108
214
.THS_

 

6'5

THLJ"

 

 

 

This is to certify that the

thesis entitled
THE NUCLEAR FREEZE MOVEPZENT AS COLLECTIVE ACTION:
A FIELD EXPERIMENT ON THE EFFECT OF REVEALINC
THE LOGIC OF FREE-RIDING ON CONTRIBUTIONS

presented by

Bradley Jay Fisher

has been accepted towards fulfillment
of the requirements for

Master of Arts degree in Sociology

 

 

Major professor

 

Date @1 r 5/63

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

MSU

LIBRARIES
‘13:,-

 

 

 

 

your record. FINES wiII
be charged-if book is
returned after the date
stamped beIow.

 

 

 

 

 

 

 

 

 

 

THE NUCLEAR FREEZE MOVEMENT AS COLLECTIVE ACTION:
A FIELD EXPERIMENT ON THE EFFECT OF REVEALING
THE LOGIC OF FREEnRIDING 0N CONTRIBUTIONS

By

Bradley Jay Fisher

A THESIS

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

MASTER OF ARTS

Department of Sociology

1983

ABSTRACT
THE NUCLEAR FREEZE MOVEMENT AS COLLECTIVE ACTION:

A FIELD EXPERIMENT ON THE EFFECT OF REVEALING
THE LOGIC OF FREE-RIDING ON CONTRIBUTIONS

By

Bradley Jay Fisher

Mancur Olson assumes that people are both rational and self-interested.
Hence, he assumes that in the absence of selective incentives, people would
prefer to be "free-riders" than to contribute toward large scale collec-
tive action. This study tests the hypothesis that one reason people often
contribute is that, contrary to Olson's assumption of rationality, they do
not understand the logic which supports free—riding. If, as Olson suggests,
almost all people already understand the logic of free-riding, then appeals
which mask or reveal this logic will generate the same rate of contribu—
tions. If, however, many people do not understand this logic, then appeals
which do not reveal the logic of free-riding should be more effective than
appeals which do reveal it. A sample of the signers and circulators of the
petition to get the Nuclear Freeze referendum on the 1982 Michigan ballot
were sent a letter asking them to purchase one or more bumper stickers
supporting the Freeze. The appeals varied as to whether they stated that
the Nucelar Freeze was a public good, i.e. nonexclusive, and whether
people were told they had high or low efficacy in bringing about this
public good. We found that manipulating the efficacy factor made no
statistically significant difference, but manipulating the nonexclusivity
factor did. Contrary to our prediction, subjects in the Nonexclusivity
Revealed condition were more likely to contribute than subjects in the
Nonexclusivity Not Revealed condition. One explanation for this result is
that by telling people they could receive the benefits without contributing

increased the credibility of the communicators and, hence, increased the

effectiveness of the appeals.

Perhaps the closest in: ever get to more profound answers

is learning to ask more profound questions!

Dr. Harry Robinson, 1975
Muskegon Community College

This is dedicated to my parents, Ethel and Louis,
and my brother, Gary, who have always provided me with
an encouraging word when I needed it most.

You are more to me than just a great family,

you are good friends as well.

ii

ACKNOWLEDGEMENTS

I wish to extend my sincere appreciation and affection to the chair-
persons of my committee, Stan Kaplowitz and Denton Morrison, for their
patience, advice, guidance, and good humor throughout the entire course
of this study.

I would also like to thank the other member of my committee,
Christopher Vanderpool, who has been an inspiration and a friend to me
since my arrival at Michigan State University.

A special note of thanks to Santo F. Camilleri whose insights helped
to clarify the original propositions of this study.

Over the last two years, several graduate students have provided me
with intellectual stimulation and emotional support: Marvin Finke, Jess
Gilbert, Akbar Mahdi, Marion McCoy, Nancy Merson, Bill Parsons, Gordon
Robinson and Kim Schopmeyer. Thank you for your friendship and concern.

Also, I would like to thank the Department of Sociology for granting
me a position as a teaching assistant over the past two years. This
position has helped me financially and provided me with a deeper apprecia-
tion for the art (and honor) of teaching.

Finally, I would like to express my deep gratitude for my companion,
Kathleen Eaton, who has been incredibly tolerant of my moodiness and has
shown me an abundance of tenderness and love. I hesitate to think what it
would have been like to deal with the stresses and adjustments of the last
two years without her. Thank you for being both friend and family to me

through times of sadness and times of joy.

iii

PERSONAL FORWARD

During the process of developing and conducting this research project,
I was confronted with several of my feelings about the nuclear arms race.
Since these feelings served as my basic motivation, as a sociologist and as
a person, for researching the Nuclear Freeze movement it seemed relevant
and responsible to share them here. In short, I consider the development,
production, and deployment of nuclear weapons as a dangerous practice and
as a sad commentary on the historical development (social evolution) of the
human species. The current policies advocated by the Reagan administration,
such as conceptualizing.a'winnable'nuclear war or a nuclear war confined
to the European theatre, bring us precariously closer to the brink of a
nuclear holocaust. Such a nuclear exchange can remain limited only in the
fantasies of military strategists. In actuality, such an exchange has a
high probability of ending civilization as we know it. It is my opinion
that despite the many flaws of our present stage of civilization, it is
still worth preserving as the basis for further change and development. I
feel a deep personal commitment to preserve the quality of the natural
environment and to enhance the quality of the social environment. The
prevalence of nuclear weaponry poses a direct threat to the continuance of
both of these environments.

I am greatly disturbed by how the current administration plays with
political rhetoric while explaining away nuclear arms escalation. The
'administrators' maintain that we must have more nuclear weapons in order

to be in a position to negotiate the reduction of such weapons. The

iv

logic (or illogic) of this argument, simply stated, is that we must have
more to have less. My feelings about this argument as well as about the
necessity for a Nuclear Freeze can best be expressed with an analogy. In
order to get a rolling piano to move in the opposite direction in which
it is going, one must first bring the piano to a complete halt. One does
not stop a moving piano by increasing its speed (unless one wants an
accident) nor can we achieve nuclear arms reduction by producing and

deploying more of such weapons.

II.

III.

TABLE OF CONTENTS

LIST OF TABLES ....... . ........ ..................... vii
LIST OF FIGURES . ................................... viii
LIST OF APPENDICES ........... ...... . ........ . ...... ix
THE PROBLEM ...... ......... ......... ..... . .......... 1
Olson's Theory ..... ..... . .............. .... ..... 1
An Alternative To Olson ................ . ...... .. 3
Hypotheses To Be Tested .............. ..... ...... 5
METHODOLOGY .................. ...... . ............ ... 10
Overview Of The Field Experiment ....... ......... 10
The Treatments ........ ......... ... .............. 11
Features Common To The Appeals ......... ..... .... 11
Operationalization Of Manipulated Variables ..... 12
Sampling Procedure .................... .......... 12

Assignment Of Subjects To Treatment Conditions .. 17

THE FINDINGS ....... ................................ 19
DISCUSSION .. ........ . ..... . ........................ 25
NOTES ......... .......................... . .......... 32
APPENDICES . ............. .... ............. . ......... 34
BIBLIOGRAPHY ...... ..... . .......... . ................ 39

vi

LIST OF TABLES

The Treatment Groups ....OOOOOOOOOOOOOOOO ...... O ........ O

The Effect of Perceived Efficacy and Non-
exclusivity on the Precentage of Contributors
by Condition: Petition Signers, First Mailing ...... ....

The Effect of Perceived Efficacy and Non-

exclusivity on the Percentage of Contributors

by Condition: Petition Circulators,

Second Mailing ..........................................

The Effect of Perceived Efficacy and Non-

exclusivity on the Percentage of Contributors

by Condition: Petition Signers and Circulators,

Combined First and Second Mailing ........... ............

Three-Way ANOVA for Percentage Contributing:
Nonexclusivity by Efficacy by Petition Signers and
CitCUlatorS O0............00.0.00.........OOOOOOOOIOOOOO.

Observed (and expected) Frequencies of Contributing
for Nonexclusivity Revealed and Not-Revealed Under

the Assumption of No Difference Between Conditions ......

vii

Page

20

20

20

21

LIST OF FIGURES

Figure Page
1. Features Common to the Appeals ........ ..... ............. 13

2. Operationalization of the Four
Treatment Conditions .................................... 14

viii

LIST OF APPENDICES

Appendix Page
A. The Fliers Mailed to the Subjects* ...................... 34

B. The 1982 Michigan Nuclear Freeze Petition Sheet* ........ 38

* Each has been reduced to 66% of its actual size.

ix

I. THE PROBLEM

Olson's Theory

 

One theorist, Mancur Olson, Jr., has made a substantial contribution
toward understanding the basis of collective action in his work The Logic

of Collective Action (Harvard University Press, 1965). His basic asser—

 

tion is that rational, self-interested individuals will not voluntarily
contribute resources toward creating a collective good they desire,
especially if a large number of persons share the desire for this collec-
tive good. A collective good is defined as any good, which, if made
available to one individual in a specified group, cannot be denied to
other members of that group. Consequently, the nonexclusivity of collec—
ltive goods provides an incentive for group members to become 'free-riders,‘
i.e. group members who do not contribute to the production of collective
goods since the goods will be produced by the efforts of others.

According to Olson, there are two main reasons why an individual
member in a large organization will not participate to help create a
collective good: 1) if the good is created, they can enjoy the benefits
of the collective good whether or not they have contributed toward its
production and 2) it is highly unlikely that any one person's contribution
will make the difference in whether the collective good is created. These
are the nonexclusivity and inefficacy features of collective action, re-
spectively. In other words, if an individual cannot be excluded from
consuming the goods produced by other members in the group or organization
and if an individual perceives his/her contribution toward purchasing the
good as inconsequential, i.e. has low efficacy, then, according to Olson,

this combination makes 'free-riding' both rational and likely.

The individual member of the typical large organization

is in a position (whereby)...his own efforts will not have

a noticeable effect on the situation of his organization,

and he can enjoy any improvements brought about by others

whether or not he has worked in support of his organization.

(Olson, p. 6)

It is Olson's conclusion that individuals in such a situation will have
no incentive to contribute resources to create a collective good.

Olson maintains that the.nonexclusivity and inefficacy features are
present in all attempts at collective action where large groups are con-
cerned and that some 'special device' is thus required to motivate group
members to act in their common interest. His basic proposition is that
collective action in a large group will occur if and only if there are
separate and selective incentives offered to those members of the group
who contribute.1 That is, there must be some side payment or private good
incentive which is separate from the public good and selectively available
only to those individualsyflm>contribute to the creation of the collective
good. These incentives may be either positive or negative. Olson's
argument suggests that members of a large group must either be offered
positive private incentives or coerced to act for the advancement of group
goals and objectives.

A central component of Olson's argument is his use of the economic
model to represent the rational individual who tries to maximize gains and
minimize costs. The rational, self-interested individual is one who cal-
culates the utility and cost of a given action and settles on that behavior
which will yield the highest gain for the least cost while ignoring any
direct consideration of the utilities of others. This economic model of
individual motivation leads Olson to conclude that private good incentives

are the necessary and sufficient conditions for individuals in a large

group to participate in collective action.

Individuals in a smaller group will have higher efficacy in that they
will perceive the acquisition of the collective good as being contingent
upon their contribution whereas members in the large organization will
have lower efficacy by perceiving that obtaining the collective good is
Eg£_contingent upon their contribution. Hence, in the small group, mem-
bers will have a higher incentive to bear the costs of producing the
collective good since their contribution is noticeable. In addition, there
is more face to face interaction in smaller groups which makes those who
attempt to free-ride more visible and, as a consequence, more vulnerable
to social sanctions. The larger group offers some degree of annonymity
to the individual member, decreasing the probability of identifying the
noncontributors. As a result, members in the large group experience less
incentive to contribute because their contribution, or lack of contribution,
will not make a noticeable difference and because the noncontributor is
less vulnerable to negative social sanctions. In short, the large group
allows the individual the option of getting something for nothing since
there is a high probability of being able to get away with it, i.e. of
avoiding the economic cost of contributing and the social cost of group

disapproval.2

An Alternative To Olson

While Olson sees selective incentives as crucial for the success of
collective action, others have offered different views. Fireman and Gamson
(1979) reject Olson's assumption that people are predominantly self—
interested. They maintain that Olson's utilitarian logic ignores the

importance of people sharing a common interest.

If individuals are thoroughly self-interested and rational,
common interests are unnecessary to collective action as well
as insufficient-~in fact they are irrelevant. (Fireman and
Gamson, p. 6)

 

Olson's argument suggests thataflJ.one need do to promote participation in
collective action is to provide some private good incentive. This implies
that it is a waste of time for the organizers of large scale social move-
ments to demonstrate to potential contributors the worthiness of the
collective good sought. Fireman and Gamson point out that such logic
suggests that individuals who are indifferent or opposed to the collective
good would be just as likely to participate as someone who supports it if
a selective incentive is offered that is worth more than the cost of parti—
cipation. They maintain that self-interest is not the sufficient condition
for collective action, but instead argue that such action is also contingent
upon developing and building on people's loyalty to a solidary group and
on their sense of responsibility to principles. Fireman and Gamson write:
It is useful to think of loyalty and responsibility not

merely as attributes of individuals but as properties of

cultural codes or belief systems. Individuals exist in a

climate with cultural beliefs about one's obligations to

those groups with which one identifies and the responsibility

to contribute one's share to just causes. (p. 32)
It is not denied that loyalty and responsibility can act as selective in~
centives, but as such they still involve a direct consideration of the
utilities of others, i.e. more than just one's self-interest.

While Fireman and Gamson question Olson's assumption that people are
basically self-interested, Kaplowitz (1982) questions whether people are
as rational as Olson assumes. Olson suggests that most people already
implicitly understand and have their actions guided by the logic of free-

riding. Consequently, individual members in a large group recognize that

collective action 1) pursues goods which are nonexclusive and 2) involve

a large number of potential contributors thereby making one individual's
contribution 'unnoticeable,‘ thus creating low levels of efficacy. If this
is true, then revealing the features of the free-rider problem to people
through appeals for contributions will not decrease the level of contribu-
tions and masking these features will not increase contributions. This
study hypothesizes that people do not implicitly recognize and act according
to either or both of the nonexclusivity and inefficacy features of large
scale collective action. Therefore, appeals for contributions which mask

or explicitly reveal these features will influence the level of response to

the appeals.

Hypotheses To Be Tested

 

As noted above, an individual's perceived level of efficacy is expected
to have an effect on his/her participation in collective action.r Campbell,
et al. (1964) have shown that voter turnout is positively correlated with
the strength of the individual's sense of political efficacy. More recent
research has substantiated the proposition that as an actor's level of
perceived efficacy increases, so does the probability that this actor will
participate in collective action (Camilleri and McMahon, 1975; Paige, 1971).
In other words, those individuals who feel their efforts will increase the
likelihood of attaining a given goal will be more likely to participate
than individuals who feel their efforts will make little or no difference
in the outcome.

Hypothesis: There will be a higher proportion of contributors among

those subjects receiving the High Efficacy appeal than

among those subjects receiving the Low Efficacy appeal.

It has been argued that masking and revealing the features of the
free-rider problem will have an effect on the level of responses to the
appeals. Clearly, it is difficult to imagine that anyone would perceive
the Nuclear Freeze as anything other than nonexclusive. For this reason,
it is expected that the nonexclusivity of the collective good will have
less influence on an individual's response than the indiviudal's perceived
level of efficacy. However, it is proposed that appeals which reveal or
do not reveal the nonexclusivity of the collective good may affect an
individual's perception of the nonexclusivity of the good and, thus, in—
fluence the level of response to the appeals.

Hypothesis 2: There will be a higher proportion of contributors

among those subjects receiving the Nonexclusivity-
Not Revealed appeal than among those subjects
receiving the Nonexclusivity-Revealed appeal.

We may distinguish between those who have already contributed to pro-
duce a collective good from those who favor it but have made little or no
contribution, e.g. sympathizers. The question to be addressed: Are
prior contributors (petition circulators) more likely to make an addi-
tional contribution toward collective action than are sympathizers (petition
signers) to make a first contribution?3 Clearly, many organizations think
so as they are especially likely to solicit donations from prior contribu—
tors before appealing to sympathizing noncontributors. In order to predict
the behavior of these two groups we must make some assumptions as to the
cause of the difference in past levels of contribution. Here are three

possible reasons for this difference:4

l. The collective good has greater utility for the prior contri-

butors.

2. The prior contributors have a greater sense of loyalty to the

group and responsibility to principles.5

3. Prior contributors are less aware of the logic of free-riding

than the sympathizers.

It is conceivable that the prior contributors place such a high value
on the collective good that this becomes their incentive for contributing
toward the production<xfthat good. Holding the level of efficacy con-
stant, Olson would agree that the greater the utility of the good to the
individual, the greater the probability that an individual will contribute
to acquire the good. Thus, if the collective good has greater utility for
the prior contributors, then we would expect a higher overall proportion
of contributors among the prior contributors than among the sympathizers.

Contrary to Olson's logic, Fireman and Gamson have presented the
argument that loyalty and responsibility act as incentives for people to
participate in collective action. Loyalty to the group assumes that
people are not just concerned withtheir own interests, but also consider
the interests of other members of the group. Hence, the loyalist may have
a greater incentive to contribute because he/she is pursuing a good which
is desired by the group and will benefit many others in addition to him/
herself. Responsibility to one's principles, however, is not necessarily
a group phenomena, i.e. it may involve the maintainence of one's self-
esteem. Responsibility refers to an individual acting on the basis of
his/her values or convictions. This sense of responsibility may serve as
an incentive to participate in collective action since the actors will

perceive themselves as doing their part, i.e.. upholding what they perceive

to be right. Consequently, if prior contributors have either a greater
sense of loyalty or have a greater sensecfifresponsibility, then in either
case we would expect a higher overall proportion of contributors among the
prior contributors than among the sympathizers.

Suppose, on the other hand, prior contributors are less aware of the
logic of free-riding, i.e. believe they have high efficacy and/or are not
fully aware of the nonexclusivity of the good. This lack of understanding
then may cause these individuals to participate in collective action even
though, according to Olson, it would not be rational to do so. But suppose
a prior contributor and a sympathizer each receive messages specifying the
same level of efficacy and providing the same information on the nonexclu-
sivity of the good. Literature suggests that a single discrepant message
is never totally effective, i.e. able to change a person's attitude to the
position advocated by the message, especially if the person has a strongly
held prior view about the issue addressed in the message (Sherif, etal.,
1965). Hence, if it is true that prior contributors are less rational6
prior to the experimental treatments, then they will also have somewhat
less rational responses after the treatments.

All of these explanations for the behavior of prior contributors lead
us to the same conclusion:

Hypothesis 3: There will be a higher proportion of contributors among

the prior contributors than among the sympathizers.

Sex was also included as a variable in this study to test the differing
orientation of men and women toward the Nuclear Freeze movement. This
variable was included for the sake of empirical interest rather than as an
additional test on Olson's theory. Consequently, we will not examine the
effect of sex in terms of the rationality argument raised earlier. There

is a limited amount of research available which suggests that women are

more strongly opposed to nuclear energy than men. The explanations for
this male/female difference are, at best, tentative. Mitchell proposes
that, "...women's child-centered biological and social roles influence
their perception of nuclear risks" (1981: 32). He argues that women are
more opposed to nuclear energy because of the risk it poses to the welfare
of future generations. If this is true, it seems reasonable to assume
that women will also be more opposed to the existence of nuclear weapons
and, hence, more supportive of the Nuclear Freeze movement. Other things
being equal, the higher utility women assign to the public good, i.e. the
Nuclear Freeze, will lead them to contribute more toward the creation of
the good.

Hypothesis 4: There will be a higher proportion of contributors among

female subjects than among the male subjects.

II. METHODOLOGY

Overview of the Field Experiment

 

The following field experiment uses Nuclear Freeze petition signers
and circulators as experimental subjects.1 The petition signers and circu-
lators were used for this experiment as it seemed logical to assume that
such individuals would (a) be more likely to value the collective good of a
Nuclear Freeze, (b) have some degree of interest in the Freeze issue since
they have already made a small investment by signing the Freeze petition,
(e) have some information and understanding of the Nuclear Freeze movement
and related issues, and (d) be more likely to respond to the appeal for
donations than would the general public. Each of the subjects was sent a
flier (see Appendix A) containing an appeal requesting a contribution to the
Nuclear Freeze Fundraisers.2

There are two independent variables which are manipulated in this
experiment: (1) Nonexclusivity Revealed/Nonexclusivity Not Revealed, and
(2) Low Efficacy/High Efficacy. The sex of the subject is an additional
independent variable in the first mailing. The dependent variable is the
level of response to the appeals, i.e. whether or not the subject contributes
in the form of purchasing a Nuclear Freeze bumper sticker and/or makes an
additional monetary contribution to the Nuclear Freeze referendum campaign.

The procedure of analysis will be, then, to compare the levels of
response to the various appeals and to assess whether there is a difference
in the response levels. Since random sampling is used, the test of
statistical significance (p < .05) is used as the criterion for the differ-
ences not being attributable to sampling error.

10

11

The Treatments

 

As mentioned above, the experimental treatments are as follows:
Nonexclusivity Revealed, Nonexclusivity Not Revealed, Low Efficacy and
High Efficacy.

Nonexclusivity refers simply to the fact that enjoyment of the public
good is not restricted, i.e. the good is available to all members of a
given group regardless of whether or not they contributed toward the crea-
tion of the public good. In this experiment, the public good to be created
is viewed as both the passage of the Nuclear Freeze referendum and the
implementation of a Nuclear Freeze.

Efficacy is defined as the perception, by the subject, of whether or
not his/her contribution will produce the intended effect, i.e. make a
difference in a given outcome.

This experiment has a 2 x'2 factorial design with the treatment groups
indicated as below:

Table l: The Treatment Groups

 

 

 

 

Low Efficacy High Efficacy
Nonexclusivity
Revealed I III
Nonexclusivity
Not Revealed II IV

Features Common to the Appeals

 

The fliers mailed to the subjects clearly indicate that the Nuclear
Freeze Fundraisers is a legitimate organization working for the passage of
the Nuclear Freeze referendum. Bumper stickers were printed and mailed to
those subjects who made contributions. The appeal for contributions was

genuine since all profits were to be used to support the Freeze movement.

12

It was decided that openly stating that the fliers were part of a research
project might seriously jeopardize the level of response to the appeals.
However, we did attempt to represent ourselves as honestly as possible.
Those features common to the appeals3 are presented inFigurelu The
only difference between the fliers occurred in the fourth paragraph (indi-
cated by the asterisk) where the independent variables were manipulated to

generate one of four experimental treatment conditions.

The Operationalization of Manipulated Variables
The four treatment conditions are shown in Figure2 as they appear, in
operational form, on the fliers as the fourth paragraph.
In treatements II and IV, Nonexclusivity Not Revealed is operational-
ized as the absence of the Nonexclusivity Revealed statement as it would
be an obvious falsehood to claim that a Nuclear Freeze is an exclusive good
which couldrun:be shared by everyone, i.e. contributors and noncontributors.
Prior contributors are operationalized as the petition circulators.

The sympathizers are operationalized as the petition signers.

The Sampling Procedure

 

The first phase of this research project involved the mailing of fliers
to a sample of 1000 subjects4 selected from the Nuclear Freeze petition
sheets turned in to the Michigan Elections Division (MED) in June, 1982.

When it was subsequently decided to initiate a second mailing5 (due to the
low response rates of the petition signers to the appeals), these same
petition sheets were used. The petition sheets (see Appendix B), which are
available as public property, contained over 350,000 signatures and addresses
of individuals wishing to see the Nuclear Freeze proposal on the November,

1982 Michigan ballot.

13

Figure 1: Features Common to the Appeals

September, 1982

Don: Cltizsn of Michigan:

is srs thrss concsrnsd citizsns trying vsrious wsys to raise funds for the
Nuclssr Frsszs.

we urge you to vote for tho Nuclear Frsszs Referendum in the general
slsction of Tuesday, November 2, 1982. Vots yes on Proposition 3.

Us hops you will msks sn odditionnl contribution to asking ths Nuclear
Frsszs Referendum psss ovorwhslmingly by purchasing and displaying s
Nuclear Frsszs bumper sticker. All profits from sslss go to promote
passage of the Michigan Nuclssr Frsszs htsrsndum.

Anything that can bs dons to hslp bring about s Nuclear Frsszs will hslp
rsdncs tho risk of nuclear war. Isn't thst rsason enough for your
contribution?

Thank you,
Brsdlsy Fishsr
Stan Kaplowitz
Dsnton Morrison

Bsst Lsnsing, Michigan

 

 

Print your Cut this and taps
nsms ad address NUCLEAR FREEZE: GIVE “BONDS AHA!‘ NEW MEANING to your snvslops
hero for our V017. YES ON PROPOSITION F. NOV. 2 as your nsiling
smiling lsbsl lsbsl

 

 

 

dlz

 

Groom on white

 

 

 

 

 

'J}

n:

:3
I sticker 31.00 Mike checks out to: 1’3 f2;
3 stickers 52.00 g g
5 sticksrs $3.00 Nuclear Freeze Fundraisers g "
IO sticksrs $430 a __
15 stickers $5.00 a 2
(Writs for bulk rats) a :5
e as
If you enclose a stamped. 3' ... '3
self-addressed envelope fi 3‘, f;
thsrs will be more money '13 __1 "" g...
for the Freeze. g if; 23 2
z c a: :-.x

 

 

 

 

 

14

Figure 2: ggperationalization of the Four Treatment Conditions

TREATMENT I

Nonexclusivity Revealed/
Low Efficacy

 

If the Nuclear Freeze campaign
succeeds you will,of course, real-
ize the benefits whether or not you
have contributed to the Freeze. /
In a state as large as Michigan, it
is very unlikely that one contribu—
tion will make the difference in
whether or not the Referendum
passes. And even if the Referendum
passes it may make little differ-
ence in whether or not a Nuclear
Freeze comes about.

TREATMENT III

 

Nonexclusivity Revealed/
High Efficacy

If the Nuclear Freeze campaign
succeeds you will, of course, real-
ize the benefits whether or not you
have contributed to the Freeze. /
Each contribution to the Freeze in-
creases the chances that the Nuclear
Freeze campaign will succeed. Your
contribution will make a difference.

TREATMENT II

Nonexclusivity Not Revealed/
Low Efficacy

In a state as large as Michi—
gan, it is very unlikely that one
contribution will make the differ-
ence in whether or not the Referen-
dum passes. And even if the Ref-
erendum passes it may make little
difference in whether or not a
Nuclear Freeze comes about.

TREATMENT IV

Nonexclusivity Not Revealed/
High Efficacy

 

Each contribution to the Freeze
increases the chances that the
Nuclear Freeze campaign will succeed.
Your contribution will make a dif-
ference.

15

It is the responsibility of the MED to receive circulated petitions
and to inspect them to ascertain whether enough valid signatures have been
gathered to certify a petition for placement on an election ballot. From
the 39,800 petition sheets received (15 names possible per sheet) by the
MED, 498 were randomly selected by a computer and subsequently copied for
the MED staff to inspect for this certification process. For convenience,
it was decided to use these same 498 sheets for the initial stage of selec-
ting experimental subjects for the first mailing. Since sex was to be
examined as a variable in the first mailing, the subjects were stratified
according to sex so as to obtain equal numbers of men and women in each
treatment group.

For the purpose of this study, the MED allowed us to use the photo-
copies of the original 498 sheets which posed several problems listed below.

1. Poor photocopy, i.e. making a petition sheet impossible to read.

2. Illegible signature.

3. Illegible or questionable address (such as Loeb Farm).

4. Duplicated names (two sheets were identical).

5. unisex names such as Leslie, Kelly, Dana, Robin, etc. which had
to be excluded in order to stratifythe subjectsaccording to sex.

6. The petition signer used first initials only (prohibiting identi—
fication of sex).

Out of the 500 names selected by the MED from the 498 sheets, 91 sig-
natures were rejected for one or more of the reasons listed above. The
first stage of selecting subjects provided 409 acceptable names, 164 men
and 245 women.

The second stage of the selection process involved going through the
sheets again and pulling one additional name from each sheet. However,

there were a few additional criteria for acceptability. The name would

16

naturally be rejected if it had already been selected. A name was also
rejected it if appeared to be the spouse of someone already selected or if
the individual lived in the same residence as a previously selected subject.
When possible, we also attempted to avoid selecting people who might be
close neighbors (this occurred only once).

The fifth name of each sheet was used if it was acceptable. If it was
not, the procedure was to proceed to the fourth name and then the third
until an acceptable name was found. If none of the first five names were
acceptable, then we proceeded to the sixth, seventh, etc. until an accept-
able name was found. Those sheets which had only one signature or were
poorly photostated were skipped. In the case of sheets containing less
than five signatures, the last name on the sheet was used if apprOpriate.
This process provided an additional 479 subjects, 224 men and 255 women.
During the second stage of selection, enough female subjects had been ac-
quired and, from then on, only acceptable male signatures were sought.

The third stage involved the selection of the remaining 112 male
subjects for the experiment. Again, the process began at the beginning
of the stack of 498 sheets and each sheet was examined for appropriate male
signatures. The remaining 112 names were selected from the first 120
petition sheets bringing'the total number of subjects to 1000.

One hundred alternate subjects (50 male and 50 female) were selected
as a precaution against incorrect addresses, e.g. "return to sender,

address unknown" or "no such number or street,"

or other problems with the
mailing of the fliers. After reviewing the list of selected subjects, some
additional questions arose as to the acceptability of some of the names

according to the criteria previously employed, e.g. the inclusion of unisex

names such as Terry and Marion andsunmaquestionable addresses. Consequently,

17

28 men and 38 women were eliminated as subjects and 66 more subjects were
selected as replacements.

The second mailing involved the selection of 600 petition circulators
as experimental subjects. The selection of subjects for this mailing also
used the original 498 sheets as the initial stage in the selection process.
The second mailing did not include a consideration of sex as an additional
independent variable and, hence, unisex names were acceptable. Since
petition circulators were also potential petition signers, a petition circu-
lator's name was rejected if it had been selected for the first mailing.

From the 498 sheets, 402 names were found to be acceptable. The re-
maining 198 subjects and 100 alternates were then selected from the remain-
ing 39,300 petition sheets on file at the MED. Every 130th sheet was pulled
and the name of the circulator on that sheet was used. If the name was not
acceptable, then the sheet prior to the selected sheet was used, i.e. the
129th, the 128th, etc. This same procedure was followed if the circulator's
name had previously been selected or if the sheet had already been pulled,

i.e. if it was one of the original 498 sheets.

Assignment of the Subjects to Treatment Conditions

Below is a brief outline of the process whereby subjects were assigned
to the various experimental treatment groups.

1. Typed mailing labels were put on envelopes.

2. The envelopes were alphabetized.

3. A master list of subjects was typed.
For the first mailing, the envelopes were also separated into alphabetized
stacks of men and women so as to be able to assess the effect of a subject's
sex upon the rate of response. The assignment of subjects to treatments

was as follows:

18

First man and woman of alphabetized stack were assigned to treatment I.
Second man and woman of alphabetized stack were assigned to treatmentII.
Third man and woman of alphabetized stack were assigned to treatment 111.
Fourth man and woman of alphabetized stack were assigned to treatmentIVJ
Fifth man and woman of alphabetized stack were assigned to treatment 1.
Sixth man and woman of alphabetized stack were assigned to treatment II.
etc.
The treatment group to which each subject was assigned was then indicated on
the master list. The form a subject would mail back to order a bumper stick-
er also indicated the treatment condition to which he/she had been assigned,
i.e. a subject receiving appeal "1" would mail his/her order form back to
Department 1.

The assignment of subjects to experimental treatment groups for the
second mailing followed the same initial three-steps indicated above, how-
ever, they were not separated into stacks of male and female subjects. The
procedure for assignment was as follows:

First subject of alphabetized stack was assigned to treatment I.

Second subject of alphabetized stack was assigned to treatment II.

Third subject of alphabetized stack was assigned to treatment III.

Fourth subject of alphabetized stack was assigned to treatment IV.

Fifth subject of alphabetized stack was assigned to treatment 1.

etc.

Again, the assignment of subjects to the various treatment groups was
indicated on the master list.

To summarize, for both mailings the subjects were assigned to the various
experimental treatment conditions through a systematic random process. Con—
sequently, it is assumed that this process made the groups experimentally

"equivalent" (Stanley and Campbell, 1963).

III. FINDINGS

If, as Olson suggests,people are already aware that they can be free—
riders, then we would expect to observe no statistically significant dif-
ference in the proportion of contributors among those subjects in the
nonexclusivity or efficacy treatment conditions. However, we have suggested
two alternative hypotheses.

Hypothesis 1: There will be a higher proportion of contributors among
those subjects receiving the High Efficacy appeal than
among those subjects receiving the Low Efficacy appeal.

The combined data in Table 4 shows that the direction of the results
is opposite from the prediction. The overall magnitude of response to the
appeals was small so the observed differences between conditions were severely
constrained by this. However, a greater proportion of subjects contributed
in the Low Efficacy condition (2.13%) than in the High Efficacy condition
(1.75%). Moreover, the results are in this same direction for both petition
signers and petition circulators (Table 2, Table 3). The three-way ANOVA
shows that the effect of the efficacy factor is not statistically significant
(p > .25).

Hypothesis 2: There will be a higher proportion of contributors among

those subjects receiving the Nonexclusivity Not Revealed
appeal than among those subjects receiving the Nonexclu-

sivity Revealed appeal.

19

Table 2:

20

The Effect of Perceived Efficacy and Nonexclusivity

on the Percentage of Contributors by Condition:

Petition Signers, lst Mailing

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Efficacy
High Low
Revealed 2.40 2.40 2.40
Non
Exclu31vity Not Revealed .40 1.60 1.00 N= Sample size of
250 per treatment
1.40 2.00 1.70 condition.
Table 3: The Effect of Perceived Efficacy and Nonexclusivity
on the Percentage of Contributors by Condition:
Petition Circulators, 2nd Mailing
Efficacy
High Low
Revealed 4.00 2.70 3.35
Non
Exclusiv1ty Not Revealed .667 2.00 1.33 N- Sample size of
150 per treatment
2.33 2.35 2.34 condition.
Table 4: The Effect of Perceived Efficacy and Nonexclusivity
on the Percentage of Contributors by Condition:
Petition Signers and Circulators,
Combined lst and 2nd Mailing
Efficacy
High Low
- Revealed 3.00 2.50 2.75
Non
Exclusivity Not Revealed .50 1.75 1.13 N= Sample size of
400 per treatment
1.75 2.13 1.94 condition.

 

21

Table 5: Three-Way ANOVA for Percentage Contributing:
Nonexclusivity by Efficacy by Petition Signers and Circulators

 

 

 

 

 

*P< .025

Sum of Mean F

Source Squares D.F. Square Ratio
Nonexclu. (A) .1056 1 .1056 5.56*
Efficacy (B) .0056 1 .0056 .29
Petition Sign.
& Circul. (C) .0150 1 .0150 .79
A x B .0306 1 .0360 1.61
A x C .0034 1 .0034 .18
B x C .0033 1 .0033 .17
A x B x C .0051 1 .0051 .27
ERROR 30.2307 1592 .0190
TOTAL 30.3993 1599

 

 

 

22

The combined data from Table 4 indicates that the direction of the
results is opposite to this prediction. Again, within the given con-
straints of the magnitude of responses involved, a greater proportion of
subjects contributed in the Nonexclusivity Revealed condition (2.70%)
than in the Nonexclusivity Not Revealed condition (1.13%). Examining the
three-way ANOVA, we find this effect to be significant (F=5.56, df=l,

1591 p < .025).

Aside from the hypotheses dealing with our two experimentally manipu-
lated variables, we also generated two other hypotheses.

Hypothesis 3: There will be a higher proportion of contributors

among the prior contributors, i.e. petition circulators,
than amng the sympathizers, i.e. the petition signers.

Looking at Tables 2 and 3, we see that a higher prOportion of petition
circulators contributed (2.34%) than did petition signers (1.70%) which
is in the direction of the prediction. This result holds for all four
treatment conditions. The three-way ANOVA shows, however, that this
result is not statistically significant (p > .25),

The data in Table 4 suggests that there might be an interaction effect
between the nonexclusivity and efficacy factors. That is, the subjects in
the Nonexclusivity Revealed condition had a higher proportion of contri-
butors if they were in the High Efficacy condition, while in contrast,
those subjects in the Nonexclusivity Not Revealed condition had a higher
proportion of contributors if they were in the Low Efficacy condition.
However, the three-way ANOVA shows that none of the interaction effects
are statistically significant.

Hypothesis 4: There will be a higher proportion of contributors

among the female subjects than among the male subjects.

23

The factor of sex was measured and balanced among conditions only in
the first mailing, i.e. for petition signers. Out of 17 responses, eleven
(11) were male and six (6) were female which is in the opposite direction
of the prediction. A two-tailed t-test was conducted on the data which
showed that the results were not statistically significant1 (pi>.2).

It is important to acknowledge that the F-test is not completely valid
because it makes two assumptions:

1. The variances among conditions are homogenous.

2. The populations are normally distributed.

The variances for each of the treatment conditions are as follows:

SE = .023475, Si = .017194, 8% 8 .02910, 8: = .004975.

Using the Fmax test (see Winer, 1971), we found that the ratio of the
largest to the smallest variance to be 5.85. With four variances and 400
observations per variance, this ratio is significant at p‘<.01. Hence, we
must reject the null hypothesis of equal condition variances. In addition,
using dichotomous data with such a low response rate (31 out of 1600) makes
it highly probable that the distribution is skewed.

However, a Chi-square test was conducted on the combined data for the
nonexclusivity factor since it is a more conservative test in that it does
not make the assumptions associated with the F—test. The data in
Tablefishows that we obtained a Chi-square of 5.56, df=1, which indicates

that the nonexclusivity factor is statistically significant at p <.025.

Table 6: Observed (and expected) Frequencies of Contributing

for Nonexclusivity Revealed and Not Revealed under
the Assumption of No Difference Between Conditions

Revealed
Non

Exclusivity Not Revealed

Total

 

 

 

 

 

= 5.559, df= 1, p <.025

 

Non
Contributors Contributors Total
22 (15.5) 778 (784.5) 800
9 (15.5) 791 (784.5) 800
31 1569 1600

DISCUSSION

Olson's conceptualization of the rational individual assumes that
such a persbn is aware of the nonexclusivity and inefficacy features of
free—riding. If Olson is correct, mentioning these features should have
no effect on whether or not an individual participates in the creation
of a collective good.

An alternative possibility is that people may not be as rational as
Olson thinks (or rational in the same way) in that they may be influenced
by revealing or not revealing the nonexclusivity and inefficacy features.
It has been hypothesized that raising perceived levels of efficacy will
make a difference as well as not revealing the nonexclusivity of the
collective good.

The results from the field experiment show that manipulating the
efficacy factor was statistically insignificant, but manipulating the non-
exclusivity factor did make a statistically significant difference. Con-
trary to both Olson and the alternative, subjects were more likely to
contribute to Nuclear Freeze Fundraisers if they were in the Nonexclusivity
Revealed condition. We are now left with the task of speculating as to why
we obtained these results. Two explanations come to mind:

1. Revealing the nonexclusivity feature may have produced feelings

of inequity in the recipients of the appeals.

2. Revealing the nonexclusivity feature may have increased the

credibility of the source of the appeals.

It can be said that an equitable situation exists when the perceived

inputs and outcomes of an individual are equal to his/her perception of

25

26

the inputs and outcomes of others. For our purposes here, inputs refer
to contributions, outcomes refer to rewards, i.e. the availability of
the collective good, and others pertains to other contributors. J. Stacy
Adams (1965) has argued that when a person perceives a situation as in-
equitable, this person will experience tension and be driven to reduce
the inequity.1 Burnstein and Wolosin (1968) write:

Inequity has been assumed to have noxious effects. Those

who benefit, i.e. those whose rewards increase or whose costs

decrease relative to others, are said to experience embar-

rassment or guilt; those injured, i.e. those whose rewards

decrease or whose costs increase relative to others, are said

to experience humiliation or anger. As a result, the indi-

vidual desires to maintain or restore equity and will act to

satisfy this desire. (p. 416)

This suggests that a person will perceive a situation as inequitable
if he/she is over-rewarded or under-rewarded. By reminding people that
they can receive benefits without making sufficient inputs, i.e. get
something for nothing, may produce perceptions of being over-rewarded.
Some donation appeals attempt to dissuade people from taking advantage
of the nonexclusivity of the good by reminding them that such behavior
will over-reward them, e.g. "Don't just sit there and let someone do your
contributing for you!".2 Hence, one explanation is that the subjects in
the Nonexclusivity Revealed condition increased their inputs, i.e. contri-
buted in order to reduce feelings of inequity.

One might question the above explanation because the petition circu-
lators, who have already provided an input, i.e. the cost of circulating
the petitions, were more likely to contribute (and in larger sums) than
the petition signers. If the inequity argument‘ holds that all one need

do, in this situation, is to increase inputs, then the circulators have

already done so and should, theoretically, contribute less.3 Since the

27

circulators did contribute more than the petition signers a more plausible
explanation is that there are other factors involved besides feelings of
inequity.

Research has shown that when a communicator takes a position that the
recipient (of the communication) perceives is opposite to the communicator's
best interests, the communicator‘ is viewed as more sincere, i.e. more
credible, and hence the message is more persuasive (Walster, et al., 1966;
Eagly, et al., 1978). If a fundraiser informs people that their contri—
butions will make little difference and/or that they can receive the collec-
tive good even if they do not contribute, such a communication would be
perceived by the recipients as making statements opposed to the communica-
tor's best interests. Such negative statements have the effect of increasing
the credibility of the communicator and, hence, increase the effectiveness
of the appeal.

This would explain why the Nonexclusivity Revealed condition generated
a higher proportion of contributors than the Nonexclusivity Not Revealed
condition. However, if this explanation is valid, why did we not observe
a significantly higher proportion of contributors in the Low Efficacy
condition than in the High Efficacy condition? We may surmise that telling
people that their contributions will not make a difference has a dual
effect. Such a statement may increase the credibility of the source of
the communication, but it may also decrease the recipients' desire to con-
tribute their resources which may be better spent elsewhere. On the other
hand, providing information which openly recognizes the nonexclusivity of
a collective good, such as a Nuclear Freeze or public television, may not
have a negative effect since people already recognize the nonexclusivity

feature.

28

While the three-way ANOVA indicated that the interaction effect
between the nonexclusivity and efficacy factors are statistically insig-
nificant, the results are highly suggestive. Examining the combined data
in Table 4, we see that there was a higher proportion of contributors in
the Low Efficacy/Nonexclusivity Not Revealed condition (1.75%) than in
the High Efficacy/Nonexclusivity Not Revealed condition (.502). This
suggests that when there are no prior negative statements,the harm from
telling people they have low efficacy is outweighed by the positive credi-
bility effect of that one negative statement.

Is it the case that if a little honesty is good, then a lot should be
better?' Comparing the two treatment conditions in which Nonexclusivity
is Revealed (Table 4), we find that there was a higher proportion of con-
tributors in the High Efficacy condition (3.0%) than in the Low Efficacy
condition (2.50%). In short, while making one negative statement, i.e.
Nonexclusivity Revealed, may increase the credibility of the communicator,
making further negative statements offers little, if any, further gain to
source credibility. Whatever gain is achieved, is more than counteracted
by the loss from telling a person they have low efficacy.

It was also hypothesized that the petition circulators would be more
likely to contribute than the petition signers. While the results were in
the direction of the prediction, the difference was not statistically
significant. However, for all four treatment conditions, the circulators
did have a higher proportion of contributors than did the petition signers.
This strongly suggests that there is a difference between those individuals
at the core of a social movement and those who are sympathizers.

The factor of sex was also shown to be statistically insignificant

but the logic of the argument remains appealing. The basic problem which

29

confronted the analysis of this variable was that we had no way of being
certain that our listed recipient, e.g. Ms. Doe, responded to the appeal
rather than her spouse. In two cases it was clear that it was the spouse
and not the designated recipient who responded. We cannot be certain that
the person who made the decision about contributing was the original
recipient of the appeal. Consequently, we cannot test the differing orien-
tations of men and women toward the Nuclear Freeze movement.

The response rate to the appeals was slightly under 2% which was quite
disheartening. Clearly, we expected that individuals who showed some
sympathy toward the Freeze movement would respond more strongly than they
did to our appeals. It is obvious that we were wrong and several possible
explanations have been considered. Before explaining these possibilities,
it needs to be acknowledged that the overall response rate does offer some
support for Olson's theory in that most of the people who gave utility to
the collective good elected not to contribute. However, as indicated, the
low response rate may be attributable to several other factors.

The first factor to consider is that both mailings were sent out in
close proximity to election day, November 2nd. During this political time
of year, people are swamped with a variety of pamphlets and pleas for
contributions and, thus, our appeal may have been one among many to get
lost in a stack or to hit the trashcan. Secondly, the bumper sticker we
advertised solicited a yes-vote for Proposition "E" on November 2nd. Con—
sequently, as that deadline approached, people may not have wanted to
purchase and display a bumper sticker which would have been outdated in a
matter of weeks. Thirdly, the Nuclear Freeze Fundraisers was not a known
organization so people may have been reluctant to contribute their resources
without a more convincing guarantee (besides our word) that the funds would,

in fact, go to support the Freeze movement. Finally, the Michigan Freeze

30

proposal was viewed as having a high probability of passage. Our appeals
sought to raise funds to promote a collective good rather than to directly
produce the good, so people may have felt they could have a more direct
effect by simply voting for the Freeze proposal. Since the proposal was
likely to pass, the response to our appeals may have been low because
people could show their support through a vote and because there appeared
to be no urgent need to contribute funds to fight a battle that had already
been won, i.e. everybody could be a free-rider.

There are also some possible modifications to this field experiment
which may serve to improve the response rate and the general quality of the
study. One possible alternative is to mail the bumper sticker along with
the appeal which is similar to the American Lung Association's strategy
when they mail "personalized address labels" with their donation appeals.
This might reassure the recipients that the organization is not a hoax
and would eliminate all possibility that the bumper sticker acts as a
private good incentive since everyone, including noncontributors, would
receive one. Inlcuding the bumper sticker would hOpefully increase the
organization's credibility and facilitate higher response rates.

In addition, follow-up interviews with recipients in each treatment
condition (respondents and nonrespondents) might provide useful information
as to the actual effect of the communications and shed some light on why
subjects choose not to contributei

Finally, it would seem prudent that similar experiments be conducted
well before the election day to avoid competing with other political
campaigns. The November deadline became an imposition on rather than an
impetus for the facilitation of higher response rates. Consequently, such

a mailing might be more successful if it were not associated with a specific

31

deadline (as well as the bumper stickers) allowing for sufficient time
for the recipients to respond.

In conclusion, the field experiment and the one statistically sig-
nificant result, i.e. the effect of nonexclusivity, suggests that there
may be a flaw in Olson's logic of collective action. People may not be
as rational as Olson thinks in that revealing and not revealing the non-
exclusivity feature of free-riding did make a difference in whether sub-
jects contributed toward collective action. We offered the conjecture
that subjects in the Nonexclusivity Revealed condition had a higher pro-
portion of contributors than those subjects in the Nonexclusivity Not
Revealed condition because they perceived the source of the communication-
as more credible and, hence, were more persuaded by the appeals for con—
tributions. Given the low response rate, the generalizations that can
be made from the results are limited,which indicates the necessity for
further replications of this study. While the results do not conclusively
refute the hypotheses of this thesis (nor do they support or refute
Olson's economic analysis of human behavior), there is evidence to
suggest that revealing the features of the free-rider problem does affect
the likelihood that an individual will contribute resources toward

collective action.

NOTES

Theoretical Background

1. Olson argues that if the collective good is produced it will be
at a suboptimal level or below the amount of the good that would be
in the best interests of the group to provide. This suboptimizing of
goods occurs because the rational, self-interested individual calcu-
lates how much of the good he/she needs and once his/her costs exceed
his/her gains, then the individual will stop contributing regardless
of the needs of other members of the group. As Olson puts it, when
one's share of the costs exceeds one's share of the additional bene-
fits, one will stop contributing toward the production of the collec-
tive good.

2. This interpretation of Olson draws heavily upon the writing of
Denton E. Morrison (1975 and 1977). See bibliography for a complete
citation of these sources.

3. Similarly, we could ask if sympathizers will have a higher pro-
portion of contributors than individuals selected from the general
public. The only problem I see with such a proposition is that sign-
ing a petition is such a minimal cost, as compared with circulating

a petition, that there may be no significant difference between the
responses of sympathizers and the general public. However, during my
experience soliciting signatures for the Freeze proposal, 252 or less
of those solicited actually signed the petition sheets which suggests
that the sympathizers, i.e. signers, are a special subset within the
general public.

4. Some other possible explanations for the differences between prior
contributors and sympathizers are:

-Prior contributors are more involved in organizational networks which
place greater pressure on them to contribute.

«Prior contributors have a higher level of perceived efficacy as justi-
fied by previous experience (circulating petitions did get the Freeze
proposal on the November ballot).

-Prior contributors have greater amounts of information than the
sympathizers, e.g. pamphlets, newsletters, etc.

5. Explanations 1 and 2 assume that the prior contributors are as
rational as the sympathizers in that they are both equally aware of
the potential for free-fiding.

6. From my own experience as a petition circulator as well as con-
versations I have had with other circulators, I found no evidence to
support this proposition. Other circulators seemed fully aware of
the potential for free-riding, yet they contributed their time and
resources toward the Freeze movement because they felt it advanced
their self-interest and the interests of others. They also contri-
buted because they felt a responsibility to act according to their
principles, i.e. they felt they had to get involved and do what they
felt was right.

32

II.

33

Methodology

1. The proposal for this study was presented to and approved by the
University Committee on Research Involving Human Subjects in August/
September of 1982.

2. The Nuclear Freeze Fundraisers was an organization fully regis-
tered and in compliance with guidelines of the Michigan Elections
Division.

3. The subjects mailed their contributions to a common address:
Nuclear Freeze Fundraisers, Box 1521, East Lansing, Michigan. However,
the department varied as to the treatment to which a subject had been
assigned. Petition signers mailed their contributions to Department

A, B, C, or D. Petition circulators mailed their contributions to
Department 1, 2, 3, or 4.

4. The first mailing was sent out on September 27th.

5. By the 11th of October, we had received only 13 responses from
the first mailing and it was subsequently decided to initiate a second
mailing. This second mailing was sent out on October 14th.

6. The name of a petition circulator appears in the lower right hand
corner of the petition sheet and must include the circulator's name
and address (see Appendix B).

III. Findings

IV.

1. For the two—tailed t-test, t= 1.2195, df= 998.

Discussion

 

1. Adams has argued that an individual (Person) is driven to reduce
feelings of inequity which may be accomplished in at least two ways.
First, the inequity may be reduced if either Person or Other increases
(or decreases) their inputs and/or outcomes. Second, Person may alter
perceptions of Person's own inputs and outcomes and/or the inputs and
outcomes of Other. Hence, feelings of inequity may be reduced if
Person actually or perceptually manipulates the inputs and/or outcomes
of Person or Other.

2. WKAR, Channel 23, Festival '83 membership drive campaign.

3. One might counter the argument that petition circulators will con-
tribute less by claiming that the collective good has a higher utility
for the petition circulators than for the signers. Consequently, the
circulators may perceive their inputs as insufficient when compared
with the outcome to be gained.

4. While we cannot be sure about the sex of the individual who
decides whether or not to contribute, we are sure that each house-
hold was in only one treatment condition. Hence, we do not encounter
the same problem with interpreting the response rates by condition.

APPENDICES

APPENDIX A:

The Fliers Mailed to the Subjects

APPENDIX A

Nonexclusivity Not Revealed/Low Efficacy Appeal

 

September . 1982

Dear Citizen of Michigan:

We are three concerned citizens trying various ways to raise funds for the
Nuclear Freeze.

We urge you to vote for the Nuclear Freeze Referendum in the general
election of Tuesday, November 2, 1982. Vote yes on Proposition B.

We hope you will make an additional contribution to making the Nuclear
Freeze Referendum pass overwhelmingly by purchasing and displaying a
Nuclear Freeze bumper sticker. All profits from sales go to promote
passage of the Michigan Nuclear Freeze Referendum.

In a state as large as Michigan, it is very unlikely that one contribution
will make the difference in whether or not the Referendum passes. And even
if the Referendum passes it may make little difference in whether or not a
Nuclear Freeze comes about.

But anything that can be done to help bring about a Nuclear Freeze will
help reduce the risk of nuclear war. Isn't that reason enough for your
contribution?

Thank you,
Bradley Fisher
Stan Kaplowitz
Denton Morrison

East Lansing , Michigan

 

 

 

 

dlz

 

 

Print your ‘ Cut this and tape
name and address NUCLEAR FREEZE: GIVE "BOMBS AHA!“ NEW MEANING to your envelope
here for our VOTE YES ON PROPOSITION 8 NOV. 2 as your mailing
mailing label , label
Green on white :3
a:
1 sticker $1.00 Make checks out to: 3'. :3
3 stickers 32.00 """'"' "" "" f. 3
5 stickers $3.00 Nuclear Freeze Fundraisers g Q
10 stickers 34.00 E .-
15 stickers $5.00 a: 2
(Write for bulk rate) ‘3’. c.
:2 5
If you enclose a stamped, "‘ ~ -- g
self-addressed envelope 5 3 .5.
there will be more money 3 (J I :—
for the Freeze. '5' a g 33
Z 6 GI:

 

 

 

 

 

 

 

 

35

APPENDIX B

The 1982 Michigan Nuclear Freeze Petition Sheet

gwszHx w

 

.56 $8 Zena. m: zcowmmn mnomum mezzo: msomn

 

 

is I.
‘2- 3:??? 3 cs 3.!— es s. ’e so...

i... .. Llplllacli 1 r .Elna—sol’egmilb‘mgmm'.§e1§wieoeBiasggseiegasielng
£25.20 £33er

 

 

6”", can: I: (at. III-On O3< o- 1”! I§< g:
‘05:! i <8: "user‘s.“ #0 <9: I :2. g i

000169

INITIATIVE PETITION

Goad—30>nn 9‘ 6.305qu .. 5e awhile. 1 .3. S0226). seas: — Is i.e....em .o ntn a-e.e .1- !l I)! ”w

 

353—0: 5:. see a: sense-Se os 30 mes-.3 1s. 20:05 I In Reasons. .3.- 3 I: 9:. 351.060. .36 veto. NM,
sen: koala—e .- ...e .53....- 1 £03.: Demo-....- 3 05.. 5e cease .am 52 .7. also: lee B .we g)— g; E

z...- 1 20:5. e a: ate—Ion .322 are I: I Sale $030.06 323?. 30 ...—I‘m II 5s— “.M A

See: nil-aginEIESs-cesros

D: IDM§ Nd >\|
£52.29 .13 nae-.2 L813: ”use .. ...8 "3.23... ... c... “a... rig .. a a ..o be can a ease h?
none 0. .3. 90:0: :0. s 3: use I use so 28 0- ar: .60: I in nu Q\N

1::- s sea-e 08.! 5s: 9.. oi: so 235-...- .e 05:: o; s anions—3e ..oe. bx R

 

38

BIBLIOGRAPHY

BIBLIOGRAPHY

Adam, Stacy J. "Toward an Understanding of Inequity" Journal of Abnormal
and Social Psychology 1963 Volume 67, Number 5: 422-436.

 

 

Burnstein, Eugene and Robert J. WOlosin "The Development of Status
Distinctions Under Conditions of Inequity" Journal of Experimental
and Social Psychology 1968, Number 4: 415-430.

 

 

Camilleri, Santo F. and Anne M. McMahon "Organizational Structure and
Voluntary Participation in Collective-Good Decisions" American
Sociological Review October, 1965 volume 4: 616-644.

 

Campbell, Angus, Philip Converse, warren Miller and Donald Stokes The
American Voter 1964 John Wiley and Sons, Inc.: New York.

 

Campbell, Donald T. and Julian C. Stanley "Experimental and Quasi-
Experimental Designs for Research on Teaching" in Handbook of
Research on Teaching 1963 N. L. Gage (ed) Rand McNally and Co.:
Chicago, Illinois.

 

 

Eagly, Alice, Wendy Wood and Shelley Chaiken "Causal Inferences About
Communicators and Their Effect on Opinion Change" Journal of
Personality and Social Psychology 1978 Volume 36: 424-435.

 

 

Fireman, Bruce and William A. Gamson "Utilitarian Logic in the Resource
Mobilization Perspective" in The Dynamics of Social Movements 1979
Mayer Zald and John McCarthy (eds) Pp. 8-44 Winthrop Publishers,
Inc.: Cambridge, Massachusetts.

 

Kaplowitz, Stan "An Experimental Design to Test Olson's Assumption of
Individual Rationality" 1982. A working paper, Department of Sociology,
Michigan State University.

Kaplowitz, Stan Personal Communication January, 1983, Department of
Sociology, Michigan State University.

 

Mitchell, Robert C. "Public Opposition to Nuclear Power in the Polls:
Knee Jerk Emotionalism or Lay Rationality?" 1981. Paper prepared
for AAAS Symposium on Nuclear Power and the Public,January 4, 1982
in Washington, D.C.

Morrison, Denton E. "Olson's Theory, A Critique and an Alternative" An

unpublished manuscript, May, 1975 Department of Sociology, Michigan
State University.

39

40

Mbrrison, Denton E. and Harriet Tillock "Group Size and Contributions to
Collective Action: An Examination of Mancur Olson's Theory Using
Data From Zero Population Growth, Inc." Research in Social Movements,
Conflict and Changg December 1977 Volume 2: 131-158.

 

 

Olson, Mancur, Jr. The Logic of Collective Action 1965 Harvard University
Press: Cambridge, Massachusetts.

 

Paige, Jeffrey M. "Political Orientation and Riot Participation" American
Sociological Review October 1971 Volume 36: 810-820.

 

Sherif, Carolyn,Muzafer Sherif and Roger E. Nebergall Attitude and Attitude

 

Change 1965 W. B. Saunders Co.: Philadephia, Pennsylvania.

Walster, Elaine, Elliot Aronson and Darcy Abrahams "On Increasing the
Persuasiveness of a Low Prestige Communicator" Journal of Experimental
Social ngchology 1968 Volume 2, Number 4: 325-342.

 

Winer, B. J. Statsitical Principles in Experimental Design 2nd Edition 1971
McGraw-Hill Book Co.: New York Pp. 206-207.