconsistency does
not contribute to reliance upon authority. We can suggest early
family situations which are characterized by more open communications
between the authorities and the child appear to develop individuals
who are more rather than less dependent upon authority, while the
reverse happens in early family situations with little communication.
It seems that inconsistency tends to extend the effects of the early
family experience rather than attenuate them. The implications of
this for the theory are great, but it also raises the question of
instrument validity. This can be discussed in the final portions of
the chapter.

The balance of personality attributes appear to conform more

closely to the theory. More liberal orientations result from groups 1

116

and 3, group 3 (inconsistency) having the most liberal subjects. The
less liberal orientations result from group 2 and 4 with group 4 having
the least liberal subjects. The only discrepancy from the theory

here is pattern 3 which appears to produce a more liberal group than
does the first process. Again, inconsistency extends the effects of
early childhood experiences rather than reducing them. Personal
efficacy results in exactly the same pattern as liberalism, with

high efficacy in group 3 and lowest in group 4. The variable trust
departs a bit further from expectations. We predict high trust for
group 1 and find that to be the case, low trust for group 4 and find
only moderately low. Groups 2 and 3 reverse our expectations. Group 3
has more cynical subjects than we expect while group 2 has less. The
only difference here that appears statistically important is that
between 1 and 3. The first process does produce more trusting subjects
while the third process produces less trusting subjects.

Finally, the orientation toward change conforms well to our
expectations with the only exception being the third group which is
higher in acceptance than group 1. Again the inconsistency produces
a more pronounced effect than the more consistent processes. Several
of the differences we find here approach statistical significance.
There is little difference between 1 and 3 or 2 and 4 but there is
substantial differences between 1 and 2, and 1 and 4, 2 and 3 and 3
and 4.

With these comments on the findings and the theory complete we
can test the several specific hypotheses constructed in the theoretical

chapter.

117

We have hypothesized groups from more Open primary and secondary
environments should be more accepting of change then those from more‘
closed environments. When we compare group 1 and 4 substantial support
for this speculation is found. Group 1 (both environments open) is
more likely to have subjects accepting change than group 4 and the
difference is significant at the .05 level using a one-tailed test
of significance appropriate to a hypothesis predicting direction.

Finally, we have attempted to.construct a critical hypothesis
which permits the investigation of two broad conceptual approaches
to personality develOpment. If the type of socialization experience
has more influence upon personality than the variation in early and
later experiences then we should find the greater differences in
the orientations toward change occurring between consistently open.
and consistently closed socialization patterns. If instead, dis—
similarity between early and later experiences has more effect upon
personality than type of experiences, we expect to find the greater
differences in the orientation toward change occurring between the
consistent (i.e. both environments open or closed) and the inconsist-
ent (one environment open and the other closed) socialization
patterns.

As one might expect, the critical hypotheses do not yield the
desired results. We do in fact find significant differences in the
means of orientations toward change for the consistently open and
consistently closed socialization patterns, which confirms the first
hypothesis. However, we also find one of the processes involving

inconsistency, group 3, produces more acceptance of change than the

{the more open and communicative the environment, the more likely

118

consistently open environment pattern. On the other hand, the other
group involving inconsistency produces less acceptance of change

than any other group except the consistently closed pattern. Our
second hypothesis is partially confirmed and partially rejected.

When the individual moves from open to closed socialization experiences
with authority he becomes less accepting of change, but when he moves
from closed to Open.experiences he becomes more accepting of change--

more so than those with consistently Open experiences.

Summagy

We have confirmed several hypotheses in this section of the
data analysis. One Of the first hypotheses investigated has been
developed to test the notionfthat early communications with authorities
in the family situation is likely to develop a trusting orientation
toward authority generally.) A significant relationship between Open

primary environment and the variable of political cynicism is found--

the individual is to trust politics and politicians: {A second
hypothesis has been used to test the idea that open communication

with authorities in the secondary environment (i.e. outside of the

n
L

immediate family and later in life) is related to the individual's

sense of personal efficacy} We again find a relationship which supports
this thinking--the more Open the sztondary environment, the higher the
sense of personal efficacy. It-is important to note that no relationship

between secondary environment and trust or between primary environment

and efficacy is present.

119

aFurther, individuals with greater degrees of discontinuity .i J?.
(i.e. personally felt crisis) in socialization should be more likely
to make personality adjustments and therefore more likely to accept
change in social and personal situationsh Discontinuity relates
positively to acceptance of change, confirming the hypothesis. We
also expect those with more education to be more likely to accept
change. When we contrast individuals having no college and those
with some college education a significant difference, which is
consistent with this hypothesis, results.

Before discussing the several hypotheses concerning consistency
it is necessary to consider some of the Observed relationships which
do not conform to expectations.

Primary socialization environment relates to reliance upon
authority, liberalism, and dogmatism. {Inhividuals emerging from
early family situations characterized by more communication and
interaction with their authorities are more likely to be dependent
upon authority, while those from closed and non-communicative early
environments are more apt to be'libera19' Evidently, the least

X/dogmatic subjects are likely to come from either most open or most.
closed primary environments. {People.from open secondary relations
are more probably accepting of change and either most or least
liberal in their politicséiiIndividuals with greater discontinuity in
socialization are less dogmatic, less dependent upon authority, liberal,

less efficacious, and moderately trusting (rather than high or low in

trust) toward other peOple. Finally, education tends to increase

iv

120

liberalism and the probability of being either extremely independent
or extremely dependent in authority relationships.

/Communication with authorities in early family situations has
several lasting effects.) Individuals having more communication are
more apt to remain dependent upon these authorities for their Opinions
later in life, they are less likely to be liberal and they are.less
dogmatic, although, closed and non-communicative relationships may
also result in less dogmatic persons.

(Those subjects who have established greater interaction and,
communication with authorities in their later life are more likely
to accept change, but are-either extremely liberal or extremely
conservative in their political views) In this case a conservative
political orientation may be accepting of change. The presence of
open secondary interaction and communication evidently permits the
individual a clearer definition of his political views, as evidenced
by these groups being more extreme and less moderate in stating their
political Opinions. But in both cases they are more able to adjust
toishense -

Discontinuity and education tend to make people liberal.

/

KPerceived crisis decreases the individual's sense of efficacy and

makes him only moderately trusting of others, but in both cases he

is still accommodative to change; Education also has the effect of
making the individual more or less dependent upon his authorities.

Again this may be due to the fact that those with education have

clearer ideas about their relationship with their authorities and are

121

therefore able to indicate either agreement or disagreement with the
opinions of their authorities.’

Finally, we have investigated the effects of consistency in
socialization upon the several personality attributes. {Those
subjects having consistently open socialization are more accepting
of change than those with consistently closed socialization, which
is consistent with another of our theoretical notions.) We have also 9
tested the idea that either the type of experience or contrast
between early and later experiences has the greatest effect upon
personality. We find both of these factors to have considerable
effect upon the acceptance of change as one orientation which is
investigated under both conditions. (The type of experience (i.e.
communicative relations with authority) appears to increase the
acceptance of change, however, if the subject moves from a closed-
environment to an Open environment he is even more likely to accept
change Inconsistency is not the important variable since those.
experiencing inconsistency in the opposite direction (i.e. open to
closed environments) are least likely to accept change. In conclusion,
if the type of experience in socialization is communicative then the
person is likely to be accepting of change, and if the socialization
process involves inconsistency from closed to open relationships then
the effects of the secondary.environment are extended or made more.
pronounced. (That is, if the inconsistency results in an open secondary
environment the subject is more accepting of change, whereas, if the
inconsistency results in a closed secondary relationship then the

individual is least likely to accept changeg> Obviously both the type

122

of experiences and the inconsistency between early and later experiences
are important, and it appears inconsistency.works to emphasize the
gffggts of the type of experiences existing presently in the person's
lifg.

The last chapter summarizes the findings of this study and uses
the data in the reformulation of the theoretical sketch guiding this

present study.

CHAPTER VI

EVALUATION OF THE THEORY

Review of Expectations and Observations

Three questions will guide this review of theory and data.
These are: (1) What are the effects of early socialization upon.
personality attributes? (2) What are the effects of later sociali-
zation upon personality attributes? (3) What are the effects of
different patterns of primary and secondary socialization upon the
personality variables under consideration?

We have theorized early authority relationships vary between
Open, communicative and closed, non-communicative interpersonal
contact between the child and his authorities. If early relations
are relatively open the child learns to relate to authority with
more information and understanding of both the authority as a person
and what the authority communicates about himself and the world.
This type of authority relationship generates less reliance upon
authority, as authority, and more dependence upon what the authority
communicates. The dependency upon the authoritative communication
rather than the authority Object contributes to an ability to
distinguish messages from the source of the message--the essential
characteristic of a less dogmatic personality. Further, these early

Open relationships with authority lead to a more trusting orientation

123

124

toward both authority and other peOple generally since early contact
in previous experiences has proven both to be trustworthy. This is
the first part of our theoretical sketch, and the relevant analysis
is now considered.

Primary socialization relates with reliance upon authority,
liberalism, political cynicism (although not with general trust
of others), and dogmatism. But the relationships we find are not
consistent with the above. First, we find subjects who report their
early relationships with authority as being more communicative tend
to be more dependent rather than less dependent upon their present
authorities for opinions about various social and political issues.
Further, these same subjects (i.e. with more Open primary socialization)
are more likely to be less dogmatic, but, those subjects from closed
primary relations are also less likely to be dogmatic. In fact, the
subjects from moderate classifications of openness and closedness
appear to be the most.dogmatic subjects in the population. That
early communicative relations engender trust.of generalized authority
is supported to some extent. The politically most trusting subjects
tend to have more open primary socialization. However, no relationship
is seen between primary environment and the general trust of others.
Subjects classified as more liberal in political values have closed~
primary relations. Obviously, such findings require reformulation
of the present theory, and this is to be the central concern following
this comparison of theory and data. The next question is--What are

the effects of secondary socialization upon personality?

125

The only major impact of secondary socialization we expect
is in regard to the attribute of personal efficacy. More Open
communications in secondary authority relations should lead to a
higher sense of personal efficacy. There is modest support of this
expectation. It is noteworthy that efficacy.does not relate to
primary socialization but does.relate to open secondary relations,
while political trust relates to open primary experiences but not
to secondary experiences..

In addition, open secondary experiences with authority relate
substantially with acceptance of change,.and subjects who are most
conservative (i.e. least liberal) and most liberal are.apt to be
classified as having open socialization experiences at the present
time in their lives.

Two other major socialization variables are appropriately
discussed at this point--degree of continuity and level of education.
We have hypothesized that people experiencing more crisis situations
in their lives (i.e. greater discontinuity) are most likely to have*
made-psychological changes and adjustments to these demands, and
are in turn, more likely to accept change in the external world.
Those in our study who are classified as experiencing discontinuity
are significantly more likely to accept change than those without
discontinuity. Perhaps it is even more interesting that we find
the continuity relating to most of the personality attributes.

There is evidence to support the following propositions concerning
perceived crisis and personality. Those who report more crisis in

socialization are also: (1) less dogmatic, (2) more independent of

126

authority, (3) more liberal in political beliefs, (4) personally
less efficacious, and (5) moderately trusting of others rather than
extremely trusting or distrusting.

We have further speculated that those with more education
are likely to relate well with authority and therefore can be more
accepting of change. The findings show subjects with college to
be more likely to accept change than those without college, and there
are several other relationships involving level Of education.

There appears to be a most complex relationship between sense
of personal efficacy and level of education. I am uncertain whether
this is a genuine finding or an artifact of the particular groups
used in this analysis. We find strongest evidence of high personal
efficacy among those with no college, one year of college substantially
reduces the number of efficacious people, a second year of college
increases efficacy once again, and there is no relationship between
efficacy and education in the group with three years of college.

A tendency exists for those with more education (i.e. two or
more years of college) to be more liberal. Those with most years
of college education are likely to be either most dependent or most
independent of authority for their opinions.

Before examining the last question guiding this evaluation,
it is apprOpriate to summarize briefly data concerning relationships
among the variOus personality variables independent of the socializa-
tion experiences.

Less dogmatic people tend to be more efficacious and more

trusting of peOple as well as politics. There is some indication

127

that efficacy relates positively to conservative political views.
The second structural attribute of personality, reliance.upon.
authority, relates to liberal political values and distrust of
peeple and cynicism about politics. Evidently both least and most
dogmatic subjects are likely to be independent of their authorities.
We have confirmed two hypotheses. Those who are less rigid
internally (e.g. less dogmatic) are likely to be less rigid about
external situations as well (i.e. more accepting of change). Also,
modest support exists for the idea that those who are independent
of authorities external to a situation are accepting of changes
in these situations. The third theoretica1.expectation finds little
support in this analysis. We have speculated that both efficacy
and trust in people relate to liberal beliefs, and liberals are more
accepting of change, so we expect efficacy, trust and liberalism to
positively relate to acceptance of change. .However, efficacy does
not relate to acceptance of change and there is a tendency for
conservatives to be more efficacious. Moderately trusting subjects
are more.like1y to accept change than either extremely trusting or
or distrusting subjects, and finally, liberals accept change but
are.a1so likely to be most resistant to it.

In regression analysis, dogmatism, liberalism and independence
of authority contribute most to the orientation toward change. The
contingency analysis suggests moderate levels of trust are also
important to the acceptance of change. The major conclusion of
this part of.the analysis is that the several psychological attributes

explain most of the variance in our predisposition to accept or resist

128

change. With these comments on relationships among the personality
variables and the orientation toward change, the final question is
discussed--What are the effects of the four socialization patterns
upon the various personality attributes?

Previously, the separate impact of primary and secondary
socialization upon personality has been considered. The present
discussion reviews the portions of the theory concerning inter-
dependencies among primary, secondary socialization experiences and
personality. We have argued that the child as he matures and
interacts with people.beyond the immediate family may find his early
experiences either reinforced or, on the other hand, later experiences
may be quite dissimilar and non-reinforcing.' Such dissimilarity or
inconsistency in experiences between primary.and secondary relations
requires personality adjustments and change. However, if the
secondary experiences reinforce earlier learning then little change
in personality can be expected. Also, when.socialization is more,
consistent the type of experiences are most determinative of later
personality, but when the socialization process is more inconsistent
the succession of early and later experiences may be critical in
determining adult personality attributes, specifically the orientation
to accept or resist change. Four socialization patterns result from
the logically possible combinations of the two dichotomized variables
of primary and secondary environments. Dividing our total population
into four groups corresponding to these theoretical patterns, this
portion of the theory has been tested. Group 1 has open primary

and secondary socialization experiences. Group 2 has primary open

129

but secondary closed. Group 3 has primary closed and secondary
open. And group 4 has both environments closed.

Each of the four socialization patterns, as represented by
the experimental groups, is discussed in terms of predicted and~
observed personality attributes resulting from each of the
processes.

The first process is composed entirely of open and communica-
tive relations with authority. Since this process represents
consistent socialization the characteristics of early childhood
relations will be reinforced and stable. Therefore, the first
process should produce subjects who are less dependent upon
authority, less dogmatic, more trusting,.and, since the open
secondary relations.should contribute to a higher sense of personal
efficacy, the subject should also be more efficacious. Further,
the subjects should be liberal since their.early personal relations
with authority have been open.and less likely to generate a general-
ized fear about authority. Because of the greater independence of.
authority as an object the subjects can more readily accept change.
in a particular situation.

It must be noted that the four socialization patterns (i.e.
experimental groups) produce significant variation in only two of
the several personality attributes--the orientation toward change
and reliance upon authority. The balance of this discussion is for
the most part based on findings in the data which may well be attri-
butable to chance alone. I believe there is justification in

discussing such admittedly weak evidence on the basis of the

130

existence of predicted directions which for these exploratory
purposes can be sifted carefully for insights into the present
problem.

The first experimental group contains subjects who are moderately
accepting of change and moderately high in their dependence upon
authority. These subjects tend to be moderately high in dogmatism,
high in trust (moderately high in political trust) and moderately
high in personal efficacy and liberal values. Although we are not
concerned with the reformulation of theory here, it is appropriate
to point out the impact of primary experiences upon personality.
These data serve to guide the following reformulation.

Open primary environment tends to produce more dependent, less
liberal, more politically trusting and less dogmatic subjects. The
effects of primary environment are relatively stable when socializa-
tion is more consistent. Dependence upon authority and trust remain
constant, but extent of liberalism increases as a result of the
secondary experiences. Dogmatism also shifts from more to less in
this socialization process, but again, these findings may be due to
chance alone.

This portion of the theory is supported by the outcomes for
the variables trust, efficacy and orientation toward change. The
theory does not do well for dogmatism, reliance upon authority or
political cynicism and liberalism. Reliance upon authority so severely
departs from the theory that it must be considered in detail in the

following portion of the chapter.

131

The second socialization pattern under,consideration involves
early and later experiences with authority which were closed and non-
communicative. It has been our thinking that if the child grows up
in such non-communicative relations he is more likely to be dependent
upon authority, less able to differentiate the authority from what
the authority says (i.e. more dogmatic), and, because of the early
and arbitrary experiences with authority less trusting of others
generally. Closed secondary relations should lead to low sense of
efficacy. Such a person may adopt more conservative attitudes
since his experiences with authority generate fear and anxiety
about authority (e.g. strong central government). And because he
relies upon the authority of tradition, custom and usual ways of
doing and thinking, he is less likely to be able to accept change.

We find, in fact, people classified as having consistently
closed experiences (group 4) are moderately low in their dependence
upon authority, low in dogmatism and moderately low in trust of
others. They are low on the scale of liberal values, low in trust
but also low in political cynicism, low in personal efficacy and
finally, highly resistant to change. Again, reliance upon authority,
dogmatism and political cynicism depart from expectations. Whereas,
the variables of trust, efficacy and orientation toward change are
relatively consistent with the theory. Liberalism appears to change
since closed primary relates to liberalism, but consistently closed
socialization tends to produce more conservative subjects.

The second two socialization patterns (represented by groups 2

and 3) involve the variable of inconsistency in socialization. These

132

patterns are expected to result in a greater likelihood of personality
adjustment. We have hypothesized people making such adjustment internal—
ly are more likely to accept change in the external world as well.
Further, the changes in personality are expected to be away from the
orientations formed in childhood and toward those predispositions
expected as a result of the type of secondary environment. First,
then, the socialization pattern involving the transition from closed
childhood relations to Open secondary experiences with authority is
discussed.

We expect the early non—communicative relations of this process
to produce more reliance upon authority, less ability to differentiate
source and message (more dogmatic) and less trust in other peOple.
But because of the adjustments required by the dissimilar secondary
experiences we expect an attenuation of orientations formed early.
The individual should become less reliant upon authority, less dogmatic,
more trusting of others, and in addition, he should develop a greater
sense of efficacy from the open secondary environment. He should be
more liberal and more accepting of change for the same reasons discussed
in relation to the first process. This hypothetical set of personality
attributes are much the same as those resulting from the consistently
Open process, but we expect the various orientations to be less
extreme than those from the open process because of the changes
required by the transition.

In fact, the results from this socialization process are more
extreme than the consistently open process. The group is character-

ized by moderately low dogmatism, high liberalism, high cynicism, low

133
trust, high efficacy, high independence of authority and the most
acceptance of change of any of the four groups.

Generally our predictions of personality orientations from this
particular process are sound. We have predicted the correct direction
(e.g. high or low) for dogmatism, reliance upon authority, liberalism,
efficacy, and orientation toward change. We missed most clearly on
the variables of trust and political cynicism. The inconsistency
does have the theorized results but as will be seen later, it is a
case of the results being right and the reasoning wrong.

Trust is the most important departure from the theory and these
people (group 3) are distrusting or politics and people. This appears
to be a result of the primary environment which is not affected by the
inconsistency in socialization to an open secondary environment. Evi-
dently, trust, both political and general, is more resistant to change
than the other variables used in the present analysis.

The last socialization pattern involves the transition from open
and communicative primary relations to closed and non-communicative
secondary relations. The open early experiences are expected to produce
less reliance upon authority, less dogmatism, and more trust. The
transition to a closed secondary environment should make the person more
reliant upon authority, more dogmatic, and less trusting. Additionally,
the closed secondary environment should reduce the sense of efficacy,
and increase the likelihood that the person is conservative and
resistant to change.

The experimental group representing this process is characterized

by high dogmatism, moderately low liberalism, moderately low cynicism,

134

moderately high trust, moderately low efficacy, high dependence upon
authority and moderate resistance to change.

Correct directions are predicted for the variables of dogmatism,
liberalism, efficacy, reliance upon authority and orientation toward
change. We have had poorer results with cynicism and trust.

We have compared the findings with the theory in this portion
of the chapter and now proceed to a reformulation of the model from
these findings and a discussion of the major conclusions. It is
necessary to remind the reader that all of the attributes except the
orientation toward change and the reliance upon authority being
discussed in terms of the predicted directions could have resulted
from chance alone. We feel justified in discussing these directional
differences even though they are not statistically significant in all
cases as a basis for evaluating an exploratory theory. We are going
to consider the logic of our model which has led to relatively good
predicitions for several of the variables, but has led to many

unsupported hypotheses along the way.
Discussion and Reformulation

To present I have evaluated the theoretical model by presenting sta-
tistically confirmed propositions for each of the theoretical expectations.
However, I have initiated this study with an intuitive sketch based on
bits and pieces of a literature review concerning the effects of
socialization and personality attributes upon the predisposition to
accept change. At this point it seems feasible to relax this analytic

posture and stray from statistical analysis toward a more interpretative

135

reformulation of the present model in the presence of findings and
considered speculation. The weaknesses and strengths of the data have
been made explicit, and although the following interpretations are
mine they can be measured against objective criteria. I offer the
following as much from the research gestalt of my own effort as from
statistical evidence with the goal of constructing a closer approxi-
mation of personality formation and change and as a basis for further
and less exploratory efforts.

First to be discussed is the variable, communication with
authorities in both early family and later life situations. We thought
that if the child's early experiences with authority were open and
communicative the subject would learn to respond to authority with
more information and understanding of the authority and what the
authority communicates about the world. These interpersonal relations
should generate less reliance upon authority as an object and lead
to more dependence upon what the authority communicates about himself
and the world. This growing dependency upon what is communicated
rather than who communicates should in turn lead to a greater ability
to differentiate messages from sources of messages and thus to a less
dogmatic personality.

When this speculation is tested, we find individuals from more
Open primary environments are more dependent upon their authorities
for opinions about various issues, and subjects from closed environ—
ments are the more independent of these opinions. In other words, we
find the opposite of what we expect and it is strongly supported in

the analysis. Consider alternative explanations. First we need to

136

question the assumption underlying our operationalization of reliance
upon authority. We have assumed that subjects who agree with their
authorities on several different kinds of issues are more dependent
upon authority, and disagreement with these authorities indicates
independence of authority. Perhaps this assumption is not warranted
and we are measuring the extent a subject does depend on what is
communicated (e.g. adult Opinions) rather than adult authority, qua
authority. But let us examine the evidence a bit more closely.

Low dogmatism is apparently characteristic of subjects from
both most open and most closed primary relations (e.g. a curvilinear
relation). We have subjects who are less dogmatic and tending to
agree with their authorities later in life and others who are also less
dogmatic but who tend to disagree with authorities in later life.
Assuming dogmatism is affected by the ability to evaluate messages
separately from source, it appears that there are two alternative
results of being able to make the distinction. Those from open early
relations make this distinction and evidently are still able to agree
with their authorities while those from closed primary relations make
the distinction and form Opinions more at variance with their authorities
later in life. The logic leads to the conclusion that the assumption is
in error. If the low dogmatic does differentiate messages from source
and therefore evaluates events and issues unencumbered by his evaluation
of the source, we should not expect to find this systematic relationship
between reliance upon authority (i.e. agreement with authority) for one
kind of low dogmatic and not for the other. In other words, we would

expect a higher correlation (i.e. linear relation) between reliance

137

upon authority and dogmatism (which we do not find), or no linear
relationship between reliance upon authority and primary environment
(which we do find). In fact, it is the moderately dogmatic subjects
who are least reliant upon authority while the extremely high and low
dogmatics are the more dependent. Low dogmatism in this study
apparently does not relate to the ability to differentiate between
source and message and evaluate them separately. Low dogmatics from
Open family situations tend to agree with the opinions of their
authorities while low dogmatics from closed family situations tend to
disagree with their authorities. This raises the question of what
does underlie the intolerance and authoritarianism represented by the
dogmatism variable if not the ability to distinguish authorities and
messages and evaluate them separately.

I suggest that the low dogmatics appear to have stronger Opinions
either disagreeing or agreeing with their authorities while higher
dogmatics are less certain about their position in regard to their
authorities. Low dogmatics appear to be the more opinionated, or at
least they have a clearer understanding of their own opinions and those
of their authorities.

Early relations with authority which are more communicative seem
to produce less dogmatic subjects who are likely to agree with authori-
ties in later life. If early relations with authority are more closed
and non-communicative the subjects tend to be less dogmatic and more
independent of their authorities' opinions later in life. Moderately

Open and closed relations in the primary environment tend to produce

138

dogmatic subjects who are evidently no more or less likely to agree
with authority Opinion.

It may be that the subjects with close, interpersonal communica-
tions with early authorities form clearer impressions of their
authorities' opinions and are therefore more able to accept the
opinions of others when they are involved in similarly Open relations
in later life. The subjects from closed early relations do not have
any communicated understanding of authority Opinions. They are aware
of dictates from authoritative sources and tend to reject the opinions
of any authority which is seen as an arbitrary source of directives
and not as a source of information and understanding. This would
help explain how peOple from these two different primary environments
both have clearer definitions of their opinions, either to agree or
disagree, in relation to those of their authorities. In one case they
agree because they understand and accept, and in the other case they
disagree because they reject the authority source. This suggests that
dogmatism may measure intolerance and authoritarianism, but it has
little to do with the ability to distinguish and separately evaluate
source and messages from source. Some are more tolerant and less
authoritarian because they have a better understanding of authority
and others are more tolerant because they tend to reject authority
generally. Those who have no clear impressions of authority, either
positive or negative, are more likely to be dogmatic (i.e. intolerant
and authoritarian).

Other effects of early communication with authorities must be

considered. Those from open primary relations are more trusting of

139

of politics and politicians, and the reverse is true of those with

closed early relations. The most important point here is that this

trust of politics is not affected by later.experiences with authority.
Those with Open primary environments whether they continue to have

open relations or not are more trusting later in life. Those with

closed early relations are more cynical whether the secondary experiences
change or not. Political trust evidently is formed from open inter-
personal relations in early life and is relatively stable throughout

the subjects' life regardless of later experiences.

We also find those with open primary relations tend to be more
conservative (e.g. less liberal), although this orientation does seem
to be more affected by later secondary experiences.

Open communication in the family tends to produce trusting views of
politics, less liberal values, dependence upon authorities and less
dogmatism. Closed early relations tend to produce less trusting
political orientations, liberal beliefs, independence of authority
Opinion and again, less dogmatism.

The other variables which seem to be affected by the secondary
environment independently of early environments are personal efficacy,
liberalism, and acceptance of change. Those with open relations in the
secondary environment regardless of the type of primary experiences are
efficacious, most liberal or most conservative (another curvilinear
relationship) and accepting of change. This indicates that later life
experiences can have substantial effect upon personality beyond the

impact of early family situation.

140

With the above findings as a basis for the reformulation of our
theory concerning the effects of early and later experiences, the results
of the variable of inconsistency in socialization are observed. The
only personality attributes that vary significantly over the four
hypothetical socialization patterns are reliance upon authority and
acceptance of change. We can use these orientations for a partial
test of the consistency hypothesis.

The patterns of open primary and open secondary should produce
subjects who are most like what would be expected from early childhood
experiences. We find that peOple from consistently open socialization
are moderately high in reliance upon authority and moderately accepting
of change. They are dependent upon authority from their primary
experiences and remain so throughout their secondary experiences. The
acceptance of change results from the open secondary experiences, since
we do not find any relationship between acceptance of change and
primary environment. The consistently closed socialization pattern
produces moderately low reliance upon authority and the greatest
resistance to change. Again we find the early effects of the primary
relations upon reliance upon authority are sustained and the resistance
to change develops from the secondary experiences.

When we investigate the process involving inconsistency we find
those with open primary relations (who should be dependent) are highest
in dependency, and those with closed primary relations are most inde-
pendent in later life. No changes have occurred but the effects of the
inconsistency are to increase the intensity of the early formed orienta-

tions, at least for this one orientation. The impact of the secondary

141

experiences are consistent with our expectation that an open secondary
environment increases acceptance of change. Those with open primary and
closed secondary relations are moderately resistant to change while the
group with closed primary and open secondary environments are most
accepting of change. We find some evidence that liberals come from
closed primary relations but with inconsistency (i.e. closed to open
environments) the subjects are most liberal and when the process is
consistent (i.e. closed to closed) the subjects are least liberal.

This suggests that consistency tends to change the orientation from
liberal to conservative while inconsistency tends to increase the
effects of early childhood relations. Both of the suggestions must

be considered in the concluding statement. Before we consider this we
need to discuss two more socialization variables--continuity and level
of education.

Discontinuity (e.g. experiences of crisis or trauma in the
socialization process) is expected to bring about personality adjust-
ments and these internal adjustments are expected to permit individuals
to be more likely to accept change in the external world as well.
Further, higher levels of education are expected to increase the
acceptance of change. Both discontinuity and more education tend to
increase the acceptance of change. Discontinuity contributes to lower
dogmatism, independence of authority, more liberalism, and reduced
efficacy, while it appears to associate with moderate levels of trust
toward others. Those with more education are more liberal and likely
to be either most or least dependent of authority for opinions at the

present time in their life.

142

We now attempt to summarize these conclusions and offer some
general thoughts for further analysis.

When the child has open and communicative interpersonal relations
with authority in childhood he forms some orientations which are
relatively stable and persist throughout life regardless of later
experiences. For example, trust of politics relates to open primary
environment and cynicism relates to closed primary environment regard-
less of the type of secondary experiences. With open primary experiences
the individual remains trusting of politics even though his secondary
experiences may be closed; the reverse is also true if the early
experiences are closed the subject remains distrusting even if the
secondary experiences are open.

Other orientations appear to develop as a result of later life
experiences. The orientations to accept change, liberalism, and personal
efficacy are apparently products of the secondary environment. PeOple
are most accepting of change when they have open secondary environment
following a closed primary environment. The same is true of the sense
of efficacy and extent of liberalism. Acceptance of change, more liberal
attitudes toward politics and higher sense of efficacy result from more
extensive communications with authority in later life.

We also have suggested that open relations with authority tend to
make individuals more able to understand and accept authority and this
in turn leads to more tolerance and less authoritarianism (i.e. less
dogmatism). On the other hand closed childhood relations lead to nega-
tive reaction to and rejection of authority generally, and this appears

to be associated with being less dogmatic. Both understanding of

143

authority and rejection of authority contribute to more open and less
dogmatic personalities. We find negative evidence that the ability to
evaluate source and message independently underlie dogmatism.

If the individual experiences discontinuity in his socialization
it appears to have considerable effect upon personality. We find the
degree of discontinuity relates to six of the seven orientations
measured in the study. It does not related to political cynicism.
Discontinuity makes individuals accepting of change in the external
world and less rigid in their internal view of the world. These
experiences contribute to less extreme positions concerning trust
(neither high or low) and a reduced sense of personal efficacy. Borh
of these findings could be interpreted as a more realistic assessment
of peOple and the individual capacity to control their environment.
Finally, these people experiencing discontinuity are more liberal in
political beliefs.

The resulting picture is a person who experiences difficulty in
life and makes adjustments to these experiences and is therefore more
likely to be less rigid about his beliefs generally and more likely to
be more accepting of changes in situations arising in his external
world. He is more tolerant of others (but not more trusting) because
he has come to be more tolerant of himself as a result of his own
experiences with crisis situations which have exposed his own weaknesses.
He tends to reject authority and feels less efficacious which may be no
more than an accurate appraisal of his experiences with people. Because
of his own difficulties he is more likely to accept weaknesses in others

and want government to do something to aid people in stressful situations,

144

thus the liberal political views. Education may increase liberalism
and evidently permits the person a clearer impression of their
authorities' opinions (i.e. more education relates to both extreme
agreement and disagreement with authorities). The educated are more
able to accept change in the world even though they are not necessarily
less dogmatic. This suggests that education may be a very important
contributor to the acceptance of change at the level of cognitive
experiences above and beyond the results of early learning.

We find some evidence that inconsistency in socialization does
not change early formed orientations but does have the effect of a
more receptive cognitive structure for the effects of the secondary
environment if the transition is from closed primary to open secondary.
This particular socialization pattern (e.g. closed primary and open
secondary) appears most important as a source of acceptance of change,
and tends to produce the most intense orientations on all of the
attributes studied except dogmatism. PeOple.with consistently open
socialization are accepting of change, but those from closed primary
and Open secondary are most accepting of change.

The regression analysis has shown that personality contributes
most to the prediction of the orientation toward change, but that
socialization experiences do contribute to better understanding and
explanation of this relatively complex predisposition. The evidence
is not impressive but does seem to support the idea that both personality
variables and socialization experiences are necessary to the explanation
of the orientation toward change even though in this study the psycho—

logical attributes are doing most of the predictive work. We are able

145

to get substantial variation from our four experimental groups for at
least two of the personality attributes which suggests that the juxta-
position of early and later experiences does have an impact upon
personality.

The inconsistency does not change the early formed orientations
but tends to render the person more receptive to secondary experiences.
Perhaps secondary experiences have the effect of expanding the person-
ality. The expansion takes the form of new orientations derived from
the type of secondary experiences. It may be that inconsistency and
discontinuity do lead to change in personality, but not change in the
sense we have thought of it. These variables appear to have the effect
of facilitating the expansion of personality under the impact of

secondary experiences.
Conclusions

Early child-authority relations characterized by intense
communication and interaction have several profound effects upon the
formation of personality. In the present study we find the child
learns from these relationships several orientations which are resistant
to change under the impact of later socialization. (The child learns to
trust authority or to distrustauthority depending upon the type of
early relations with his immediate authorities, and this orientation
persists into later life regardless of the type of later experiences
with authority.) If early authorities are open and accessible to the
child, he learns to trust political authority, while if they are

r,

[closed and non-communicative he learns to distrust politics and

146

politicians. There is some indication that liberal political
orientations are more likely to emerge from closed primary sociali-
zation, but liberalism appears to be more affected by events and
relationships in the secondary environment.

Dogmatism and reliance upon authority are molded by these same
early experiences with authorities, although not in the way we had
expected. Less dogmatic subjects are products of either extremely
open or extremely closed early relations, and again, the effects
seem to be resistant to change. The subjects from open primary
environments tend to be more dependent upon the Opinions of their
authorities for their own thinking about various social and political
issues later in life. Those with closed interpersonal contact in
childhood are more independent of authority opinion in later life.

We should point out that both reliance upon authority and dogmatism
are presently conceptualized as being structural attributes of person-
ality that affect the way a person believes rather than what he believes.
We-can conclude with some confidence from the present study that the
child does learn to relate to authority in the early family situation
(i.e. learns to trust or distrust authority; to accept opinions or
reject opinions of authority; and to be more or less tolerant and
authoritarian XEOre or less dogmatié7) and the orientations formed at
this point in the child's life are relatively resistant to change

from later socialization. Also we can point out here that both
dogmatism and reliance upon authority show substantial relationship

to the various substantive or content orientations (what the person

believes). Dogmatism relates positively to efficacious attitudes

P's.

147

about personal environment and trust of peOple and politics.
Independence of authority relates positively to liberalism, distrust
of people and cynicism about politics. The curvilinear relation
between the two structural attributes (i.e. dogmatism and reliance
upon authority) leads to one of the more exciting theoretical
suggestions of the study.

Rokeachl has suggested that the underlying variable affecting
dogmatism is the ability to distinguish messages from source of
message and evaluate the two separately. The more dogmatic person
operates on a fusion of source and message and the less dogmatic
person evaluates events in the situation separately from his evalua-
tion of the source of the messages in the situations. We find
negative evidence concerning this argument in the present study.

This analysis shows low dogmatics emerge from both extremely
closed and extremely open primary relations-with authority; however,
low dogmatics from Open primary relations are more likely to agree
with the Opinions of their authorities in later life while low
dogmatics from closed family situations are more likely to disagree
with authorities later in life. We have theorized that dogmatism is
not related to ability to distinguish source and message, but instead,
it appears to be an outgrowth of two entirely different modes of
relating to authority generally. Low dogmatism (tolerance and less

authoritarianism) may result from relationships which are characterized

 

1M. Rokeach, op. cit., 1960.

148

by communication and understanding of the authority and what the
authority says, or from situations in which.there is no information
and understanding and only rejection of the source of arbitrary
dictates. Intolerance and authoritarianism appear likely only when
there is no clear impression of the authority. When there is either
positive understanding or negative rejection the individuals are
more likely to be less dogmatic (i.e. less authoritarian and more
tolerant).

As the psychoanalytic perspective has directed our attention
to the impact of primary relations upon personality, the cognitive
approach has led us to look for effects of later life socialization.
We have relatively strong evidence concerning the impact of secondary
socialization upon personality and particularly upon the substantive
attributes of personality.

Theoretically, we had expected to find open, communicative
relations with authority in later life to have the effect of
increasing personal efficacy. We have found this to be the case.
Efficacy is related to open secondary experiences while not related
to the type of primary experiences. More importantly we find even
more substantial effect from secondary relations upon other attributes
of personality.

We find relationship between Open primary relations with authority
and the variables of liberalism and acceptance of change. Open communi-
cations and interpersonal relations with authority in later life do

have an impact upon personality supporting the notion that later life

149

experiences are of critical importance to the development of orienta-
tions relevant to political behavior.

We have theorized discontinuity in socialization as having the
effect of changing orientations formed early. That is, experiences of
crisis or personal trauma require internal adjustments in personality,
and they are eXpected to have the additional effect of making individu—
als more receptive to change in external situations, as well. In fact
the variable of discontinuity has substantial impact upon the personality
attributes presently studied. Subjects who have reported at least one
crisis event (e.g. one or more such events in the list of ten most
important events in their lives) are more likely to be less dogmatic
internally and more accepting of change externally. They are also
less dependent upon authority, more liberal, less efficacious and
moderately trusting of other peOple. Discontinuity has the predicted
effect upon personality. Evidently, people experiencing such crisis
events are changed by the experiences.

We also thought level of education might contribute to the
develOpment of oreintations and particularly to the acceptance of
change. We find evidence of this as well as those with more education
being more liberal and more likely to be either most or least dependent
upon their authorities. Again this suggests that later life experiences
(i.e. secondary experiences with authority, discontinuity, and educa-
tion) all have substantial effects upon the formation of personality.

Finally, we had thought that inconsistency between early and
later socialization experiences with authority should produce changes

in personality and acceptance of change in the external world. The

150

findings require a re-evaluation of our conceptualization of personality
change. Rather than conceptualize change as from one type of orientation
formed in childhood to an opposite view in later life, we have come to
view change as an expansion process. When the transition (inconsistency)
from early to later experiences involves open primary to closed secondary
there appears to be a contraction effect. Although the personality
does absorb new orientations from the secondary environment, the
effects are not as strong or the person does not seem to be as receptive
to the new environmental circumstance and experiences. When the
inconsistency is from closed primary to open secondary the person is
most sensitive to the impact of later socialization and develops more
intense orientations as a result of the type of secondary experiences
with authority. We cannot conclude that consistency tends to make
personality more stable as we had expected. In fact, we find the
orientation of liberalism switches direction from more liberal to more
conservative in the consistently closed socialization pattern while
the Opposite occurs in the consistently open pattern. It may be that
liberalism is more affected by later experiences and therefore the
evident change is not so real, but we can only conclude that if the
socialization pattern is more consistent then the type of socialization
experiences becomes the more important determinant of personality.
Another general conclusion of the study has to do with usefulness
of conceptualizing personality orientations as a product of both
socialization experiences and other personality attributes. We have
substantial evidence that varying patterns of socialization have impacts

upon personality. The four experimental groups in our study

151

(representing each of the theoretical socialization patterns) do produce
significant variation in the orientation toward change and reliance

upon authority. This multivariate model appears to have merit in the
explanation of the particular orientation to accept change.

The results of the regression analysis and multiple correlations
suggest the psychological variables are the most important contributors
to the acceptance of or resistance to change, but our predictive
capacity is improved by utilizing the combination of socialization and
personality variables. The variables dogmatism, liberalism, secondary
environment and continuity are the most important predictors of the
predisposition to accept or resist change.

In final evaluation of this study I offer the following major
conclusions. We have found personality predispositions are relatively
complex phenomena and even the use of multi-variate schemes of analysis
provide only limited insight into their formation and operation. We
have demonstrated the utility of using such models in the more important
quest for understanding of the process of becoming a person. Perhaps
the most suggestive portion of the study has been the use of the four
sequential patterns of socialization experiences in relation to varying
personality attributes. Clearly, if further research into the relation-
ship of socialization, personality and behavior is to be meaningful it
must be conceptualized as an ongoing and changing process rather than

a static set of relationships existing at some moment.

BIBLIOGRAPHY

Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., and Sanford,
R. N. The Authoritarian Personality. New York: Harper,
1950.

Agger, Goldstein, and Pearl. "Political Cynicism: Measurement
and Meaning," The Journal of Politics, XXIII (Aug.),
pp. 477-506.

Almond, G. A. "Comparative Political Systems," Comparative
Politics: Notes and Readings. Edited by Macridis and
Brown. Homewood, Illinois: The Dorsey Press, 1964,
pp. 50—64.

Almond, G., and Verba, S. The Civic Culture. Princeton, N. J.:
Princeton University Press, 1963.

Anderson, C. "A Developmental Study of Dogmatism During
Adolescence with Reference to Sex Differences," Journal of
Abnormal and Socia1;Psycholo , 65 (1962), pp. 132-135.

Baggaley, Andrew R, Igtermediate Correlational Methods. New York:
John Wiley & Sons, 1964.

Blalock, H. M. Social Statistics. New York: McGraw-Hill, 1960.

Blumer, H. "Attitudes anthhe Social Act," Social Problems, 1955,
pp o 59-640

Brookover and Holland. "An Inquiry into the Meaning of Minority
Group Attitude Expression," American Sociolo ical Review,
April, 1952, pp. 196-202.

Campbell, A., Converse, P., Miller, W., and Stokes, D. The
Americag Voter. New York: John Wiley and Sons, 1964.

Campbell, A.,.gt..§l. The Voter Decides. Evanston, 111.:
Row, Peterson and Co., 1954.

Campbell, D. T. "Social Attitudes and Other Acquired Behavioral

Dispositions," Psychology: A Study of a Science. Edited
by Koch. New York: McGraw-Hill, 1963.

152

153

BIBLIOGRAPHY--Continued

Child, I. L. "Socialization," Handbook of Social Paychology.
Edited by Lindzey. Cambridge, Mass.: Addison and Wesley
Co., 1954.

Converse, P., and Depeux, G. "Politization of the Electorate in
France and in the United States," Public Opinion Quarterly,
26 (1962)’ ppe 1’230

Crittenden, J. "Aging and Party Affiliation," Public Opinion
Quarterly, 26 (1962), pp. 648-657.

Cronbach, L. J. "Coefficient Alpha and the Internal Structure of
Tests," Psychometrika, Vol. 16, No° 3 (Sept., 1951),
pp. 297-334.

Croxton and Cowden. Applied General Statistics. New York:
Prentice-Hall, Inc., 1939.

Davies, J. "The Family's Role in Political Socialization,"
The Annals, 361 (September, 1965), pp. 10-19.

DeFleur and Westie. "Verbal Attitudes and Overt Acts: An
Experiment on the Salience of Attitudes," American Sociological
Review, Vol. 23 (December, 1958), p. 667.

DeFleur, M. L., and Westie, F. R. "Attitude as a Scientific Concept,"
Social Forces, October, 1963, pp. 10-30.

Doob, L. W. "The Behavior of Attitudes," Ps cholo ical Review, 1947,
pp. 135-156.

Duverger, M. Political Parties. New York: John Wiley and Sons,
1963.

Easton, D. "An Approach to the Analysis of Political Systems,"
World Politics, April, 1957, pp. 383-400.

Easton, D., and Hess, R. "Youth and the Political System,"
Culture and Social Character. Edited by Lipset and Lowenthal.
Glencoe, 111.: The Free Press, 1961.

Easton, D., and Dennis, J. "The Child's Image of Government," The
Annals, 361 (September, 1965), pp. 40-57

Eisenstadt, S. N. From Generation to Generation. Glencoe, Ill.:
The Free Press, 1956.

154
BIBLIOGRAPHY--Continued

Elkins, S. "Slavery and Personality," Studying Personality Cross
Culturally. Edited by Kaplan. Evanston, 111.: Row, Peterson
and Co., 1961.

Erikson, E. H. Childhood and Society. New York: Norton and Co.,
1950.

Erikson, E. H. Young Man Luther. New York: Norton and Co., 1958.

Eysenck, H. J. The Psychology of Politics. London: Routledge and
Kegan Paul Ltd., 1954.

Froman, L. "Learning Political Attitudes," Western Polétical
Quarterly, June, 1962, pp. 21-38.

Froman, L. "Personality and Political Socialization," Journal of
Politics, 23 (1961), pp. 15-25.

Green, B. "Attitude Measurement," Handbook of Socia1;Psychology.
Edited by Lindzey. Cambridge, Mass.: Addison and Wesley
Co., 1954.

Greenstein, F. "The Benevolent Leader: Children's Image of
Political Authority," Ameriean Political Science Review,
December, 1960, pp. 934-943.

Greenstein, F. "Personality and Political Socialization: The
Theories of Authoritarian and Democratic Character," The
Annals, 361 (September, 1965), pp. 93-111.

Hagen, E. On the Theory of Social Change. Homewood, 111.:
Dorsey Press, 1962.

Hagood and Price. Statistics for Sociologists. New York: Holt &
Co., 1952.

Hess, R. and Easton, D. "The Child's Changing Image of the President,"
Public Opinion Quarterly, 24 (1960), pp. 632-644.

Horton, J. E., and Thompson, W. E. "Powerlessness andilolitical
Negativism: A Study of Defeated Local Referendums,” American
Journal of Sociolo , 47 (1962), pp. 485-492.

Huntington, S. P. "Conservatism as an Ideology," Comparative Politics:
Notea_and Readings. Edited by Macridis and Brown. Homewood, Ill.:
The Dorsey Press, 1961, pp. 276-284.

155

BIBLIOGRAPHY--Continued

Hyman, J. Political Socialization. Glencoe, 111.: The Free Press,
1959.

Katz, D. "The Functional Approach to the Study of Attitudes,"
Public Opinion Quarterly, 24 (1960), pp. 163-204.

Killan, L. M. "The Adjustment of Southern White Migrants to
Northern Urban Norms," Social Forces, Vol. 32, Oct., 1953, p. 69.

Krech, D. and Crutchfield, R. Theory and Problems of Social
Psychology. New York: McGraw—Hill, 1948.

Lambert, W. E., and Klineberg, O. "A Pilot Study of the Origin and
Development of National Stereotypes," International Social
Science Journal, 11 (1959), pp. 221-237.

Lane, R. "Fathers and Sons: Foundations of Political Beliefs,"
American Sociolo ical Review, 24 (1959), pp. 502-511.

Lane, R. Political Ideology. Glencoe, 111.: The Free Press, 1962.

Lane, R. "The Need to be Liked and the Anxious College Liberal,"
The Annals, 361 (September, 1965), pp. 71-80.

Lane, R., and Sears, D. Public Opinion. Edgewood Cliffs, N. J.:
Prentice-Hall, 1964.

LaPiere, R. T. "Attitudes Versus Action," SocialpForces, Vol. 13
(Dec., 1934), pp. 230-37.

Leblanc, M. "Acculturation of Attitudes and Personality Among
Katangese Women," Journal of Social Ps cholo , 47 (1958),
pp. 50-61.

Levin, M. L. "Social Climate and Political Socialization," Public
Opinion Quarterlyg 25 (1961), pp. 596-606.

Levine, M. "The Role of the Family in Authority Systems," Behavioral
Science, V. (1960), pp. 20-42.

Lipset, S. M., and Wolin, S. S. (eds.). The Berkeley Student Revolt.
Garden City, New York: Anchor Books, Doubleday and Co., Inc.,
1965.

Litt, E. "Civil Education, Community Norms and Political Indoctrina-
tion," American Sociolo ical Review, 28 (1963), pp. 69-75.

156

BIBLIOGRAPHY--Continued

Lohman and Reitzes. "Deliberately Organized Groups and Racial
Behavior," American Sociological Review, Vol. 19 (June, 1954),
p. 342.

Maccoby, E., Matthews, R., and Morton, A. "Youth and Political Change,"
Public Opinion Quarterly, 18 (1954), pp. 23-29.

Marvick, D. "The Political Socialization of the American Negro,"
The Annals, 361 (September, 1965), pp. 112—126.

 

McClosky, H. "Conservatism and Personality," The American Political
Science Review, 52 (1958), pp. 27-45.

Mead, M. Growing Up in New Guinea. New York: Mentor Books, 1930.

Merton, R. "Social Structure and Anomie," American Sociological
Review, 3 (1938), pp. 672-682.

Middleton, R., and Putney, S. "Political Expression of Adolescent
Rebellion," Americap7Journal¥of Sociolagy, 68 (1963), pp. 527-
535.

Milbrath, L. "Latent Origins of Liberaliseronservatism and Party
Identification: A Research Note," Journal of Polltics, 24
(1962), pp. 679-688.

Milbrath, L. Political Participation. Chicago, I11.: Rand McNally,
Co., 1965.

Mitchell, W. The American Polity. Glencoe, 111.: The Free Press,
1963.

Nagel, S. "Off—The-Bench Judicial Attitudes," Judicial Decision—
Making. Edited by Schubert. The Free Press of Glencoe,
Collier-Macmillan Ltd., 1963, pp. 29-54.

Pearlin, L. I. "Shifting Group Attachments and Attitudes Towards
Negroes," Social Forces, Vol. 33 (1954), pp. 47-50.

Pinner, F. "Student Trade Unionism: Anticipatory Socialization and
Role Seeking," Sociolo of Education, 37, No. 3 (Spring, 1964),

Pinner, F. "Parental Overprotection and Political Distrust," The
Annals, 361 (September, 1965), pp. 58-70.

157

BIBLIOGRAPHY—~Continued

"tn

Pool, Ithiel De 5013. the Mass Media and Politics in the
Modernization Process,” Communicati; s and Political
Development. Editied by Pye. Princeton, N. J.: Princeton
University Press, 1963, pp. 128-151.

 

Pressy, S. "Changes from 1923 to 1943 in the Attitudes of Public
School and University Students," Journal of Psychology, 21
(1946), pp. 173_188.

Pye, 1. A. "Personality, Identity, and Political Ideology,"
Political Decision Makers. Edited by Marvick. New York:
Free Press, 1961.

 

Rokeach, M. lhe Open and Closed Mind: Investigations into the
Nature of Belief Systems and Personality Systems. New York:
Basic Books, 1960.

Rokeach, M. The Three Christa of Ypsilanti. New York: Knopf, 1964.

Rokeach, M. "The Nature of Attitudes," International Encyc10pedia
of Social Science. New York: MacMillan, 1966.

 

Rosenberg, M. "MisanthrOpy and Political Ideology," American
Sociological Review, 21 (1956), pp. 690-695.

 

Sarnoff, 1., and Katz, D. "The Motivational Basis of Attitude Change,"
.loutnal of Abnormal and Social Psychology, 49 (1954), pp. 115-
124.

 

1.1
.1!

Siegel, S. Nonparametric Stat .tiia. New York: McGraw—Hill Co.,
1956.

Smith, B., Brunet, J., and White, R. Opinions and Personality.
New York: John Wiley and Sons, 1956.

Sorauf, F. Perspectives on Political Science. Columbus, Ohio:
Charles E. Merrill Books, 1966.

Srole, 1. "Social Integration and Certain Corollaries: An Explora-
tory Study," American Sociological Review, 21 (1956), pp. 709-
7160

Taylor, J. "A Personality Scale of Manifest Anxiety," Journal of
Abnormal and So:ia1 Psychology, 48 (1953), pp. 285-290.

Troldahl, V. G., and Powell, F. A. "A Short—Form Dogmatism Scale for
Use in Field Studies," Social Forces, 44 (December, 1965),
ppo 211‘214.)

 

158
BlBLIOGRAPHY-—Continued

Vetter, G. B. "What Makes Attitudes and Opinions 'Liberal' or
'Conservative, ” Journal of Abnormal and Social Psychology,
42 (1947), pp. 125-130.

Walker and Lev. Statistical Inference. New York: Holt, Rinehart
and Winston, 1953.

Wallace, A. "Revitalization Movements," American Anthropologist,
58 (1956).

Whiting, 1., and Child, I. L. Child Training and Personality: A
Cross Cultural Study. New Haven, Connecticut, 1953.

Wilkins, Kutner, and Yarrows. "Verbal Attitudes and Overt Behavior
Involving Racial Prejudice," Journal of Abnormal and Social
Psychology, Vol. 47 (1952), p. 650.

Willis, R. H. "Political and Child-Rearing Attributes in Sweden,"
Journal of Abnormal and Social Psychology, 53 (1956).

Ziblatt, D. "High School Extracurricular Activities and Political
Socialization," The Annals, 361 (September, 1965), pp. 20—31.

 

:\

‘lllllllllllllllllES