'FFR 07‘10957333 g
(kw???) __ _ _‘ , _.

ABSTRACT

VALUE CHANGE, PERCEIVED INSTRUMENTALITY, AND ATTITUDE CHANGE

by Charles C. Hollen

This study was conducted as part of an ongoing series of studies at
Michigan State University investigating experimentally induced changes in
values, attitudes, and behavior. The purpose of this study was to test
the effect of an induced value change on changes in instrumentally re-
lated attitudes.

In the theoretical viewpoint which guided this investigation, values
are seen as part of the belief system of an individual (Rokeach, 1968a).
A belief is considered to be "any simple proposition, conscious or un-
conscious, inferred from what a person says or does, capable of being

preceded by the phrase 'I believe that...'" (1968a:113). A belief system

 

consists of "the total universe of a person's beliefs about the physical
world, the social world, and the self" (1968a:123). In this system, a
33123 is "an enduring belief that a specific mode of conduct or end-state
of existence is personally and socially preferable to alternative modes

of conduct or end-states of existence" (1968a:160). A value system is

 

"an hierarchical organization--a rank ordering--of ideals or values in
terms of their importance" (1968a:124). Finally, an attitude is defined
as "a relatively enduring organization of beliefs around an object or
situation predisposing one to respond in some preferential manner" (1968a:
112).

The following hypotheses were advanced as the focus of this inquiry:

Charles C. Hollen

1. A significant change in the relative importance of a particular
value to an individual may be induced by presenting a persuasive message
which makes the individual aware of an inconsistency between his own value
ranking and those of significant others.

2. An induced increase in the importance of a value will lead to
increases in favorable attitudes toward objects which are perceived as
facilitating the attainment of that value and increased disfavor toward
objects perceived as hindering the attainment of the value.

3. Given an induced value change, the amount of change in each
instrumentally-related attitude will be in direct proportion to the degree
of perceived instrumentality possessed by that attitude object, i.e. the
degree to which that attitude object is perceived as instrumental to the
attainment of the value.‘

In other words, we may change a person's attitudes indirectly, by
changing his values. An increase in the importance ascribed to a particular
value will lead to an increase in favorability toward objects seen as
facilitating the attainment of that value, and a decrease in favorability
toward objects seen as hindering the attainment of that value. The amount
of positive or negative attitude change should be relatively greater in
the case of objects which are perceived as highly instrumental, either posi-
tively or negatively, to the attainment of a value, and less in the case
of objects whose "perceived instrumentality" is relatively slight.

An experiment was conducted to test the hypotheses, using question-
naires administered to volunteer college student subjects. There were
three experimental sessions, the first consisting of the pre-tests and

the second, one week later, including the experimental treatment and the

Charles C. Hollen
immediate post-tests. The third experimental session, administered to
a subgroup of the subjects one month later, consisted of the same post—
tests.

Initial attitude position was measured by asking subjects to indicate

 

their degree of agreement or disagreement with each of a series of
attitudinal statements, each one proposing some social action. Second,
initial value importance was measured by asking subjects to rank a set

of 18 values from most to least important. Third, perceived instrumen-

 

tality was measured by asking subjects to respond again to the same set
of attitude statements, this time by indicating to what extent they felt
each social action would help or hinder the attainment of a particular
value. Statements pertaining to two values were included, the value to
be manipulated (A E2£lfl.2§ Beauty) and a "control" value, (A florid £5
Peace). This concluded the first experimental session. One week later,
subjects were given a written persuasive message designed to increase the
importance of the value, A K2£li.2£ Beauty. A control group received no
manipulation. After receiving the persuasive message, subjects were
administered post-tests of value importance and attitude position, using
the same questions which had been used in the pre-tests. One month later,
a subgroup of the subjects were given the post-tests for a second time,
in order to measure the persistence of the induced changes.

Value change was then computed by taking the difference between the

 

pre-test and post-test rankings of each of the values. Attitude change

 

was similarly computed, by taking the differences between the pre- and
post-measures for each attitude item. Perceived instrumentality was taken

directly from subjects' ratings.

Charles C. Hollen

Apparently value importance can be successfully changed using per-
suasive techniques. Our results show that the importance assigned to A
flgyld of Beauty was significantly changed immediately following the
persuasive message and this change had persisted and slightly increased
one month later. Previous studies have shown that changes in value im-
portance are highly persistent over periods of a year or more (Rokeach,
1971b). Thus we may conclude that we may induce significant and persis-
tent changes in value-importance through persuasion.

Our results show that a change in value importance did produce
changes in attitudes toward related issues. Moreover, the results
showed that attitudes changed toward those actions which were seen as
highly instrumental to value attainment. Thus the study provides further
support for the conception of instrumental linkages between values and
attitudes. Given an increase in value-importance, the more a particular
action was seen as facilitating the attainment of that value, the more
the individual's attitude toward that action changed toward increased
favorability.

There was also a tendency, though not as clear-cut, for individuals
who had changed their values to become less favorable toward actions seen
as hindering the attainment of the value in question. Thus it would
seem that it is possible to change a number of attitudes at once, through
a single manipulation, by directing that manipulation toward a value to
which the attitudes are seen as instrumental.

Results from a delayed post-test showed that the obtained attitude
change did not decrease over time. Indeed, they tended to increase

slightly (though not significantly). The results of studies of attitude

Charles C. Hollen
change typically show decreases over time as the effects of the manipu-
lation "wear off" and individuals' attitudes regress toward their initial
positions. It is extremely interesting to note that in this study the
opposite effect occurred.

The results of this study indicate that (a) individuals' values are
amenable to influence through traditional persuasive methods, (b) changes

.

in values bring about changes in attitudes toward instrumentally related
objects, and (c) these changes in attitudes do not decrease over time.
Thus it would seem to illustrate a potentially powerful means of persuasion
and influence. First, changing attitudes indirectly through changing values
would appear more economical for persuaders, in that a number of
attitudes may be influenced "in one fell swoop" through the manipulation
of a single value. Second, it would appear that attitude changes obtained
through this indirect method may be more lasting than attitude changes

obtained without regard to (or perhaps even in spite of) the individual's

values.

VALUE CHANGE, PERCEIVED INSTRUMENTALITY, AND ATTITUDE CHANGE

by

Charles C. Hollen

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Psychology

1972

ACKNOWLEDGMENTS

I wish to express my deep thankfulness to all those who helped
in the planning and execution of this research.

I am particularly grateful to my committee chairman, Dr. Milton
Rokeach, for his always generous contributions of time, patience,
incisive questions, and wise counsel. I would also like to thank
the other members of my committee-~Dr. Eugene Jacobson, Dr. Verling
Troldahl, and Dr. William Crano--who provided encouragement and many
helpful comments. I particularly appreciate the aid of Mrs. Catherine
Scheer in the preparation of computer programs for analysis of the
results.

Finally, I am deeply indebted to my wife Karen, who provided an
abundance of encouragement, prodding, and general moral support. The

value of her aid and comfort is incalculable.

ii

CONTENTS

INTRODUCTION . .

Studies of Indirect Changes

Summary and Evaluation

The Role of Personal Values in Human Behavior
The Relation Between Attitudes and Values
Definitions of Terms Used in This Research
Hypotheses

METHOD

Overview of Method

Subjects

Materials and Procedure
Attitude Pre-Test

Value Importance Pre-Test
Perceived Instrumentality Test
Second Experimental Session
Experimental Treatment
Value Importance Post-Test
Attitude Post-Test
Matching Questions

Third Experimental Session
Constructed Variables

RESULTS

Value Change

Value Change — Delayed Post-Test

Attitude Change

Attitude Change - Delayed Post-Test

The Relation Between Value Change and Attitude Change
Perceived Instrumentality and Attitude Change

DISCUSSION

Value Change Through Persuasion

Value Change, Perceived Instrumentality, and Attitude Change
Delayed Action Effects in Value and Attitude Change
Changing Attitudes Through Value Change

Conclusions

REFERENCES

APPENDICES . . . . . . . . . . . . . . . .

Verbal Instructions for Session One

Values and Attitudes Questionnaire

Verbal Instructions for Second Session
Values and Attitudes Questionnaire - Part II

UCUJ>

iii

10
15
17
27
27

29

29
30
3O
31
32
32
33
34
35
35
35
36
36

38

38
40
42
44
46
48

52

52
53
5A
56
57

59
62
62
63

69
70

Table

LIST OF TABLES

Mean changes in rank order of 18 terminal values
for experimental and control groups . . . .

Mean changes in rank order of 18 terminal values
for subgroups of experimental and control subjects
tested immediately and one month later . . . .

Mean changes in agreement-disagreement ratings
of 11 attitude statements for experimental and
contrOI groups C O O O O O O O O O O O O O I O 0 0

Mean changes in attitudes for subgroups of experi-
mental and control subjects tested immediately

and one month later 0 O O O O O O O O O O O O 0

Mean changes in attitude for experimental subjects
grouped according to amount of value change . .

Mean instrumentality ratings and mean incremental
changes in attitudes of experimental subjects

iv

Page

39

41

43

45

47

49

INTRODUCTION

In past studies, attitudes have frequently been treated as if they
existed independently, in isolation from one another and from other
cognitive elements. Recently, however, it has begun to be recognized
that in order to fully understand the functioning of attitudes in human
behavior, we must attempt to place them within some organized system of
relationships to other entities. Conceptions of linkages between an
attitude and other attitudes and beliefs have been advanced by various
theorists (McGuire, 1960a; Rosenberg, 19603; Rokeach, 1968a). It has
been suggested that we must obtain evidence concerning the cognitive and
motivational systems in which an attitude is embedded in order to under-
stand and predict the ways in which attitudes determine social behavior
(Rokeach, 1968).

An effective method of studying the extent and strength of attitudinal
linkages is through the study of "repercussion effects," i.e. the changes
in related attitudes and beliefs when a particular attitude is changed
through experimental manipulation. The "cognitive consistency" models of

V
human thinking and action (Heider, 1946; 1958; Newcomb, 1953; Osgood and
Tannenbaum, 1955; Cartwright and Harary, 1956; Festinger, 1957; Abelson
and Rosenberg, 1958; Rokeach, 1968a) would all agree that pe0p1e tend
to behave in ways which will maintain an internally consistent belief system.
Thus, an induced change in one belief should arouse inconsistencies with
related beliefs, leading to "readjustments" to restore consistency. These
"readjustments" may take a variety of forms, including the restoration
of the manipulated belief to its original position (we see this occurring
in the ubiquitous Hregression" or "decay" effects). However, if the

1

2
manipulated change is accompanied by sufficient motivating force, then
it will persist, and will set off "repercussions" as other beliefs are
changed to accommodate it. The processes through which such "cognitive
diffusion" effects occur are not well understood, and until very recently,
have received little attention from investigators.

Studies of Indirect Changes

 

Empirical results pertaining to indirect change effects are not
numerous in the literature on attitude change. Only a few studies have
included measures of attitudes toward additional issues not explicitly
mentioned in the persuasive message.

I’Hovland, Lumsdaine, and Sheffield (1949) conducted a series of
studies of the impact of orientation films on military recruits. They
found that large and significant Opinion changes on specific issues dealt
with by the films were not accompanied by changes on implicit, generalized
conclusions which were intended as the actual target of the films:} Thus
opinion changes toward the effectiveness of the Royal Air Force in the
Battle of Britain did not lead to changes in general confidence in British
fighting strength. They suggest that a general attitude toward some object
is a function of many specific opinions about particular aspects of the
object. They conclude that "it would appear that changes in specific
opinions can occur without being accompanied by changes in general orien-
tation toward an issue" (1949: 70). In a further effort to ascertain
whether effects on general attitudes require a "sinking-in" period, they
conducted a study in which one group was post-tested five days after
experimental treatment while another group was tested nine weeks later.
Their results showed some tendency for the effects on general conclusions

to increase with the passage of time. Unfortunately these results are

3

equivocal, since they occurred on some issues but not on others. They
cautiously interpret these results as indicating some degree of support
for the hypothesis that changes in Specific Opinions may lead to changes
in general orientations gradually over time.

iStotland, Katz, and Patchen (1959) report a study in which subjects
were given case history material regarding the roots of prejudice. This
information set in motion a process of change which led to alterations
in their Opinions on specific issues pertaining to racial prejudice. These
Opinion changes were not present on an immediate post-test, but were found
on a delayed post-test administered several weeks later.

Carlson (1956) administered a persuasive message designed to increase
subjects' favorability toward "allowing Negroes to move into White neigh-
borhoods." He also obtained measures of attitudes toward five related
issues, (a) Mexican housing segregation, (b) Jewish housing segregation,
(c) command of White enlisted men by Negro Officers, (d) command of White
enlisted men by Mexican officers, and (e) command of White enlisted men
by Jewish Officers. He found that attitude change on the "target" issue
was accompanied by significant change on all five related issues. His
results provide no evidence concerning the time-delay factor, since his
study included only a single post-test, administered three weeks after
the experimental treatment. He hypothesized that there would be differ-
ential amounts of change on the five related issues, according to a
"gradient of generalization" based on the similarity of the attitude
Objects and situations. This hypothesis was not confirmed by his results.
He concludes that "further study Of the process of generalization of change
of social attitudes will depend upon systematic analyses of the gradients

of conceptual similarity involved" (1956: 260).

4

Tannenbaum and Gengel (1966) demonstrated that a change in attitude
toward a given concept will be accompanied by changes in attitudes toward
sources (persons) who previously made assertions about that concept. Sub-
jects were given a brief message describing a fictitious "Dr. McLay" as
a prOponent of teaching machines, "Dr. Samuels" as Opposing teaching
machines, and "Dr. Spence" as neutral. Following this linking message
the subjects were administered persuasive messages designed to change
their attitudes toward teaching machines. For half the subjects the
communication favored teaching machines and for the other half the com-
munication Opposed teaching machines.b/U:ilizing Osgood's congruity theory
(Osgood and Tannenbaum, 1955) Tannenbaum and Gengel predicted that a
favorable change in attitude toward teaching machines should produce a
favorable change toward the source with the positive linkage (Dr. McLay)
and an unfavorable change toward the source with the negative linkage
(Dr. Samuels). Comparably, the unfavorable change toward teaching
machines should also be accompanied by changes in attitudes toward the
two "sources,” but in the Opposite direction. In general, their results
quite thoroughly confirm the hypotheses, although the negative changes
were somewhat less than predicted, and significant positive changes toward
the neutral source were Obtained. These effects were probably due to a
generally favorable impression Of all three sources, imparted by the
linkage message, even though it was not intended to change attitudes. It
could be argued that the linkage message itself rather than the attitude-
change message may have been reSponsible for what appear to be the "indirect
change effects." This argument could not be effectively refuted from
the data were it not for the existence of the "negative persuasion" group.

For this group, a negative change toward teaching machines was accompanied

5

by a significant positive change toward the source who had been nega-
tively linked with teaching machines. This result is the reverse Of
what would be expected on the basis of the linkage message alone, given
that subjects' initial attitudes toward teaching machines are neutral
or mildly favorable. To counter anticipated questions regarding the
possible presence of demand characteristics, the linkage message was
administered by a different experimenter for an ostensibly different
purpose. Follow-up interviews with subjects indicated that the two
messages were in fact perceived as independent of each other.

In a second study, Tannenbaum (1966) Obtained evidence indicating
that a change in attitude toward a given concept will produce a change
in attitude toward a source who has been linked with that concept, and
in addition, a "second-order" change, toward an additional concept about
which the source has made an assertion. In a design similar to the previous
one, the concepts "teaching machines" and "Spence learning theory" were
used, along with the source "Prof. Samuels." Four different linking
messages were used, containing different combinations Of positive and
negative linkages between the source and the two concepts. Results fully
supporting the hypotheses were obtained, including predicted negative
changes which the previous (Tannenbaum and Gengel, 1966) study had failed
to Obtain. Thus not only did attitude change generalize from a manipulated
concept to a linked source, but also from the source to an additional linked
concept. These findings provide strong support for the existence of at
least certain types of indirect change effects, and by inference, for the
general conception of indirect changes predictable from balance or con-
sistency theory. It must be pointed out, however, that no evidence was
found (nor was any sought) to show that these indirect changes were delayed

or required a "sinking-in" process. The post-test was administered

6

immediately following the persuasive message, and there was no other
post-test.

McGuire (1960a, 1960b) has conducted a series of studies, one focus
of which is the logical repercussions Of change. Subjects are asked to
rate the truth probability of three syllogistically related statements--a
major premise, a minor premise, and a conclusion--which are randomly
dispersed among other statements in a questionnaire. On a subsequent
occasion, persuasive messages are administered which argue for the truth
value of the minor premise. Immediately and again later, the impact of
the persuasion on the premise and on the unmentioned but logically related
conclusion is retested. Each subject receives persuasive messages relating
to several syllogisms. McGuire provides the following example of a set of
statements used in his studies: (1) ”Any form of recreation that consti-
tutes a serious health menace will be outlawed by the City Health Authority."
(2) "The increasing water pollution in this area will make swimming at
the local beaches a serious health menace." and (3) "Swimming at the
local beaches will be outlawed by the City Health Authority." In addition
to ratings of truth probability, McGuire also obtains ratings of the
desirability of the state of affairs described in each statement. He
uses these data to test a "wishful thinking postulate," i.e. that change
will be greater on statements which are rated as highly desirable. Thus,
an individual will behave to increase logical consistency among his beliefs,
but this tendency will be tempered or distorted somewhat by a tendency
toward "affective consistency" i.e. consistency with desired or hOped-
for states Of affairs. McGuire also advances a "cognitive inertia"
postulate, stating that the amount of change will be less than the amount

required to achieve complete logical consistency, and that this change

7

will occur through 3 delayed-action process. The results Of McGuire's
studies have demonstrated repeatedly that persuasion directed at a
"premise” will produce change, not only on it, but also on the unmentioned
"conclusion.” Unfortunately, McGuire's results, examined closely, are
not strong and unequivocal evidence of delayed-action effects on derived
issues, even though he interprets them as providing significant support
for this effect. Ideally, in order to conclude that indirect changes
were delayed, one would expect tO find large, significant changes on the
premise but little or no change on the conclusion at the time of the
immediate post-test, then a large and significant change in the conclusion
on a subsequent, delayed post-test. In McGuire's studies, both the
premise and the conclusion changed significantly on the immediate post-
test, and both regressed back toward initial positions on the delayed
(one week later) post-test. However, he found a slight (p.<.10) tendency
for the reversion Of the conclusion to be less than that predicted by
the reversions on the premises. He interprets this as evidence for
”continued seepage of the impact" from the explicit issue tending to
overcome "the Opposing decay effect on the derived issue." (19603: 350).

Regardless of whether his results adequately illustrate delayed-
3ction effects, we are indebted to McGuire for his demonstrations Of
another type of indirect change effect, that which occurs according to
logic. In addition, his prOposals concerning "wishful thinking" and
"cognitive inertia" are instructive.

Dillehay, Insko, and Smith (1966) report two studies using the
same syllogisms, communications, and instructions as McGuire. The results
of these studies for the direct and indirect effects of persuasion are

consistent with McGuire's findings. The second of these studies is

8

particularly helpful since it utilized an after-only design. A potential
criticism of McGuire's studies can be based on his use of a before-after
design. Even though the syllogistic statements were scrambled and embedded
among other statements, the pre-test may have served to sensitize the
subjects to the logical connections between the statements. Dillehay,
Insko, and Smith's second study showed that apparently this is not the
case, since they found results consistent with McGuire's, on their immediate
post-test which was not preceded by a pre-test. Unfortunately, they do
not mention whether in their post-test, the attitudes toward the conclusions
were tested before or after the attitudes toward the premises. Secondly,
they did not administer the two post-tests to two separate groups. Thus,
their design does not preclude the presence Of sensitization effects of
the immediate post-test on the delayed one. Whether or not such effects
may have been present, an interesting finding of their study was that at
the time of the delayed measurement there was no significant difference
between the message and nO-message groups on the derived, unmentioned
issues. 923.yggk_after experimental treatment there was "no remaining
impact of the communications on the conclusions" (1966: 653). The failure
of this replication casts further doubt on McGuire's somewhat weak results
concerning delayed action effects.

Rokeach, Reyher, and Wiseman (1968) conducted a study designed to
show that a change in a "more central" belief will produce greater changes
among related beliefs than will a change in a "less central" belief. 55
belief statements were selected to represent five types of beliefs, varying
systematically in degree of centrality. These five types (in descending
order of centrality) were: (A) Primitive beliefs, unanimous consensus,

i.e. "my name is...," (B) Primitive beliefs, zero consensus, i.e. "I

9
believe that my mother loves me," (C) Authority beliefs, i.e. "the
philosophy Of Dwight Eisenhower is basically 3 sound one," (D) Derived

beliefs, i.e. "birth control is morally wrong,"

and (E) inconsequential
beliefs, i.e. "Elizabeth Taylor is more beautiful than Dinah Shore."
It was postulated that a more central belief should possess more and
stronger functional connections (linkages) with other beliefs than a less
central one. Hence an induced change in a Type A belief should produce
greater changes among related beliefs than 3 Type B change, and so on in
descending order for each of the remaining belief types. Belief statements
to represent each of the five types were selected on the basis of intui-
tive judgments "that they should be sufficiently interconnected and have
implications for one another, so that a change occurring in one or more
beliefs of one type could reasonably be eXpected to lead to changes in
the remaining beliefs." (Rokeach, 1968: 26). However, the precise nature
of the relationships among the "related" beliefs, logical or otherwise,
was not systematically formulated, nor were any predictions made concerning
which particular "related" beliefs would be expected to change as 3 result
of change in a manipulated belief.

Belief changes were induced by hypnotic suggestion. Subjects were
selected who were capable of being placed in a deep hypnotic trance and
who evidenced complete posthypnotic amnesia. In a series of experimental
sessions, subjects' responses to the 55 belief items were Obtained immedi-
ately prior to hypnosis, while under hypnosis, and following removal Of
hypnotic trance but while under post-hypnotic suggestion. The only results
which the authors have presented pertaining to the "indirect changes" hy-
pothesis concern simply the sheer amount of change, averaged across subjects

and across beliefs. In their "raw" form, and after having been "adjusted"

10
to take into account differential amounts of change in the manipulated
beliefs, the results show that substantial indirect change effects were
obtained. However, they show very little difference in the amount of
indirect change produced by manipulating each Of the belief types.
Although the results differ very slightly in the predicted direction,
which is encouraging, the differences appear insignificant. (The authors
do not report a test Of the significance Of the differences.) It is
apparent that further study of the relation between belief centrality and
repercussion effects would be useful. On both theoretical and common-
sense grounds, the authors' hypothesis appear to be a reasonable and valid
one. Unfortunately, Opportunities to study repercussion effects of this
type are sharply limited, since extremely central beliefs such as one's
own name are not readily manipulable except by extreme measures such as
hypnosis, and it cannot be known whether the types of indirect changes
which occur under hypnosis are comparable to those which occur naturally.

Summary and Evaluation

 

In summarizing and evaluating the evidence concerning indirect change
effects, several general conclusions may be drawn, some of which are more
in the nature Of tentative prOposals rather than conclusions, since they
lack strong or unequivocal empirical support at this time.

1. Indirect change effects are limited by certain inherent prOperties
of belief-system linkages. In almost every case it has been found that
when persuasive messages do produce changes in related, unmentioned issues,
these changes are less than the amount required to achieve complete
consistency. McGuire (1960b) has prOposed that this is due to "cognitive
inertia." However, this approach to the problem is characteristic Of those

who apply a label to a phenomenon instead Of attempting to understand or

ll

explain it. A variety Of mechanisms might be prOposed to account for
"cognitive inertia." Most plausibly, following Hovland e£__l. (1949)
it seems reasonable to assume that an individual's position on a "related"
issue is not determined solely by his position on the manipulated issue,
but rather is itself linked to a number of related beliefs, each of which
determines a certain amount Of the variance, as it were. The existence of
this type of internal belief-system relationships has been strongly supported
by the work Of Fishbein (1963, 19653, 1965b, 1966) which shows that an
individual's attitude toward an Object can be conceptualized and measured
as an additive function of his beliefs about its' characteristics and his
evaluations Of those characteristics. This theoretical expectation is
also implicit in the work of Rosenberg (1956, 1960) nd others who have
studied the relation between values and attitudes. However, no study has
been undertaken to attempt to predict the amount of change in a related
attitude or belief from prior knowledge of the amount of variance accounted
for in it by the manipulated belief.

Additional constraints on the occurrence of repercussion effects may
be postulated in the form of belief system isolation and compartmentalization.
As Rokeach (1960), Rosenberg (1960), and other cognitive consistency
theorists have pointed out, all individuals' belief systems contain numerous
disconnections and inconsistencies.V/ECZryone must maintain certain "mental
blocks” and defensive avoidances in order to maintain sanity in a world
of violently conflicting stimuli. In addition, each individual to some
extent maintains his own "psycho-logic." Thus we may expect considerable
idiosyncratic differences affecting the extent to which a change in Belief A
is perceived as necessarily implying a change in Belief B. Few of these

factors have been systematically investigated, but all of them must be

12
taken into account before we can reach complete understanding and prediction
of the nature and extent of repercussion effects.

2. NO single theoretical model appears adequate to account for all
types of indirect change effects. Repercussion effects appear to proceed
through a variety of different types of interconnections. Past studies
show changes going from specific beliefs to general orientations (Hovland,
Lumsdaine, and Sheffield, 1949), from a specific issue to similar issues
(Carlson, 1956), from a concept to a linked source (Tannenbaum and Gengel,
1966), and from a logical premise to a deduced conclusion (McGuire, 19603,
1960b). The empirical results obtained by McGuire appear to support his
"logical model" as a candidate for consideration as a possible general
model within which others may be subsumed. However, Dillehay, Insko,
and Smith (1966) have noted that McGuire's results provide no evidence
whatsoever that repercussion effects occur according to logic. In discussing
their own results, they point out that:

Perhaps changes occur in derived unmentioned issues less as a result

Of the logical structure of the relationships between explicit and

implicit issues than of the extralogical cognitive relatedness of

the prOpositions. If persuasion produces a change in acceptance

of Fact A about Object X, this change may influence the acceptance

of other facts about the Object, especially facts that have an ex-

periential relationship to Fact A or are "reasonably" related to A.

What we are suggesting is that the process underlying the observed

changes in McGuire's studies and ours may be due to a psychological

process based on experience and/or judged reasonableness and not

to valid logical structures among beliefs. The apparently logical

repercussions found in these studies may be due to the fact that we

only looked at derived issues that were logically related. (1966:

654)

The only way to clearly prove or disprove the validity of this
contention would be to show evidence of indirect change effects between
cognitions which are related eXperientially or reasonably but not logically.

This may be easier said than done. The attitudes toward sources and

concepts utilized by Tannenbaum and Gengel (1966), for example, can be

13
rather simply and directly linked by a pseudo-logical premise ("Teaching
machines are valuable and useful Objects, therefore anyone who advocates
them must be an intelligent and worthy person, and anyone who Opposes
them must be either stupid or dishonest.") However, in the case of
results such as Carlson‘s (1956) it may not be so easy to make a "logical"
connection. In order to establish a logical connection between Negro
housing integration and issues such as the command of WASP enlisted men
by Jewish or Mexican officers, one would require a rather long and tenuous
chain of premises. It would seem rather more parsimonious to conclude that
repercussion effects such as these took place more through "reasonable"
or "experiential" connections than through logical deductions.

The type of indirect effects found by Hovland _£_al, (1949) would ap- L///A
pear best explained by the type of model advanced by Fishbein (1963, l9653,,/’/
1965b, 1966) and by Rosenberg (1956, 1960) in which an individual's attitude
toward an Object is seen as a joint function of his beliefs linking it with
other Objects, concepts, values, or goals, and his evaluations Of or attitudes
toward these "related objects." I shall review these models in more detail in
a subsequent section of this paper. In regard to the present concern, it may
be stated that although these models appear potentially capable of accounting
for some types Of indirect changes, it does not appear that they are suffi-
ciently general to account for all the interconnections within a belief system,
hence all the ways in which repercussion effects can occur. In addition, ex-
cept for Hovland SE gl.'s (1949) suggestive but inconclusive results, studies
of indirect changes based on these models have not yet been carried out.

3. Indirect changes are delayed, and may occur days, weeks, or
months following the manipulation. For example, in a recent study con-
ducted by Rokeach (1971b) a "delayed action effect" was found, through which

attitude change took place following changes in values. A manipulated

14
change in value importance was followed by changes in related attitudes
on post-tests taken three months to 17 months later. These indirect
attitude changes were not present on a post-test taken three weeks after
the manipulation. In this particular study, which will subsequently be
discussed in more detail, the repercussions began to occur only after
three weeks had elapsed, and some time prior to three months afterwards.
It is difficult, of course, to precisely pinpoint the process Of indirect
change over time. TO do so would require a series Of closely-Spaced post-
testS. And the time factor would vary according to the nature Of the
manipulated and the related cognitions.

The time required for indirect changes may be a function of the
closeness or remoteness of the affected cognitions to the one manipulated--
their distance from it in "belief Space." We may envision a manipulated
change in belief as setting off a chain of repercussions, beginning with
immediate effects on closely linked beliefs and followed by second- and
third-order changes as effects Spread throughout the surrounding belief
Space in ever-widening circles. NO Study has been undertaken to test
this postulate directly, although such 3 study would seem eminently worth
doing.

4. Some types of beliefs may be more powerful sources of repercussions
than others. Rokeach, Reyher, and Wiseman's (1968) compelling arguments
and their suggestive results indicate that a change in a central belief
may lead to a greater number and/or magnitude Of indirect changes than
will a change in a peripheral belief. This study is one of a mere handful
in the literature which have addressed themselves to the neglected problem
Of systematic differences among beliefs or attitudes having to do with their

importance or unimportance. In a sense it may be seen as an attempt to

15
distinguish those sources of repercussion effects which are important,
which represent significant and lasting changes in an individual's life,
from those which are trivial, superficial, and transitory. Based on
Rokeach, Reyher, and Wiseman's finding, we may postulate that a change
in the importance of a value will lead to a greater number and/or
magnitude Of indirect changes than will a change in some less funda-
mental cognition.

A person's attitude toward an Object (person, group, issue, etc.)
consists Of a cluster of beliefs about that object. This cluster includes
a variety of types Of beliefs. A crucially important subset of beliefs
within this cluster are his beliefs about the object's potentiality for
facilitating or blocking the attainment of important values. A favorable
attitude toward an object will be associated with beliefs that it (the
attitude object) tends to facilitate the attainment of important values,
while an unfavorable attitude will be associated with beliefs that the
attitude Object tends to block the attainment of important values. These
beliefs may be referred to as "instrumentality beliefs." They constitute
one type of belief system interconnections, concerning which we may
prOpose the following mechanism for indirect belief change which is the
thesis of the present research: If we can succeed in substantially altering
the importance an individual ascribes to a particular value, he will also
change his attitudes toward objects which he sees as instrumental to the
attainment of that value. The amount of change in each attitude will be
in direct proportion to that attitude object's degree Of instrumentality
to the value in question.

The Role of Personal Values in Human Behavior l [

Rokeach (19683) has pointed out that half a century Of attitude

 

research has resulted in a proliferation of fragmentary and micrOSCOpic

l6
knowledge of attitudes but has not led to any satisfactory theoretical
conception of the role of attitudes in social behavior. He prOposes that
the concept of value be accorded greater attention in our attempts to
predict attitudes and behavior. He argues that value is a more powerful
and efficient eXplanatory concept than the concept of attitude, since:
(1) values are more fundamental components of the individual's belief system
than attitudes, (2) values are determinants of attitudes as well as of
behavior, (3) values are relatively few in number and are organized
hierarchically, while attitudes are extremely numerous and possess no
inherent organizing structure, and (4) the concept of value is a more
clear and unequivocal locus of motivational dynamic, to account for the
purposiveness or goal-directedness Of human behavior. In Rokeach's
conception, an individual's day-to-day behavior is organized around his
efforts to attain certain values. Thus, his attitudes may be understood
as favorable or unfavorable reactions to objects in his environment
which he sees as instrumentally related to his values, either directly
or indirectly. The relevant question concerning any attitude object
thus becomes: which values does the individual see this Object as facili-
tating or hindering? It is not suggested that direct instrumentality
relationships are the sole determinants of attitude intensity. In some
cases, an attitude may be formed directly through experience with the
attitude object itself. For example, a child may form a negative attitude
toward a particular dog as a result of being bitten by it. In other
cases, an attitude may be induced through associations with other attitude
objects, as in the well-known case of induced favorability toward some
issue or person, through the suggestion of a respected authority figure.

However, the original source of all positive evaluations of external

17
objects is the individual's value system. His values, of course, are a
product of socialization, and thus reflect the culture in which he lives,
as well as the influences of his own personal childhood and personality.
Thus the influences of culture and personality on attitudes and behavior
may be understood through their role in shaping values, which are viewed
as fundamental determinants of attitudes and behavior.

The type of indirect change proposed here--value change leading to
attitude change-~is not intended as a comprehensive model for indirect
changes. It represents only one specific type of indirect change -
albeit an important one. Earlier, in stating the thesis of this research,
we postulated that "instrumentality beliefs” constitute the functional
link between attitudes and values. To defend this assumption, we shall
review briefly the evidence from previous studies of the relation between
attitudes and values.

The Relation Between Attitudes and Values

 

An Operational definition of value was introduced into the literature
of social psychology in 1931 by Vernon and Allport (Vernon and Allport,~
1931). They constructed a standardized, generalized measuring instrument
for personal values. The availability of this test stimulated a number
of investigations of personal values, including studies of their relation-
ship to attitudes. Early studies of values and attitudes typically were
correlational, seeking to identify which values and which attitudes were
related. For example, Pintner (1933) correlated college students' scores
on the Allport-Vernon scale with their attitudes toward church, prohibi-
tion, war, and the Negro. His results, like those of most Of the other
correlational and exploratory studies, yielded moderate (.20 to .40)

correlations, most Of'WhiCh seemed consistent with common-sense expectations.

18

G. H. Smith (1946) found evidence indicating that individuals who
possess the same positive value may at the same time possess Opposing
attitudes, depending on whether they perceive the Object of their attitude
as favoring or Opposing their value. In a study of values and attitudes
toward Russia, he found that individuals who approve of Russia affirm
that one can worship as he pleases there, while those who Oppose Russia
deny that a person may worship as he pleases in the U.S.S.R. Both groups
agree on religious freedom as a value. This example revealed that the
relation between values and attitudes may not be a simple one.

Brewster Smith (1949) conducted an interview study to test the
hypothesis that a value for liberty would be related to negative attitudes
toward Russia, while a value for economic security would be related to
positive attitudes. His first hypothesis was confirmed, but he found no
relationship between economic security as a value and attitudes toward
Russia. This led him to hypothesize a difference in levels between liberty
and economic security, the former being usually thought of as an abstract
universal good, but the latter usually having a much more personal focus.
Upon looking again at his data, he found that peOple who mentioned economic
security as important to them tended to possess less information about
Russia and to be less interested in world events and in taking part in
community affairs, while peOple holding a high value for liberty showed a
broader sc0pe of interests. He concludes by prOposing that a person will
tend to perceive and judge the focus of an attitude in terms of one of
his personal values to the extent that (a) the value is important to him,
occupying a central position in his value hierarchy; (b) the information
available to him about the focus (the Object) contains a basis for engaging
the value; and (c) the SCOpe of the value and of the person's interests

is broad enough to extend to the focus of the attitude.

19 X

A study conducted by Woodruff and DiVesta in 1948 represents a
landmark in the literature on attitudes and values. Not only was it the
first attempt to go beyond the correlational approach and study the
effects of change, in addition it presents for the first time an attempt
to Specify the functional relation between attitudes and values. Woodruff
and DiVesta use the term "concepts" to refer to what we herein call
"instrumentality beliefs." They hypothesize that:

An individual's attitude toward any object, prOposition, or cir-

cumstance will be favorable if, according to his concepts, that

object seems to favor the achievement of his strong positive

values. Conversely, one's attitude toward any object, prOposition,

or circumstance will be unfavorable if, according to his concepts,

the object seems to threaten his strong positive values. (1948: 648).
They also hypothesized that a change in an individual's concepts would lead
to a change in his attitude. They used a test previously develOped by
Woodruff (1942) to measure the importance of a number of values, and a
combination of structured and Open-ended questions to measure attitudes
and concepts. They found a correlation Of .80 between 83' attitude scores
(their positions on the issue of whether fraternities and sororities Should
be abolished) and their "concepts" scores, which consisted of the net
total of the number and strength of their beliefs about the value-attaining
and value-blocking prOperties of fraternities and sororities. Later, after
having been subjected to favorable prOpaganda pertaining to the Greek
system, the subjects' attitudes and concepts were re-tested. It was found
that a change in attitudinal position on the issue was accompanied by
changes in instrumental "concepts." Thus both Of the hypotheses were
confirmed. The authors conclude by suggesting that their results support

the general thesis that "values are directive factors in behavior, and

attitudes are an aspect of behavior, and not causative factors" (1948: 653).

20

Several years later, Rosenberg (1956) conducted a study very similar
to the correlational portion of Woodruff and DiVesta's, but with more
up-to-date terminology and with more measurement precision. Whereas
Woodruff and DiVesta had stated their hypotheses simply in terms of Sign
(favorable or unfavorable) Rosenberg attempted to account for degree of
favorability. He hypothesized:

The degree and Sign of affect aroused in an individual by an object

(as reflected by the position he chooses on an attitude scale)

vary as a function of the algebraic sum of the products Obtained by

multiplying the rated importance of each value associated with that

object by the rated potency of the object for achieving or blocking

the realization of that value. (1956: 367).
Rosenberg used two separate card-sorting procedures to measure "value
importance" and "perceived instrumentality." Subjects were asked to
rate a group of value statements in terms of (l) the importance of the
value as a source of satisfaction, and (2) the extent to which the value
tends to be attained or blocked through the agency of the attitude "Object"
(allowing Communists to address the public). An index was computed by
multiplying a subject's "value importance" rating by his "perceived
instrumentality" rating for each of the values, then summing this set of
products. These indices were found to be monotonically related to atti-
tudinal position. In addition to confirming the principal hypothesis,
Rosenberg also found that attitudinal position varied significantly with
"value importance" and with "perceived instrumentality" scores, when each
was considered while the other was held constant.

Rosenberg has also conducted several studies of the effect of
"affect change" on perceived instrumentality. (19603, 1960b) In a
reversal of the typical persuasion procedure which seeks to create or

alter instrumentality beliefs in order to change an attitude, Rosenberg

sought to demonstrate that hypnotically induced attitude change could

21

produce changes in beliefs concerning the value-instrumentality of the
object. He points out that this sort of process can be seen to occur
in everyday life. A positive or negative affect toward an object is
acquired or changed directly (for example, through rewarding or punishing
eXperiences with the object itself) and then beliefs concerning the
value-instrumentality of the object are acquired or invented, to "ration-
alize" the acquired affect. In a series of replicated studies using a
variety of attitude issues, Rosenberg utilized posthypnotic suggestion to
induce a reversal of attitudinal affect toward a particular object. He
found that this led to four different types of instrumentality change:
(1) a simple reversal from positive to negative or negative to positive,
(2) a reduction in extremity of an instrumentality rating without change
in its Sign (3) the reduction of an object-value relationship to the
absence of any relationship, and (4) the addition of new Object-value
relationships. He also found numerous instances of value change, in which
subjects substantially increased or decreased the rated importance of one
or more values in order to reduce cognitive inconsistency. He points
out, however, that "subjects tend to rely on instrumentality change signifi-
cantly more than upon value change in reorganizing their cognitive struc-
tures to accommodate imposed reversals of attitudinal affect" (1960b: 63).

Scott (1959) conducted a study of cognitive consistency and attitude
change, using a modified version of Rosenberg's procedures. He used
measures of value importance and perceived instrumentality to predict the
degree of favorability of each subject toward an attitude object. Subjects
whose predicted attitudes corresponded to their actual attitudes were called
"cognitively consistent" and the Others were called "inconsistent." An

attempt was then made to induce a change in attitude on the issue. It

22
was found that cognitively consistent S's were significantly more likely
to retain their initial attitudes than inconsistent S's. This result was
interpreted as indicating that "the consonance of an attitude with other
cognitive elements, such as values and expectancies, serves to stabilize
it and increase its resistance to change under externally imposed pressures"
(1959: 229). Thus one function which values may serve for attitudes is L////
to provide "bolstering power” when an attitude meets with challenge.

Carlson (1956) conducted a study designed Specifically to test the
traditional underlying assumption of rational persuasion attempts, i.e.
that attitudes toward an object may be changed through altering the person's
perception of the instrumentality of the object as a means for attaining
valued goals. He measured value importance and perceived instrumentality
using procedures Similar to Rosenberg's. Subjects were given persuasive
messages designed to increase their awareness that the Object (allowing
Negroes to move into White neighborhoods) would lead to the attainment of
four important values. On postmanipulation measures the experimental
group was found to have changed significantly both their judgments of the
instrumentality of the Object for each of the values, and their attitudinal
position toward the Object.

DiVesta and Merwin (1960) report evidence that attitudes toward
teaching as a career can be made more favorable by changing the perceived
instrumentality of teaching for the satisfaction Of the achievement need.
Two experimental groups were exposed to two different panel discussions
pertaining to the teaching profession. Both of the panel discussions
maintained that teaching is a desirable career, but one emphasized achieve-
ment-oriented satisfactions which it allegedly provides, while the other

spoke of freedom from competitiveness and the frequent vacations. It was

23

found that the achievement panel produced Significant positive changes

in the perceived instrumentality of teaching for satisfaction of achieve-
ment need and in attitude toward teaching, while the nonachievement panel
produced significant negative change on both of these dependent variables.
In a control group, there was no significant change. The correlation
between attitude change and the products of the achievement scores by the
instrumentality change scores was a low but Significant .26.

A series of studies conducted by Fishbein (1963, 19653, 1965b, 1966)
have Shown that attitudes toward a person or group of peOple can be
predicted on the basis of individuals' beliefs or cognitions about the
characteristics of the given person or persons and the value they place
on these characteristics. Subjects are asked to indicate the extent to
which Negroes (for example) possess certain characteristics, such as
dark skin, tall, athletic, lazy, etc. Next, subjects indicated on evalu-
ative semantic differential scales the value they placed on each of the
characteristics. Combining these two measures by a "sum of the products"
procedure, an estimated attitude score is computed. The correlation between
this estimated score and actual attitude scores obtained via semantic
differential procedure, was found to be quite high (.80) (Fishbein, 1963).
The correlations of Obtained attitude scores with value scores only and
belief scores only were .47 and .02 respectively. Evidently measures of
values alone or of "attributional beliefs" alone are not as effective for
estimating attitudes as the combination of the two. In addition to the
foregoing study of attitudes toward Negroes (1963) Fishbein has also
obtained high correlations in predicting group leaders' attitudes toward
group members (1965b) and individuals' attitudes toward political candi-
dates (1966), through these procedures. These studies provide strong

support for a slightly different type of belief system interconnection,

24
i.e. between values and attitudes through what might be called "attri-
butional beliefs." These may be seen to differ somewhat from what we
have termed "instrumentality beliefs." The difference has to do with the
nature of the values in question. Thus Rosenberg has pointed out that we
may distinguish two classes Of values, and differentiate their relation-
ships with attitudes:

The objects to which attitude objects are linked in relevant
cognitions tend to be of different classes depending on the

type to which the attitude object itself belongs. Thus, atti-
tudinal cognitions about liked or disliked persons and groups

are usually concerned with their positive or negative attributes

and "defining characteristics." On the other hand, cognitions

about the actions of a person or about favored or disfavored

social actions, prOposed policy changes, legislative develOpmentS, and
so on, are usually concerned with positive or negative goals or
with general values whose attainment is seen as fostered or hampered
through the agency of the attitude object. (1965: 122).

 

 

 

Thus we may readily see that Fishbein's focus is on the former type of
value-attitude relationships, while the other studies reviewed focus on
the latter. This distinction between two classes of values parallels
Rokeach's distinction between "instrumental values" (modes of behavior)
and "terminal values" (end-states of existence). ._____
Until recently, few attempts had been made to study the effect of
value change on change in related attitudes. This was probably because
it has been assumed that it would be difficult or impossible to induce
a change in value in the single, brief encounter characteristic of most
experimental treatments. However, two Studies recently conducted by
Rokeach (1971b) with the help of the present author provide evidence that
the importance of a personal value can successfully be increased through
experimental manipulation. Both studies utilized the same design and

materials. The relative importance assigned to the values Equality and

Freedom were measured by administration of a value ranking scale containing

25
these values along with a number of others. Then subjects were Shown
some results from earlier studies, and were given the exPerimenter's
interpretations of these results. The purpose Of this treatment was to
arouse an awareness of inconsistency among those who had ranked Equality
"low" on the scale, by linking Equality with Freedom and with an attitude
(Civil rights demonstrations). It was proposed to the subjects that if
they valued Freedom highly (which most college students do) they should
also value Equality highly. In addition, if they favored civil rights
demonstrations (which many of them did) they should value Equality highly.
The results of all three studies showed significant positive changes in
the relative position of Equality on post-tests taken at a number of
intervals up to 15 months later. Subjects' attitudes toward a number
of specific Equality-related issues (housing integration, school integration,
racial intermarriage, etc.) were pre-tested and later post-tested by a
different experimenter, on separation occasions, under different auSpices.
It was found that the experimental manipulation was successful in inducing
significant immediate changes in value importance. These value changes
were followed by significant attitude changes which were found to be
present on the delayed post-tests but which were not present at the time
of the immediate post-tests. In addition, the changes in values and
attitudes were accompanied by Significant changes in behavior in the
experimental subjects. The results Show a significant increase in actions
undertaken in support of the induced values and attitudes, including
responses to solicitations to join the NAACP. These studies paved the
way for the present study, by indicating that value change could be
obtained through experimental manipulation and that it does lead to changes

in related attitudes, as well as in behavior.

26
A study recently conducted by Homant (1970) found that a communication

linking an attitude object to an important value produced a significant
increase in favorability toward the attitude object. Subjects' attitudes
toward a "sliding scale tuition plan" were measured. Then one group

of subjects were given a persuasive message arguing that Sliding scale
tuition would be instrumental to the attainment of Freedom, 3 highly
important value. Another group received a message linking Sliding scale

tuition to the attainment Of A_Comfortable Life, a relatively unimportant

 

value for these subjects. It was found that attitudes toward the tuition
plan increased significantly among those who received the high value
(Freedom) communication. The outcome of Homant's study was interpreted
in terms of changes in perceived instrumentality. However, perceived
instrumentality was not measured. Therefore the presence of perceived
instrumentality changes, and the extent to which they may have led to
attitude changes, is purely a matter of Speculation. The study does Show
that an attitude may be manipulated by Specifically invoking its instru-
mental relationship to an important value.

Additional evidence regarding value change and attitude change is
provided by a study conducted by Penner (1971). White subjects were
given an experimental manipulation Similar to that used in Rokeach's
studies (1968b, 1971b) designed to increase the importance of the value
Equality. Later their attitudes and behavior toward a black person were
measured. It was found that, in comparison to a control group which
received no such manipulation, the experimental subjects showed somewhat
more favorable attitudes and behaviors toward the attitude object. The
differences were not all significant, but the results were in the predicted
direction. The results were interpreted as supportive of the relation

between value change and attitude/behavior change. As in the Homant (1970)

27
study, the degree of instrumentality between value and attitude was assumed,
rather than explicitly measured.

Definitions of Terms Used in This Research

 

In the theoretical vieWpoint which guided this investigation, values
are seen as part Of the belief system of an individual (Rokeach, 1968a).
A belief is considered to be "any simple prOposition, conscious or un-
conscious, inferred from what a person says or does, capable of being

preceded by the phrase 'I believe that...'" (19683: 113). A belief system

 

consists of "the total universe Of a person's beliefs about the physical
world, the social world, and the self" (1968:a 123). In this system, a
yglgg is "an enduring belief that a specific mode of conduct or end-state
of existence is personally and socially preferable to alternative modes

of conduct or end-states of existence" (19683: 160). A value system is

 

"an hierarchical organization--a rank ordering--Of ideals or values in
terms Of their importance" (1968: 124). Finally, an attitude is defined
as "a relatively enduring organization Of beliefs around an Object or
Situation predisposing one to respond in some preferential manner" (19683:
112).

Hypotheses

 

On the basis of the previous evidence and the theoretic rationale
presented earlier, the following hypotheses are advanced as the focus
of this inquiry:

1. A significant change in the relative importance of a particular
value to an individual may be induced by presenting a persuasive message
which makes the individual aware of an inconsistency between his own value
ranking and those of significant others.

2. An induced increase in the importance of a value will lead to

increasesiixfavorable attitudes toward objects which are perceived as

28
facilitating the attainment of that value and increased disfavor toward
Objects perceived as hindering the attainment of the value.

3. Given an induced value change, the amount of change in each
instrumentally-related attitude will be in direct proportion to the degree
of perceived instrumentality possessed by that attitude object, i.e. the
degree to which that attitude object is perceived as instrumental to the
attainment of the value.

In other words, we may change a person's attitudes indirectly, by
changing his values. An increase in the importance ascribed to a particular
value will lead to an increase in favorability toward objects seen as
facilitating the attainment of that value, and a decrease in favorability
toward objects seen as hindering the attainment of that value. (It may
also be postulated that a decrease in value importance should lead to
decreases in favorability toward instrumentally related attitude objects.
This will not, however, be tested in the present study.) The amount of
positivecnrnegative attitude change should be relatively greater in the
case of Objects which are perceived as highly instrumental, either posi-
tively or negatively, to the attainment of a value, and less in the

case of objects whose "perceived instrumentality" is relatively Slight.

29
METHOD

Overview 2f Method

An experiment was conducted to test the hypotheses, using question-
naires administered to volunteer college student subjects. There were
three experimental sessions, the first consisting of the pre-tests and
the second, one week later, including the experimental treatment and the
immediate post-tests. The third experimental session, administered to a
subgroup of the subjects one month later, consisted of the same post-tests.

Initial attitude position was measured by asking subjects to indicate

 

their degree of agreement or disagreement with each of a series of atti-
tudinal statements, each one proposing some social action. Second, initial

value importance was measured by asking subjects to rank 3 set of 18

 

values from most to least important. Third, perceived instrumentality was

 

measured by asking subjects to respond again to the same set of attitude
statements, this time by indicating to what extent they felt each social
action would help or hinder the attainment of a particular value. State-
ments pertaining to two values were included, the value to be manipulated
(AM p_f Beauty)and a "control" value (A M E Leg). This con-
cluded the first experimental session. One week later, subjects in the
experimental group were given a written persuasive message designed to
increase the importance of the value, é:fl2£l§.2£ Beauty. After receiving
the persuasive message, subjects were administered post-tests of value
importance and attitude position, using the same questions which had
been used in the pre-tests. One month later, a subgroup of the subjects
were given the post-tests for a second time, in order to measure the

persistence of the induced changes.

30

Value change was then computed by taking the difference between the

 

pre-test and post-test rankings of each of the values. Attitude change

 

was similarly computed, by taking the difference between the pre- and
post-measures for each attitude item. Perceived instrumentality was taken
directly from subjects' ratings.

Subjects

Michigan State University students enrolled in Introductory Psychology
served as subjects in this experiment. In the Introductory Psychology
course, students are required to spend a certain number of hours as sub-
jects in experiments. They choose among the available experiments by
signing their names on a sheet posted on the bulletin board which states
the nature of the experiment. The present experiment was listed as "A
Study of Values and Attitudes."

The experiment was administered in groups to a total of 320 subjects.
Subjects were randomly assigned to the experimental and control groups.
The experimental group consisted of 176 subjects, and the control group
contained 143 subjects. (The experimental group was deliberately made
somewhat larger than the control group in order to insure sufficient data
to perform the desired analytical breakdowns.) Of the subjects, 56% were
males and 44% were females. Males and females were approximately equally
distributed between the experimental and control groups. A large majority
of the subjects (57%) were freshmen, a minority were sophomores (26%),
with a small percentage of juniors and seniors (17%).

Materials and Procedure

 

Subjects were initially given a brief set of verbal instructions

(See Appendix A). They were told that that the experiment consisted of

31
two sessions, and that they should report for the second session a week
later. Then they were given a six-page questionnaire and told to read
the instructions carefully, fill it out, and return it. This question-
naire was labeled "Values and Attitudes Questionnaire” and began with a
set of instructions describing the experiment as "a scientific study of
students' values and attitudes" and soliciting their cooperation. (See
Appendix B). Subjects were first asked to answer a brief series of
"background" questions, pertaining to their age, sex, grade point average,
political and religious affiliations, etc. These questions were included
primarily to mask the pretest-treatment-posttest nature of the design by
making it appear to be a survey.

Attitude Pre-Test

 

The attitude pre-test consisted of 16 statements, each one proposing
some social action (e.g. "The U.S. should promptly and completely withdraw
its military forces from Vietnam”). Subjects were asked to "mark each
statement on the line to the left of the statement number, according to
how much you agree or disagree with it." They were provided with the

following "key":

+1 I agree slightly -1 .I disagree slightly

+2 I agree moderately -2 I disagree moderately
+3 I agree quite a bit -3 I disagree quite a bit
+4 I agree strongly -4 I disagree strongly

+5 I agree very strongly -5 I disagree very strongly

Eight attitude items pertained to A Eggld 2: Beauty, proposing actions
such as increased air and water pollution control, restrictions on hunting
and fishing, a ban on non-returnable bottles, etc. Another eight items
pertained to A HQEIQ.§E.§E§EE: and included withdrawal from Vietnam, 3

strong nuclear arsenal, increased foreign aid, 3 strengthened Peace Corps, etc.

32
The Sixteen items were mixed together in random order. Within each of
the two sets of eight "value-instrumental" items, Six were "positive,"
i.e. were intended to be perceived as facilitating the attainment of the
value, and two were "negative," i.e. were intended to be perceived as
hindering the attainment of the value. This was done in order to test
whether increased value-importance would not only lead to enhanced
favorability toward "positively instrumental" actions, but also would
lead to increased disfavor toward actions perceived as ”negatively instru-
mental" to the value in question.

Value Importance Pre-Test ,z’

 

The pre-test of value importance consisted of Rokeach's Terminal Value
Ranking Scale, which had previously been developed and standardized

(Rokeach; 19713). Subjects are given a set of 18 values (A Comfortable

 

Life, 53 Exciting Life, A Sense 2; Accomplishment, etc.--See Appendix B),

 

 

each one printed on a gummed label, along with a breif defining phrase.
Subjects are asked to peel off these labels and arrange them in rank order
of importance "for you, as guiding principles in your life." Included in
the 18 values are ‘A‘nglg'gf Beauty and A HEELS. t ngge, the two values
of particular interest in this study. Each value is accompanied by a
brief defining phrase. A Eggld gf Beauty is defined as "beauty of nature

and the arts," and A World EE Peace is defined as "free of war and conflict."

 

 

Perceived Instrumentality Test 4'
After the completion of the value-ranking task, subjects were asked

to respond again to the same attitude statements used in the attitude pre-

test. This time they were asked to rate each proposed action "according

to how strongly you feel it would help or hinder the attainment of"

33
the particular value to which it pertains. The subjects were again asked
to mark each statement on a line preceding the statement number, and were

given the following "key":

+1 would help slightly -1 would hinder slightly

+2 would help moderately -2 would hinder moderately

+3 would help quite a bit -3 would hinder quite a bit
+4 would help strongly -4 would hinder strongly

+5 would help very strongly -5 would hinder very strongly

Twelve of the sixteen actions, including the eight directly relevant to
‘A Eggld‘gf Beauty plus four additional items randomly chosen from among
the other eight, were rated according to how strongly they would help
or hinder the attainment of A Egglg_gf Beauty. Similarly, the eight items
relevant to Azflggld.§£.§e§gg plus four randomly chosen beauty items, were
rated according to their instrumental relationships to é.22£ gflgg Peace.

The purpose of obtaining instrumentality ratings for four "non-
relevant" actions was to provide, for comparison purposes, some "control"
items which Should have little or no perceived relationship to the value
in question, hence should not change as a result of a change in the
importance of the value.

This concluded the first experimental session. On the last page of
the questionnaire, subjects were given an opportunity to express any
comments they wished to make about the questionnaire.

Second Experimental Session

 

One week later, at the beginning of the second session, subjects were
again given a brief set of verbal instructions. (See Appendix C.) Then
they were given a seven-page questionnaire entitled "Values and Attitudes
Questionnaire - Part II," containing the experimental treatment and the

post-tests. This questionnaire was described as "a continuation of the

34

study in which you participated previously." It began with a set of
instructions, including the following:

We will be asking you some of the same questions again, which

we asked you in Part 1. Please answer these questions carefully

and honestly, according to how you feel about them today. Do

not worry about how you may have reSponded to them previously.
Experimental Treatment

Next, the following instructions were presented:

First, we would like to know your opinions about an issue of

major importance. Please read carefully the following selection.

After you have finished it, you will be asked to give your re-

actions to it.

This is an excerpt from a speech given recently by Dr. Milton

Rokeach, a professor of psychology at Michigan State. Dr.

Rokeach is the author of the value ranking scale used in this

questionnaire. In his speech, he discussed some findings from

a previous study using the value scale;
Following this introduction, a brief (225 words) written message was
presented, the purpose of which was to increase the importance to the
subjects of A World 2f Beauty. In this message, the source (Dr. Rokeach)
reports that the subjects in a previous study had ranked A World pf Beauty
unexpectedly low (unimportant). He expresses disappointment at this, and
briefly advocates reasons why the value Should be considered important by
Americans. He draws the readers' attention to the fact that young people
and better-educated people tend to rank A World 2; Beauty higher than the
general public, and interprets these results as grounds for cautious
optimism. (For the complete text of the persuasive message, see
Appendix D.)

Included as part of the experimental induction were five questions
immediately following the persuasive message. These questions were

designed to focus and reinforce the effect of the persuasive message

through a "Socratic" process. Subjects were asked to respond to each of

35
the following statements using the same +5 to -5 agree-disagree rating
scale used on the attitude pre- and post-tests:

l. The spoiling or destroying of our natural environment is
a serious problem in America today.

2. Especially in our large cities, the environment has already
reached the point of being detrimental to the physical and
psychological health of its inhabitants.

3. This problem will rapidly become worse unless we take im-
mediate and substantial steps to solve it.

4. Most Americans place too little importance on preserving
the beauty of nature.

5. This problem will become one of the most important ones
facing our country during the next decade.

Control subjects received a substitute written message of comparable
length which was followed by five "Socratic” questions. This message and
the accompanying questions pertained to an unrelated topic (child rearing
practices and student activism) and were designed to have no effect on
their value for A Egglg 9f Beauty and their attitudes toward the environ-
mental issues. Questionnaires containing the "experimental" and "control"
messages were randomly mixed together before passing them out to the
subjects in group administrations.

Value Importance Post-Test

 

The post-test of value importance consisted of the same terminal value
ranking scale administered in the pre-test, using the same instructions.
Attitude Post-Test

The attitude post-test consisted of the same 16 items used in the
pre-test, and used the same +5 to -5 scale of agreement, disagreement.

Matching Opestions

 

Next, the following instructions were presented:

36

Your answers to the following three questions will enable us
to ”match" this questionnaire with the one you filled out pre-
viously (Part 1) without requiring you to reveal your name.

It is very important that you answer these questions again,
even though you already answered them once before.

 

Then subjects were asked to indicate their sex, place of residence, and
date of birth, which had been asked at the beginning of the Part 1 ques-
tionnaire. Subjects were not asked to give their names on either of the
two questionnaires, in order to preserve the protection of their privacy
through anonymity.

Finally, subjects were again offered the opportunity to comment on
the questionnaire if they wished. This concluded the second experimental
session.

Third Experimental Session

A subsample of subjects were asked to report back for a third session
one month later. These subjects were selected simply on the basis that
they were the first batch of subjects who were run through the experiment.
There were 70 subjects solicited for this purpose and of these, 61 were
eventually successfully contacted and re-tested. Of these, 39 turned
out to be experimental subjects and 22 were control subjects. At the time
of the third experimental session, they were given a second post-test,
consisting of the same set of value and attitude post-test items which
were used in the first post-test, using the same instructions and procedures.
Constructed Variables

An index of value change was constructed by subtracting each subject's
post-test ranking of each of the values from his pre-test ranking of the
same value. Thus if he ranked é:32£l§H2£ Beauty 16th in order of importance

initially and 8th on the post measure, this would represent a change of +8.

37

An index of attitude change was constructed for each of the 16 attitude

 

items by subtracting a subject's pre-test rating from his post-test

rating.

38

RESULTS

Value Change

 

The results pertaining to Hypothesis 1 are of immediate and primary
importance, since these results will indicate whether the experimental
manipulation did in fact succeed in producing significant value change.
Shown in Table l are the mean pre-test to immediate post-test changes in
value rankings for all eighteen values, including the manipulated value,
.é.H2£l§.2£ Beauty. It Should be noted that the signs of these changes
are reversed, so that a change "upward" toward increased value importance
is represented by a positive number.

It may be seen that the experimental subjects changed their rankings

Of A World 2f Beauty 3.34 units in a positive direction. Their mean

 

ranking of the value on the pre-test was 11.77 and their mean post-test
ranking was 8.43. The control group changed from a pre-test mean of 11.09
to a post-test mean of 11.00, a positive change of .09 units. A one-way
analysis of variance of the change scores was conducted, and it was found
that the difference in value change between experimental and control groups
is highly significant (p <1 .001).* Significant differences between the
two groups were also found on three of the other eighteen values. It

may be seen that the control group's mean importance ranking of A Egglg

32 Peace decreased .76 units, and their rankings of Family Security and

 

*In this analysis and all subsequent analyses, results are presented in
terms of mean changes, and statistical comparisons are made in terms of
analyses of variance of change scores. Preliminary analyses of variance
conducted for all values and attitudes indicated that the experimental

and control groups were not Significantly different in their pre-test
means. However, it was felt that, rather than using post-test comparisons,
it would be simpler and more accurate to present the mean changes and the
accompanying statistical tests.

39

Table 1. Mean changes in rank order of 18 terminal values for experimental
and control groups.

 

 

Values Experimental Control
(N=l76) (N=l43)
A comfortable life -.54 -.15
An exciting life -.11 .41
A sense of accomplishment -.23 -.34
A world at peace .44 -.76**
A world of beauty 3.34*** .09
Equality .38 ..03
Family security -.12 .60*
Freedom -.15 .55*
Happiness -.53 -.31
Inner harmony -.22 .11
Mature love -.31 .23
National security -.10 .15
Pleasure -.23 -.02
Salvation -.45 -.15
Self-respect -.43 -.14
Social recognition -.33 -.11
True friendship -.40 -.15
Wisdom -.Ol -.03

 

*p < .05, **p‘( .01, ***p‘< .001, significant difference between experi-
mental and control groups, one-way,analysis of variance of change scores.

40

Freedom increased .60 units and .55 units, respectively. These changes
in the control group may be due to the "control" message used in place
of the "experimental treatment" message. In any case they are not im-
portant to the purpose of this study. Changes among the remaining
values were slight, and did not differ Significantly between the two groups.

In summary, the results shown in Table 1. indicate clearly that the
experimental manipulation was successful in producing a highly significant
positive change in the experimental subjects' rankings of é;22£l§.2£
Beauty, on a post-test administered immediately following the experimental
manipulation.

Value Change - Delayed Post-Test

 

The results shown in Table 2 were obtained from subgroups of 39
experimental and 22 control subjects who were tested immediately following
the experimental manipulation along with all other subjects, and who were
additionally retested approximately one month later. Table 2 shows the
value changes for these groups on both the immediate and delayed post-
tests. These results will be relevant to our subsequent presentation of
attitude change results pertaining to Hypothesis 2.

As may be seen in the table, the experimental subjects had changed
their rankings oflA'W931g_gf Beauty 3.31 units at the time of the immediate
post-test and 3.87 units on the delayed post-test. Both these changes
were highly significant (p«< .001) according to a one-way analysis of
variance comparing the differences in amount of change between experimental
and control groups.

None of the other 18 values changed significantly on either post-test,
except for Pleasure, which was changed in a positive direction by the

control group at the time of the second post-test.

Table 2. Mean changes in rank order of 18 terminal values for subgroups

of experimental and control subjects tested immediately and

one month later.

 

 

 

Values Immediate Delayed

Exper. Control Exper. Control

(N=39) (N=22) (N=39) (N=22)
A comfortable life -l.25 -.32 -.46 -.41
An exciting life -.72 .41 -.67 .14
A sense of
accomplishment -.54 -.59 -1.26 -.18
A world at peace .92 -.23 1.51 -.41
A world of beauty 3.31*** .55 3.87*** -.14
Equality .18 .05 .62 -.36
Family security .13 -.18 .10 -.27
Freedom -.92 -.73 -.28 -.19
Happiness -.79 .23 -.21 -.18
Inner harmony 1.03 -.32 .51 .27
Mature love -.56 .41 -.69 1.05
National security -.33 .09 -.36 -.59
Pleasure .36 .82 -.41 1.32*
Salvation -.67 -.36 -.74 -.59
Self-respect .05 1.32 -.54 .68
Social recognition -.02 .18 .31 -.27
True friendship -.10 -1.05 1.18 .14
Wisdom -.O3 -.27 -.13 -.01

 

*p <..05, **p < .01, ***p'< .001, significant difference between experi-
mental and control groups, one-way analysis of variance of change scores.

42
It Should be noted that there was no substantial "delayed action
effect" as far as value change was concerned. The immediate and the
delayed changes in importance 0f.éjflfl£l§.2£ Beauty are not substantially
different. Thus the experimental manipulation apparently took effect
immediately, and this effect was sustained over time, increasing slightly
during the month following the manipulation.

Attitude Change

 

Table 3 shows the mean changes in eleven* attitudes for the experi-
mental and control groups. Above the line are the seven attitudes rele-
vant to A 22319 pf Beauty, and below the line are the four "control"
items. It may be seen that the experimental manipulation was successful
in producing significant changes in three of the seven attitudes. For
attitudes toward highway beautification, banning automobiles from cities,
and prohibiting nonreturnable bottles, the experimental group changed
Significantly more than did the control group, in a positive (more favorable)
direction. For the two negatively-worded attitude items, unrestricted
hunting and fishing and allowing trash burning, for which it was predicted
that the experimental group would change negatively (i.e. become less
favorable) the experimental group changed in a positive direction but
less than the control group. The fact that on both of these items the

experimental group changed very little and changed less than the control

 

*Here and elsewhere we originally planned to show attitude change results
for twelve items, i.e. all eight Beauty items and the additional four
"control" (Peace) items which were included in the instrumentality ratings.
Thus we are concerned with all twelve items which were rated for per-
ceived instrumentality. However, of the original twelve items, one,
pertaining to water pollution, was found to be suffering from a severe
ceiling effect on the pre-test, having been initially rated in the most
extremely favorable category by a majority of the subjects. This pre-
cluded its use as a measure of attitude change. Therefore it was excluded
from the analysis.

43

Table 3. Mean changes in agreement-disagreement ratings of 11 attitude
statements for experimental and control groups.

 

 

Attitude Experimental Control
(N=l76) (=143)

 

 

Highway beautification .69* .13
Unrestricted hunting and fishing .20 .41
Ban automobiles from cities 1.46** .76
Abolish neon Signs .78 .49
Prohibit nonreturnable bottles .56** -.22
Allow trash burning .02 .26
Limit children to two .70 .41
Increase economic aid .34 .13
Fight communist aggression -.19 -.14
Liberate women .08 -.15
Legalize marijuana -.13 -.01

 

*p‘< .05, **p14 .01, significant difference between experimental and
control groups, one-way analysis of variance of change scores.

44
group may mean that the intended effect was present to some degree but
was overbalanced by other influences such as agreement bias.

For the remaining two attitudes, toward neon signs and population
control, the experimental group changed a greater amount than did the
control group and in the predicted direction, but these changes were not
statistically Significant. The fact that some attitudes changed more
than others was an anticipated and intended effect of this study. The
differences in amounts of change among the various attitudes are a function
of their differing amounts of perceived instrumentality, as will be shown
in the subsequent section dealing with the relation between instrumentality
and change.

Hypothesis 2 is partially supported by the results Shown in Table 3.
The hypothesis stated that an induced increase in the importance of a value
will lead to increases in attitudes toward objects which are perceived as
facilitating or hindering the attainment of that value. The results
presented thus far show significant positive changes toward objects (social
actions) seen as facilitating the attainment of A HQElQ.2£ Beauty. They
do not Show significant negative changes toward objects seen as hindering
the attainment of the value. However, results to be presented below will
help to clarify this relationship. In addition, the following results will
further clarify a particular implication of Hypothesis 2.

Attitude Change - Delayed Post-Test

Table 4 shows the immediate and delayed attitude change results,
taken from the same subgroups of subjects whose changes in values were
shown in Table 2. The obtained differences between experimental and control
groups were significant in only one case, that of the attitude toward

prohibition of non-returnable bottles. At first glance, it may seem

45

Table 4. Mean changes in attitudes for subgroups of experimental and
control subjects tested immediately and one month later.

 

 

 

 

Attitudes Immediate Delayed
Exper. Control Exper. Control
(N=39) (N=22) (N=39) (N=22)
Highway beautification .26 .02 .64 .01
Unrestricted hunting and
fishing .05 .45 -.05 1.03
Ban automobiles from cities 1.41 .68 2.08 1.14
Abolish neon signs .82 .63 1.62 1.18
Prohibit nonreturnable
bOttleS 058).: -095 087* -041
Allow trash burning .77 .18 .15 -.09
Limit children to two .31 -.05 .49 .23
Increase economic aid -.18 -.05 .28 .23
Fight communist aggression -.97 -.86 -.87 -1.18
Liberate women -.64 .55 -.44 .09
Legalize marijuana .08 .14 .36 -.23

 

*p‘< .05, significant differences between the experimental and control
groups, one-way analysis of variance of change scores.

46

strange that the changes in attitudes toward bottles are significant
while the other numerically larger changes are not significant. However,
it may be seen that for the bottles item, the control group changed in
a negative direction while the experimental group was changing positively.
Thus the difference between experimental and control groups is actually
largest for the bottles item. The failure of the remaining items to
achieve statistical significance may be due to the relatively small N's
of these two groups (39 and 22). However, the results are crucial in
that they Show the presence of positive change and the presence of a
delayed action effect on attitude change. For all but one of the seven
relevant attitudes, the experimental group changed in the predicted
direction more than did the control group. And in every case the experi-
mental group had changed their attitudes substantially more at the time
Of the delayed post-test than they had on the immediate post-test. For
example, highway beaUtification had changed .26 units immediately and
.64 units a month later. Banning automobiles had changed 1.41 and 2.08
units respectively. Abolishing neon signs had changed .82 units imp
mediately and 1.62 units during the following month. These results run
strongly contrary to the usual findings of regression toward initial
positions in delayed post-tests of attitude change.
The Relation Between Value Change and Attitude Change

If value change produces attitude change, then those subjects who
changed their rankings of A H2311 9; Beauty the most should also have
changed their attitudes significantly more than those subjects whose value
rankings changed very little. These results are shown in Table 5. A
comparison was made between subjects who changed their rankings of‘A
y9;;g_9§ Beauty very little (one unit or less), somewhat (2 to 5 units),

or substantially (6 units or more). Table 5 shows that in nearly every

47

Table 5. Mean changes in attitude for experimental subjects grouped
according to amount of value change.

 

 

 

 

Attitude Low (21) Med (2-5) High (£6)
Highway beautification .15 .79 1.04*
Unrestricted hunting and fishing .28 .21 .13
Ban automobiles from cities .72 1.53 2.02**
Abolish neon signs .52 .86 .92
Prohibit nonreturnable bottles .05 .63 .98*
Allow trash burning .58 -.12 -.38*
Limit children to two .83 .38 .89
Increase economic aid .55 .52 -.02
Fight communist aggression .11 -.25 -.40
Liberate women .19 .02 .05
Legalize marijuana .02 -.22 -.17

 

*p‘< .05, **p <f.01, Significant difference between the three groups,
one-way analysis of variance of change scores.

48

case those whose values changed more changed their attitudes more than did
those whose values were not changed as substantially. These differences
were significant in four instances, including a significant negative
change on the trash burning issue, which was in the predicted direction.
Again, as in the previous comparisons, one-way analyses of variance were
used to compare the differences in the changes of the three groups. The
differences between the groups were found to be significant for the atti-
tudes toward highway beautification, banning automobiles from cities,
prohibiting nonreturnable bottles, and allowing trash burning. For the
attitudes toward unrestricted hunting and fishing, abolishing neon signs,
and limiting family Size, the differences were not significant. These
results confirm the relation between value change and attitude change in
the sense that the more an individual changes his value ranking, the more ,-
he changes his attitudes.
Perceived Instrumentality and Attitude Change

One of the foremost interests of this study is in the question of
perceived instrumentality and attitude change. It was argued that, given
an induced value changeenuiresultant attitude changes, those attitudes
which change the most should be those which are perceived to be most
instrumental to the value. Results pertaining to this hypothesis are
found in Table 6. Subjects had rated each of the eleven social actions
according to its degree of value-instrumentality (i.e. how much it would
help or hinder the attainment of‘A‘nglg.gf Beauty). For each of these
attitudinal issues, the mean instrumentality rating was computed, for
the experimental group. In addition, an "increment attitude change" was
computed for each of the issues, by subtracting the mean attitude change

of the control group from the mean change of the experimental group.

49

Table 6. Mean instrumentality ratings and mean incremental changes in
attitudes of experimental subjects. (N=l76)

 

 

Attitude Mean Instru- rank Mean Incre- rank
mentality order mental change order

 

Prohibit nonreturnable

 

bottles 3.30 l .78 1
Highway beautification 3.18 2 .56 3
Ban automobiles

from cities 3.01 3 .70 2
Limit children to two 1.91 4 .28 5
Abolish neon signs 1.77 5 .29 4
Increase economic aid 0.67 6 .21 7
Liberate women 0.38 7 .23 6
Legalize marijuana 0.15 8 -.12 8
Allow trash burning -3.99 1 -.24 1

Unrestricted hunting
and fishing -3.73 2 -.21 2

Fight communist
aggression -1.13 3 -.05 3

 

The Spearman rank-order correlation between mean instrumentality and mean
incremental change for the eight positively-instrumental items is .94.

50
The resultant figures comprise the most accurate representation of the
effect of the experimental manipulation, since they represent the incre-
ment of change produced in the experimental group, using the control
group as a baseline. Finally, both the instrumentalities and the incre-
ment changes were ranked in order from largest to smallest. The results
are shown in Table 6. The eight positively instrumental items are listed
and ranked separately from the three negatively-instrumental items. This
was done because the changes occurring in the negative items are of a
different order of magnitude than those occurring in the positive ones.
For example, the item with the highest positive instrumentality (3.30)
changed .78 units while the item with the highest negative instrumentality
(3.99) changed only .24 units. Thus the changes were much less for the
negative items. Evidently other influences (such as agreement bias) were
present which partially counteracted the effects of the manipulation on
the negatively-instrumental attitudes. It should be noted, however,
that the changes taking place in the negative items are in the right
direction, i.e. toward decreased favorability, as was predicted.

The Spearman rank-order correlation between instrumentality means
and change means for the eight positive items is .94. For the three
negative items, instrumentality and change are ordered correspondingly.
These results indicate a very high degree of correspondence between the
relative degrees of perceived instrumentality and attitude change. The
greater the instrumentality, the greater the attitude change. Thus ~~
Hypothesis 3 is supported by these results. It appears that the degree
of attitude change produced by a change in value importance is affected
markedly by the extent to which the attitude object in question is seen

as instrumental to the value in question.

51
Referring back to Table 3 for a moment, it may be seen that the
three attitudes which changed significantly were those with the highest
instrumentality ratings (i.e. mean instrumentality ratings of 9.30, 9.18,
and 9.01, while the two positive attitudes which did not change signi-
ficantly had much lower instrumentality ratings (i.e. 7.91 and 7.77).
It is apparent that the degree of perceived instrumentality is a good

predictor of the attitude change following a change in value importance.

52
DISCUSSION

Value Change Through Persuasion

 

This study was conducted as part of an ongoing series of studies at
Michigan State University investigating experimentally induced changes
in values, attitudes, and behavior. Rokeach (1971) has pointed out that
there are two well-known methods of inducing changes in an individual's
attitudes or values. These are (a) to induce him to engage in behavior
incompatible with his attitudes or values, and (b) to expose him to
information about the attitudes or values of Significant others which
are incompatible with his own. A third approach, which has been used in
the previous studies in this series, has been to expose the individual
to information designed to bring to his awareness states of inconsistency
existing within his own value-attitude system. This third type of ap-
proach is in contrast to the traditional persuasive message, in that it
primarily emphasizes exposing the individual to information about himself
rather than information about others or their views. This approach has
been used in the previous attempts to experimentally induce changes in
value importance, and has been highly successful.

The question arises, then, as to whether values, being more funda-
mental, stable, and resistant to change than attitudes, require special
techniques of experimental manipulation, such as the "revealed inconsis-
tency" approach described above. Or are values amenable to the more
"traditional" types of persuasion? The present study offers the first
opportunity to answer this question, since it is the first of the series
of studies to utilize a traditional persuasive message.

Apparently value importance can be successfully changed using

conventional persuasive techniques. Our results show that the importance

53
assigned to A Eggld pf Beauty was significantly changed immediately
following the persuasive message and this change had persisted and slightly
increased one month later. Previous studies have shown that changes in
value importance are highly persistent over periods of a year or more
(Rokeach, 1971b). Thus we may conclude that we may induce Significant and
persistent changes in value-importance through persuasion.
Value Change,yPerceived Instrumentality, and Attitude Change

Previous studies have demonstrated the existence of perceived
instrumentality as a linkage between values and attitudes (Rosenberg,
1956, 1960, Fishbein, 1966). Our results show that a change in value
importance did produce changes in attitudes toward related issues. MOre-
over, the results showed that attitudes changed toward those actions
which were seen as highly instrumental to value attainment. Thus the
study provides further support for the conception of instrumental linkages
between values and attitudes. Given an increase in value-importance,
the more a particular action was seen as facilitating the attainment of
that value, the more the individual's attitude toward that action changed
toward increased favorability.

There was also a tendency, though not as clear-cut, for individuals
who had changed their values to become less favorable toward actions seen
as hindering the attainment of the value in question. Thus it would seem
that it is possible to change a number of attitudes at once, through a
single manipulation, by directing that manipulation toward a value to
which the attitudes are seen as instrumental.

The manipulation may seem to have succeeded only partially, in that

some of the related attitudes changed significantly but others did not.

54

However, it is important to note that this was an anticipated and in-
tended effect of the experimental design. Attitude items were purposely
preselected to represent attitudes both high and low in perceived in-
strumentality. This was done in order to show that a manipulated value
change would lead to changes only in attitudes strongly instrumentally
related to the value and not in attitudes weakly or negligibly related.
Thus the mixture of significant changes, slight changes, and unchanged
attitudes is entirely consistent with our expectations.
Delayed Action Effects in Value and Attitude Change

The results of this study pertaining to delayed effects are not
entirely unequivocal, since because of the small N's tested, they did
not achieve statistical significance. However, they show a partial trend
toward increased attitude change over time, as a result of immediate
value changes. These delayed increases in attitude change run counter
to the usual tendency for attitude changes to regress on delayed post-
tests. It would appear that the value changes tended to take place
immediately, as a result of the experimental manipulation which was
aimed directly at the value itself and did not mention the attitudinal
issues. This change in value importance then apparently led to immediate
changes in attitudes, which were also present at the time of the delayed
post-test. There was a moderate (though not significant) tendency for
attitude change to increase over time. However, we should emphasize that
there were also significant immediate changes in attitudes as well,
and that the delayed changes are only present as an increase beyond
the immediate changes. Thus it would not be proper to conclude that

the value change-attitude change process is entirely a matter of

55
delayed action. The finding of immediate attitude changes in this study
is in conflict with some earlier results. In the study by Rokeach (1971b)
described earlier, attitude change did not take place immediately. In
fact, there was a slight negative change (boomerang effect) in the im-
mediate post-test, which was then followed by a significant positive
change on the delayed post-tests. By comparison, in Rokeach's study‘
the attitude change took much longer to take effect, whereas in the
present study it took place immediately and then continued to increase
over time. Perhaps this difference is due to the fact that in Rokeach's
study there was a greater distance in belief space between the manipulated
cognition (value) and the indirectly affected cognitions (attitudes).
Rokeach used civil rights attitudes and the value Equality, while the
present study used A Eggld 9: Beauty and related attitudes toward ecological
issues. A.more likely explanation is that civil rights attitudes are
more entrenched and resistant to change. Issues such as banning non-
returnable bottles, prohibiting trash burning, etc., were quite new and
had not been subjected to public debate at the time that this study was
conducted. Thus attitudes on these issues were less well formed and well
defended, hence more easily and quickly changed.

Both in the present study and in Rokeach's earlier studies (1971b)
attitude change increased over time. The results of studies of attitude
change typically show decreases over time as the effects of the manipulation
"wear off" and individuals' attitudes regress toward their initial posi-
tions. It is extremely interesting to note that in this study and in

Rokeach's studies the opposite effect occurred.

56
Changing Attitudes Through Value Change

On the basis of these findings, it may be argued that the "indirect"
approach to changing attitudes (i.e. through changing values) used here
is a potentially powerful method in that it results in more lasting
changes. The theoretical rationale for this conclusion is as follows:
First, it has been theoretically argued that value changes are more en-
during than attitude changes, because values are more fundamental and
stable elements in the cognitive system. (Rokeach, 19683). Thus we may
logically deduce that an attitude change brought about through a related
value change will be more enduring than an attitude change induced "directly."

In addition, it may be argued that one likely cause of "regression"
in attitude changes is value-inconsistency. If an individual changes his
attitude without changing the importance he assigns to the value or values
to which that attitude is instrumental, and without changing his "perceived
instrumentalities," this will result in inconsistencies in his value-
attitude system which will lead to a tendency to return to his original
attitude position in order to restore consistency. This effect may be a
testable, predictable one, and bears consideration as a focus of further
inquiry.

It has earlier been suggested that persons wishing to effect lasting
changes in attitudes might be more successful by directing their attention
toward attempts to change values rather than limiting their efforts to
changing attitudes themselves. The results of this study lend direct

support to this assertion.

57

Conclusions

 

The results of this study indicate that (3) individuals' values
are amenable to influence through traditional persuasive'methods,

(b) changes in values bring about changes in attitudes toward instru-
mentally related objects, and (c) these changes in attitudes do not
decrease over time. Thus it would seem to illustrate a potentially
powerful means of persuasion and influence. First, changing attitudes
indirectly through changing values would appear more economical for
persuaders, in that a number of attitudes may be influenced "in one
fell swoop” through the manipulation of a single value. Second, it
would appear that attitude changes obtained through this indirect
method may be more lasting than attitude changes obtained without re-
gard to (or perhaps even in spite of) the individual's values.

Rokeach (1971) has previously discussed the ethical problems
involved in direct attempts to influence values. Restraint and caution
may be appropriate in setting out to manipulate values in the experimental
laboratory. However, in the world outside the experimental laboratory
attempts to influence values proceed apace. Every day, values are
brought under great influence and pressure, often indirectly but some-
times directly and explicitly, in the churches, schools, and political
forums of our society. It is this author's view that the most important
effects of persuasion are the effects on individual's values. For
example, every commercial message presented over television extols a
product, directly or indirectly, in terms of that product's instru-
mentality to the attainment of a value, such as cleanliness, good health,

pleasure, excitement, social acceptance, etc. In so doing, these

58
commercials constantly reinforce, by implication, the importance of
the values. It may be that the effects of television commercials
(and many other types of persuasive messages) in evoking and reinforcing
values is far more powerful and lasting than the obviously intended
effects on attitudes toward a particular product or brand. These
effects on values may often be unintended by the communicator. Never-
theless they occur, and they have been neglected by social psychologists
and other students of persuasion, attitude change, and social influence.
In our emphasis on studies of attitude change, we have seemingly over-
looked the more powerful, though subtle, effects on values which are a
result of persuasion. We as yet know little about this potent source
of influence. It should be obvious that further study of the process
and effects of value change will prove eminently useful and beneficial
to those who seek to understand and predict changes in attitudes and
behavior and those who set out to change others' attitudes and behavior

through persuasion.

REFERENCES

Abelson, R. and Rosenberg, M. Symbolic psycho-logic: a model of attitudinal
cognition. Behav. Sci., 1958, 3, 1-13.

 

Carlson, E. Attitude change through modification of attitude structure.
A, abnorm. soc. Psychol., 1956, 52, 256-261.

Cartwright, D. and Harary, F. Structural balance: a generalization of
Heider's theory. Psychol. Rev., 1956, 63, 277-293.

 

Dillehay, R., Insko, C., and Smith, M. Logical consistency and attitude
change. J, Pers. soc. Psychol., 1966, 3, 646-654.

 

Di Vesta, F., and Merwin, J. The effects of need-oriented communications on
attitude change. A, abnorm. soc. Psychol., 1960, 60, 80-85.

Festinger, L., A theory 9f_cognitive dissonance. Stanford: Stanford
University Press, 1957.

 

Fishbein, M. An investigation of the relationships between beliefs about
an Object and the attitude toward that object. Hum. Relat., 1963,
16, 233-239.

 

Fishbein, M. A consideration of beliefs, attitudes, and their relationship.
In I. D. Steiner and M. Fishbein (Eds.), Current studies £3 social
psychology. New York: Holt, Reinhart, & Winston, 1965 (a).

 

 

Fishbein, M. Prediction of interpersonal preferences and group member
satisfaction from estimated attitudes. A, pers. soc. Psychol.,
1965, 1, 663-667(b).

 

Fishbein, M. Beliefs, attitudes, and behavior. In S. Feldman (Ed),
ngnitive consistengy: motivational antecedents and behavioral
consequents. New York: Academic Press, 1966.

 

 

Heider, F. Attitudes and cognitive organization. A, Psychol., 1946,
21, 107-112.

Heider, F. The pgychology g: interpersonal relations. New York: Wiley,
1958.

 

 

Homant, R. Values, attitudes, and perceived instrumentality. Unpublished
doctoral thesis. 1970 Michigan State University.

Hovland, C., Lumsdaine, A., and Sheffield, F. Experiments 92 mass communi-
cation. Princeton: Princeton Univ. Press. 1949.

 

 

McGuire, W. A syllogistic analysis of cognitive relationships. In C.
Hovland and M. Rosenberg (Eds.), Attitude organization and change.
New Haven: Yale Univ. Press. 1960(3), 65-111.

 

McGuire, W. Cognitive consistency and attitude change. A, abnorm. soc.
Psychol. 1960(b), 60, 345-353.

 

59

60

McGuire, W. Direct and indirect effects of dissonance-producing messages.
A, abnorm. soc. Psychol. 1960(c), 60, 354-358.

Newcomb, T. An approach to the study of communicative acts. Psychol.
Rev., 1953, 60, 393-404.

Osgood, C., and Tannenbaum, P. The principle of congruity in the prediction
of attitude change. Psychol. Rev., 1955, 62, 42-55.

 

Penner, Interpersonal attraction toward a black person as a function of
value importance. Personality, 1971, 2, 175-187.

 

Pintner, R. A. comparison of interests, abilities, and attitudes.
A. abnorm. soc. Psychol., 1933, 27, 351-357.

Rokeach, M. (Ed), The gpen and closed mind. New York: Basic Books, 1960.

Rokeach, M. Beliefs, attitudes, and values. San Francisco: Jossey-Bass.
1968(3).

 

Rokeach, M. A theory of organization and change within value-attitude
systems. Journal 2: Social Issues, 1968, 24, 13-33. (b).

Rokeach, M. The measurement of values and value systems. In G. Abcarian
& J. W. Soule (Eds.) Social psychology and political behavior.
Columbus, Ohio: Charles E. Merrill, 1971(3).

Rokeach, M. Long-range experimental modification of values, attitudes,
and behavior. Amer. Psychol. 1971, 26, 453-459 (b).

 

Rokeach, M., Reyher, J., and Wiseman, R. An experimental analysis of the
organization of belief systems. In M. Rokeach (Ed), Values, attitudes,
and behavior. San Francisco: Jossey-Bass. 1968, 22-61.

 

Rosenberg, M. Cognitive structure and attitudinal affect. A, abnorm. soc.
Psychol., 1956, 53, 367-372.

 

Rosenberg, M. An analysis of affective—cognitive consistency. In C. Hovland
and M. Rosenberg (Eds.), Attitude organization and change. New Haven:
Yale Univ. Press, 1960(3), 15-64.

Rosenberg, M. Cognitive reorganization in response to the hypnotic
reversal of attitudinal affect. A, Pers., 1960(b) 28, 39-63.

Scott, W. Cognitive consistency, response reinforcement, and attitude
change. Sociometry, 1959, 22, 219-229.

Smith, G. The interrelationships of attitudes toward Russia and some
general desires. A, Psychol., 1946, 21, 91-95.

Smith, M. Personal values as determinants of a political attitude. J:
Psychol., 1949, 28, 477-486.

61

Stotland, E., Katz, D., and Patchen, M. The reduction of prejudice through
the arousal of self-insight. A, Pers., 1959, 27, 507-531.

Tannenbaum, P. Mediated generalization of attitude change via the principle
of congruity. J, Pers. soc. Psychol., 1966, 3, 493-500.

 

 

Tannenbaum, P. and Gengel, R. Generalization of attitude change through
congruity principle relationships J, Pers. soc. Psychol., 1966, 3,
299-304.

Vernon, P. and Allport, G. A test for personal values. A, abnorm. soc.
Psychol., 1931, 36, 231-248.

Woodruff, A. Personal values and the direction of behavior. Sch. Rev.,
1942, 50, 32-42.

Woodruff, A. and DiVesta, F. The relationship between values, concepts,
and attitudes. Educ. psychol. Measmt., 1948, 8, 645-660.

APPENDIX A

Verbal instructions for session one

 

Hello, my name is Charles Hollen. I'm conducting the experiment on
values and attitudes that you signed up for. This experiment involves
two sessions, one now and another one a week from today. Will it be
possible for you to come back here a week from now at this same hour? All
right.

The experimental task we're asking you to do is to fill out this
questionnaire. This is not a test. It's a method we use to obtain
information for our research. Also we hope that the time you spend thinking
about your Opinions and your goals in life in order to answer these ques-
tions will be a beneficial learning experience for you. Many of the
students who have previously filled out the questionnaire commented that
it was a very thought-provoking experience for them.

You are not required to Sign your name to the questionnaire if you
don't want to. This is to assure you that the information you are giving
us will be confidential. We aren't going to be asking you anything particu-
larly personal, but we make it a policy not to require names so that peOple
will feel free to give their honest Opinions.

Please read the instructions very carefully before you begin. If
you have any questions feel free to come and ask me. When you have
finished, bring the questionnaire back to me.

Don't forget that in order to complete the experiment and receive
your credit for it, you must report back for the second session a week

from now.

62

APPENDIX B
VALUES AND ATTITUDES QUESTIONNAIRE

Instructions - Please read carefully before proceeding further.

 

This is a scientific study of students' values and attitudes. Your honest
cooperation would be deeply appreciated. In addition to providing us

with information, we hOpe that this questionnaire may help you to learn
something about yourself.

In the course of filling out the questionnaire, you will be asked to
answer the same qdestions several times in different contexts. Although
this may seem repetitious, it is necessary so that we may make certain
comparisons. Each time you answer a particular set of questions, please
answer them carefully and honestly, without worrying about how you may
have responded to them previously.

Each set of value and attitude questions is preceded by some instructions.
Please read the instructions carefully and make sure you understand them
before answering any questions. Please answer each page of the question-
naire in the order in which it is given; do not skip ahead or turn back.

First, we are interested in knowing something about your background.
Please answer all the questions asked below.

 

 

 

 

 

 

 

 

 

Sex: Place of Residence: city state

Date of Birth: month day year

Grade Point Average: Class: Freshman
SOphomore
Junior
Senior

 

Which of the following most closely exemplifies your political vieWpoint
in general?

 

 

 

 

Radical
Liberal
Middle-of—the-road
Conservative
C
Religious DO you attend religious
Preference: None services? Never
Catholic A few times a year
Jewish About once a month
Protestant Twice a week or Oftener

63

64

Given the choice between reading a good book or attending a party, which
would you rather do?

Read a book
Attend a party

Below you will find 18 statements, each one prOposing some social action.
We would like to know how you feel about each of these issues. We have
tried to cover different and Opposing points of view. You may find your-
self agreeing with some of the statements and disagreeing with others.
Whether you agree or disagree with any statement, you can be sure that
many other peOple feel the same as you do.

Mark each statement on the line to the left of the statement number,

according to how much you agree or disagree with it. Mark it with a + (plus
Sign) if you agree or a - (minus Sign) if you disagree, and a number from 1

to 5 according to how strongly you feel in each case, from 1 (slightly) to

5 (very strongly). Thus, for example, if you agree slightly with a statement,
mark it +1; if you disagree very strongly, mark it -5, etc.

Please read each statement carefully and be sure you understand its meaning
before responding to it.

+1 I agree slightly -1 I disagree slightly

+2 I agree moderately -2 I disagree moderately

+3 I agree quite a bit -3 I disagree quite a bit
+4 I agree strongly -4 I disagree strongly

+5 I agree very strongly -5 I disagree very strongly

1. The U.S. Should promptly and completely withdraw its military
forces from Vietnam.

2. Factories which pollute the air or water should be forced to
clean up or shut down.

3. The U.S. should appropriate the necessary funds for a continued
strong and active program of Space exploration.

4. More funds should be Spent on highway beautification (land-
scaping, etc.).

5. The United Nations should be given stronger powers to help
settle disputes among member nations.

6. Restrictions on hunting and fishing should be decreased.
7. Automobiles should be banned from the downtown areas of cities.

8. The U.S. should continue to build up its arsenal of nuclear weapons,
as a deterrent force against potential aggressors.

9. The Peace Corps should be expanded and strengthened.

65

+1 I agree slightly -1 I disagree slightly

+2 I agree moderately -2 I disagree moderately

+3 I agree quite a bit -3 I disagree quite a bit
+4 I agree strongly -4 I disagree strongly

+5 I agree very strongly -5 I disagree very strongly

10. Neon Signs should be abolished.

 

11. The U.S. should increase its economic aid to develOping
countries.

 

12. Nonreturnable bottles should be prohibited.

 

13.

 

14.

 

15.

 

16.

 

l7.

 

18.

 

The U.S. and Russia Should stop experimenting with chemical
and biological warfare.

All forms of discrimination against women should be ended.
Penalties for the possession of marijuana should be reduced.
People Should be allowed to burn trash or leaves.

Families should limit themselves to two children.

The U.S. Should continue helping other nations such as Vietnam
in their struggles against Communist aggression.

 

VALUE SURVEY - INSTRUCTIONS

On the next page are 18 values listed in alphabetical order.

We are inter-

ested in finding out the relative importance of these values for you, as
guiding principles in your life.

Each value is printed on a gummed label which can be easily peeled off and
pasted in the boxes on the left-hand side Of the page.

Study the list carefully and pick out the one value which is most important

for you.

Then pick out the value which is second most important for you.
off and paste it in Box 2.

values.

Work slowly and carefully.
your answers.

place.

Peel it Off and paste it in Box 1 on the left.

Peel it
Then do the same for each of the remaining

The value which is least important goes in Box 18.

If you change your mind, feel free to change
The labels peel Off easily and can be moved from place to

The end result should truly show how you feel about these values.

66

 

 

 

 

 

 

 

 

 

 

b

 

and
"A

 

 

..de
CON

 

3:21:

 

 

 

 

“Animal's-1.3
oovcxw

 

 

 

l A COMFORTABLE LIFE

(a prosperous life)

" mar-mm “‘“33'
1 AN EXCITING LIFE

(a stimulating, active life)

—---——-— w” ‘— -——-—’

Eja—-—-
‘3

 

_Hr-r-——--

7...- ‘-

A SENSE OF ACCOMPLISHMENT
(lasting contribution)

 

- II-r-—-— ._: q—-——1 “.1 H—w—H—Y-—-.=..— ._-_
v r14 — -_n “L “F

l, _

I A WORLD AT PEACE

1' (free of war and conflict)

\\ —-— .31-r3”? W”: ‘ TW' ~‘v-—-+ ----;

j A WORLD OF BEAUTY
(beauty of nature and arts)

 

warn-—

 

-
a
LA! I'L E—HLfi-E -_-—...,

i EQUALITY (brotherhood,
I equal opportunity for all)

‘~. _ _ '- ’ '1 -‘», __..

3' FAMILY SECURITY
l (taking care of loved ones)

 

l FREEDOM

l (independence, free choice)
I'V-‘W “ - - TémmW—ltr' fi'“

l HAPPINESS ‘

l (contentedness)

~.-- WET“* "."'.".'“'“"
l INNER HARMONY

l, (freedom from inner conflict)
"- mqr==

MATURE LOVE

(sexual and spiritual intimacy)

 

( NATIONAL SECURITY

(protection from attack)
.--~--W-. _ _ 33 “TF3“,
I PLEASURE

I. (an enioyable, leisurely life)

\__ _ , il'fi-I I'll." WVI’H' ’
I . ' - -

l SALVATION
l (saved, eternal life)

\
'.
-mY—rfi____fl
a - Li I _ ‘ . . . ~
I

SELF-RESPECT

(self-esteem)
I.“ " a firm... WWW—“-

 

(respect, admiration)

 

TRUE FRIENDSHIP
(close companionship)

__ d? '(‘I'r- ‘ 'h m?

l WISDOM (a mature
' understanding of life)

.._.'.-.. . .- -. 1’!" a. .1" 1 "' "' "’1' __-_'_ 7*, ' "I '

l
l
{ sOCIAI. RECOGNITION

 

 

 

 

 

..-

no

67

Below you will find 12 statements, each one proposing some social action.
We are interested in finding out how much you feel each of these actions
would help or hinder the attainment of the value, A WORLD AT_PEACE.

 

Regardless of whether you favor or Oppose each of the actions, your judgments
Should be based on how strongly you feel they would help or hinder the
attainment of A World at Peace. Please rate each one according to the
following +5 to -5 scale.

If you feel that a particular action would have no effect (would neither
help nor hinder the attainment Of A World at Peace) rate it "0."

+1 would help slightly -1 would hinder slightly

+2 would help moderately -2 would hinder moderately

+3 would help quite a bit -3 would hinder quite a bit
+4 'would help strongly -4 'would hinder strongly

+5 would help very strongly -5 would hinder very strongly

l. The U.S. should promptly and completely withdraw its military
forces from Vietnam.

2. The United Nations Should be given stronger powers to help
settle disputes among member nations.

3. The U.S. should appropriate the necessary funds for a continued
strong and active program of space exploration.

4. Automobiles should be banned from the downtown areas of cities.

5. The Peace Corps should be expanded and strengthened.

6. Families Should limit themselves to two children.

7. Neon signs Should be abolished.

8. The U.S. should increase its economic aid to develOping countries.

9. The U.S. and Russia Should stOp experimenting with chemical and
biological warfare.

10. The U.S. should continue to build up its arsenal of nuclear
weapons, as a deterrent force against potential aggressors.

 

ll. Penalties for the possession of marijuana Should be reduced.

 

12. The U.S. should continue helping other nations such as Vietnam
in their struggles against Communist aggression.

68

Below are 12 statements, each one proposing some social action. We are
interested in finding out how much you feel each of these actions would
help or hinder the attainment of the value, A WORLD 9§_BEAUTY.

Regardless of whether you favor or Oppose each of the actions, your
judgments should be based on how strongly you feel they would help or
hinder the attainment of A World of Beauty. Please rate each one accord-
ing to the following +5 to -5 scale.

If you feel that a particular action would have no effect (would neither
help nor hinder the attainment of A World of Beauty) rate it "0."

+1 would help slightly -1 would hinder slightly

+2 would help moderately -2 would hinder moderately

+3 would help quite a bit -3 would hinder quite a bit
+4 would help strongly -4 would hinder strongly

+5 would help very strongly -5 would hinder very strongly

1. Automobiles should be banned from the downtown areas of cities.

2. More funds should be spent on highway beautification (land-
scaping, etc.).

3. Restrictions on hunting and fishing should be decreased.

4. The U.S. should increase its economic aid to developing countries.
5. Nonreturnable bottles should be prohibited.

6. Families should limit themselves to two children.

7. The U.S. Should continue helping other nations such as Vietnam
in their struggles against Communist aggression.

8. Neon signs should be abolished.

9. Factories which pollute the air or water should be forced to
clean up or Shut down.

10. Penalties for the possession of marijuana should be reduced.

 

11. People should be allowed to burn trash or leaves.

 

12. All forms of discrimination against women should be ended.
If you would like to make any comments, either general or specific, about
this questionnaire, please use the back of this page. Your comments will

be carefully read and heeded.

Thank you for your participation in this study.

APPENDIX C

Verbal instructions for second session

Hello. You're here for the second experimental session, aren't
you? May I have your name, please? (Check second session signup sheet,
to verify that subject has already participated in session 1). Today
I'm going to ask you to fill out another questionnaire. This question-
naire is very similar to the one you filled out last time. In fact, part
of it is exactly the same questions you were asked before. One of the
things we're interested in is to find out again how you feel about these
values and social actions. Your answers may be the same as before or they
may be different. Please don't worry about whether they are the same as
before or not. Just answer each question as carefully and honestly as
you can.

When you have finished the questionnaire, bring it back to me and
I'll Sign your card to give you credit for participating in the experiment.

Don't forget to read the instructions carefully before you begin.

69

APPENDIX D
VALUES AND ATTITUDES QUESTIONNAIRE - PART II

Instructions - Please read carefully

 

This is a continuation of the study in which you participated previously.
If you did not already complete Part I, you Should not fill out this portion.

In this portion of the questionnaire, we will be asking you some of the

same questions again which we asked you in Part I. Please answer these
questions carefully and honestly, according to how you feel about them today.
DO not worry about how you may have responded to them previously.

First, we would like to know your opinions about an issue of major importance.
Please read carefully the following selection. After you have finished it,
you will be asked to give your reactions to it.

 

This is an excerpt from a speech given recently at an environmental teach-
in by Dr. Milton Rokeach, a distinguished professor at Michigan State.

Dr. Rokeach is the author of the value ranking scale used in this question-
naire. In his speech, he discusses some findings from a previous study
using the value scale:

***

"I recently conducted a survey of the values of the American public.
We Obtained the value rankings of a national sample of American
citizens from all walks of life.

I was disheartened to find that most Americans rank A World 2; Beauty
as one of the least important values. Apparently most Americans have
not yet realized the vital importance of cleaning up and preserving
our natural environment.

We are suddenly confronting a new man-made peril--the poisoning of
our environment with noxious doses of chemicals, garbage, fumes, noise,
sewage, heat, ugliness and urban overcrowding.

We Americans are by far the worst polluters in the world. The average
American in his lifetime uses up 35 times as much of the world's
resources and produces 50 times as much garbage and waste pollution as
an average citizen of the rest of the world.

The challenge of the '70's will be to stOp polluting and destroying
our world. We must face the deadly chemical and physical changes
wrought by wanton pollution of earth, sky, and water.

If we lack the inclination to deal with these problems, the effects
will lie heavy on ourselves and our children--on their lives, their
health, their right to a wedge of happy space in the crowd, their
access to a tree, air to breathe, and water the right color.

70

71

In the past, man could always leave his own depredations behind and
move on to some part of the planet still unspoiled. Today there is
no escape. Our cities, where most of us live, are ugly and dirty.
Ten more years of doing nothing will make them much worse.

The earth is limited in its natural bounty. A World g£_Beauty must
become as important to us as A World §£_Peace. We must cherish this
earth and save it.

The results Of my study along with other evidence shows that many
Americans appear indifferent toward the threat to their future. This
is partly due to ignorance and partly because of apathy--"Let someone
else worry about it."

However, our study did reveal some sources of hope. Even though the
American public, as a whole, ranked A World 9E Beauty very low on
the list, we found that young people and the better-educated peOple
ranked it higher than did the Older and less well-educated.

Perhaps our nation's young people, particularly our college students,
will lead us in the campaign to restore and preserve our environment.
Recently we have witnessed rapidly increasing interest in this issue
among college students.

Young peOple should be more concerned because they have longer to live
in this world. But the task of restoring and maintaining a world of
beauty Should be of utmost concern, not just to the young, but to
every American."

drink

We are interested in finding out your Opinions about this issue. Please
rate each statement below from +5 to -5, according to how much you agree
or disagree with it.

+1
+2
+3
+4
+5

I agree slightly -1 I disagree slightly

I agree moderately -2 I disagree moderately

I agree quite a bit -3 I disagree quite a bit

I agree strongly -4 I disagree strongly

I agree very strongly -5 I disagree very strongly

l. The spoiling or destroying of our natural environment is a
serious problem in America today.

2. Especially in our large cities, the environment has already
reached the point of being detrimental to the physical and
psychological health of its inhabitants.

3. This problem will rapidly become worse unless we take immediate
and substantial steps to solve it.

4. Most Americans place too little importance on preserving the
beauty of nature.

72

5. This problem will become one Of the most important ones facing
our country during the next decade.

If you wish, you may use the space below for additional comments.

73

VALUE SURVEY - INSTRUCTIONS

On the next page are 18 values listed in alphabetical order. We are
interested in finding out the relative importance of these values for
you, as guiding principles in your life.

Each value is printed on a gummed label which can be easily peeled off
and pasted in the boxes on the left-hand side of the page.

Study the list carefully and pick out the one value which is most important
for you. Peel it off and paste it in Box 1 on the left.

Then pick out the value which is second most important for you. Peel it
off and paste it in Box 2. Then do the same for each of the remaining
values. The value which is least important goes in Box 18.

Work Slowly and carefully. If you change your mind, feel free to change
your answers. The labels peel off easily and can be moved from place to
place. The end result Should truly show how you feel about these values.

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

123456718

 

75

Below you will find 18 statements, each one proposing some social action.
We would like to know how you feel about each of these issues. We have
tried to cover different and Opposing points of view. You may find your-
self agreeing with some of the statements and disagreeing with others.
Whether you agree or disagree with any statement, you can be sure that
many other people feel the same as you do.

Mark each statement on the line to the left of the statement number,
according to how much you agree or disagree with it. Mark it with a

+ (plus Sign) if you agree or a - (minus sign) if you disagree, and a
number from 1 to 5 according to how strongly you feel in each case, from
1 (slightly) to 5 (very strongly). Thus, for example, if you agree
slightly with a statement, mark it +1; if you disagree very strongly,
mark it -5, etc.

Please read each statement carefully and be sure you understand its
meaning before responding to it.

+1 I agree slightly -1 I disagree slightly

+2 I agree moderately -2 I disagree moderately

+3 I agree quite a bit -3 I disagree quite a bit
+4 I agree strongly -4 I disagree strongly

+5 I agree very strongly -5 I disagree very strongly

1. The Peace Corps should be expanded and strengthened.

2. Nonreturnable bottles should be prohibited.

3. Families should limit themselves to two children.

4. The U.S. should increase its economic aid to developing countries.
5. People should be allowed to burn trash or leaves.

6. More funds should be Spent on highway beautification (landscaping,
etc).

7. The U.S. should continue to build up its arsenal of nuclear
weapons, as a deterrent force against potential aggressors.

8. Automobiles should be banned from the downtown areas of cities.

9. Penalties for the possession of marijuana should be reduced.

 

10. Neon signs Should be abolished.

 

11. All forms of discrimination against women Should be ended.

 

12. The U.S. should appropriate the necessary funds for a continued
strong and active program of Space exploration.

 

13. Restrictions on hunting and fishing Should be decreased.

+1

+3

+5

76

I agree slightly -1 I disagree slightly

I agree moderately -2 I disagree moderately

I agree quite a bit -3 I disagree quite a bit

I agree strongly -4 I disagree strongly

I agree very strongly -5 I disagree very strongly
14. The United Nations Should be given stronger powers to help

 

15

 

 

 

settle disputes among member nations.

. The U.S. Should continue helping other nations such as Vietnam
in their struggles against Communist aggression.

16. The U.S. and Russia should stop experimenting with chemical and
biological warfare.

17. Factories which pollute the air or water Should be forced to
clean up or Shut down.

18. The U.S. should promptly and completely withdraw its military

 

forces from Vietnam.

77

For scientific reasons, we need to know what impression you have formed
of the purpose of this study. In the space below, please explain as

thoroughly_and specifically as you can, the purpose 9£_this study, as
you understand it.

 

 

 

 

 

 

 

 

 

 

 

 

 

Your answers to the following three questions will enable us to "match"
this questionnaire with the one your filled out previously (Part 1)
without requiring you to reveal your name. It is very important that

you answer these questions again, even though you already answered them
once before.

 

Sex: Place of Residence: city state

 

Date Of Birth: month day year

 

If you would like to make any comments, either general or specific, about
this questionnaire, please feel free to do so, in the Space below or
on the back of this page. Your comments will be carefully read and heeded.

 

 

 

 

 

Thank you for your participation in this study.

"‘IIIIITTITTTES