INFORMATION TO USERS
This reproduction was made from a copy o f a docum ent sent to us for microfilming.
While the most advanced technology has been used to photograph and reproduce
this docum ent, the quality o f the reproduction is heavily dependent upon the
quality o f the material subm itted.
The following explanation o f techniques is provided to help clarify markings or
notations which may appear on this reproduction.
1.The sign or “ target” for pages apparently lacking from the document
photographed is “ Missing Page(s)” . If it was possible to obtain the missing
page(s) o r section, they are spliced into the film along with adjacent pages. This
may have necessitated cutting through an image and duplicating adjacent pages
to assure complete continuity.
2. When an image on the film is obliterated with a round black mark, it is an
indication o f either blurred copy because o f movement during exposure,
duplicate copy, or copyrighted materials th a t should n o t have been filmed. For
blurred pages, a good image o f the page can be found in the adjacent frame. If
copyrighted materials were deleted, a target note will appear listing the pages in
the adjacent frame.
3. When a map, drawing or chart, etc., is part o f the material being photographed,
a definite m ethod o f “sectioning” the material has been followed. It is
customary to begin filming at the upper left hand com er o f a large sheet and to
continue from left to right in equal sections with small overlaps. If necessary,
sectioning is continued again—beginning below the first row and continuing on
until complete.
4. For illustrations that cannot be satisfactorily reproduced by xerographic
means, photographic prints can be purchased at additional cost and inserted
into your xerographic copy. These prints are available upon request from the
Dissertations Customer Services Departm ent.
5. Some pages in any docum ent may have indistinct print. In all cases the best
available copy has been filmed.

Uni
International
300 N. Zeeb Road
Ann Arbor, Ml 48106

8503269

S a x t o n , J u d ith E la in e

THE INTERACTION O F POLITICAL ALIENATION AND INFORMATION IN A
LOCAL MILLAGE ELECTION CAMPAIGN: A CA SE STUDY

M ichigan Sta te U niversity

University
Microfilms
International

Ph.D.

300 N. Zeeb Road, Ann Arbor, Ml 48106

Copyright 1984
by
Saxton, Judith Elaine
All Rights Reserved

1984

THE INTERACTION OF POLITICAL ALIENATION
AND INFORMATION IN A LOCAL
MILLAGE ELECTION CAMPAIGN:
A CASE STUDY

By
Judith Elaine Saxton

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

Department of Telecommunication

1984

Copyright by
Judith Elaine Saxton
1984

ABSTRACT
THE INTERACTION OF POLITICAL ALIENATION
AND INFORMATION IN A LOCAL
MILLAGE ELECTION CAMPAIGN:
A CASE STUDY
By
Judith Elaine Saxton

Researchers from a variety of fields have inves­
tigated the role played by alienation in political parti­
cipation and media use.

The research has shown that aliena­

tion is integral to these processes.

Two dimensions that

have appeared in many of the studies as being interrelated
are trust in government and political efficacy.
This study examined how political alienation,
defined as trust in government and political efficacy,
related to information acquisition in a local school millage
election campaign by proposing and testing a model of aware­
ness, information and information source.

The model

predicted that varying levels of awareness, correctness
of and source of information would be found based on scores
on trust in government and political efficacy scales.

The

scales were made situation specific, referring to "school
board" rather than government.
Data were collected from 109 registered voters in
Shiawassee County, Michigan, between April 12 and June 10,
1979.

An attempt was made to recontact all respondents

during the week following the June 11 election.

Judith Elaine Saxton
Major findings of the study included:
1)

Levels of awareness differed among the four

alienation groups.
2)

Different levels of information did not exist,

nor were significant differences as to source of infor­
mation moted, among the four groups.
3)

The proportions aware of and holding correct

information about the election did not increase signifi­
cantly as the election neared.
4)

Age was negatively, but not significantly,

related to favorability.
5)

The presence of children in the home was a

significant factor as was casting a previous vote for a
school millage.
6)

Favorability increased as education' increased.

This did not occur with income or occupation level.
7)

Alienation was found to have little impact on

either the likelihood of casting a favorable vote or the
self-report of the actual vote on the millage.
The major conclusion was that the school millage
was a low information,

low interest election situation.

The study proposed that one cause was the media situation
in the county:

a plethora of media from other areas

overpowering the local media,
mation about the election.

the source of most infor­

ACKNOWLEDGEMENTS

Few projects are completed in a vacuum.
case here.

That is the

There are several people without whose help and

support, the completion of this project would not have
been possible.
First, Dr. Thomas Muth, my major advisor,

saw me

through the four years I was in residency at Michigan State
University.

His willingness to give of his time to chal­

lenge and console me was beyond normal expectations.

With­

out his continued advice and support, I wouldn't have
lasted a month,

let alone through to the end.

Dr. Martin Block served as the advisor for this
dissertation.

He unfailingly stood by me through the many

trials and tribulations of undertaking research in the
"real world".

Through countless delays and trips to

Shiawassee County, he kept me believing that the election
actually would be held.
Dr. Ada Finifter and Dr. Charles Atkin served as
the remaining two members of my guidance committee.

They

were constantly willing to talk through the various prob­
lems associated with this project.

Additionally, Dr.

Finifter first instilled in me a fascination for trying to
understand why people vote as they do and Dr. Atkin, a
questing to discover the impact of information on that

ii

process.
Marin Pearson Allen, and her family, provided
friendship and support throughout the process of writing
this dissertation.

Their basement in Silver Spring,

Maryland, often looked like a battleground as I held
"dissertation camp" there
My students and colleagues at Liberty Baptist
College had to put up with me through the countless hours
of revisions and rewrites.

Their patience, understanding

and prayers will never be forgotten.
My mother, Margaret Saxton Smith, never gave up
believing that this project would be finished, even when
I did.

Her continued belief in me often provided the

strength necessary to continue.
Finally, my Lord and God provided the ultimate
guidance for this project and for my life.

He has provided

meaning and direction to a life that was wandering.
to Him, above all else, that my thanks must go.

It is

TABLE OF CONTENTS

ACKNOWLEDGEMENTS............................................ ii
LIST OF TABLES.............................................. vi
LIST OF FIGURES............................................ vii
CHAPTER I

INTRODUCTION....................................1
Shiawaseee County Background................ 6
Justification of the Stu d y .................. 7
Organization of the Dissertation.......... 10
Endnotes..................................... 12

CHAPTER II

REVIEW OF LITERATURE..........................18
Media U s e ....................................18
Demographics............................. 18
Political Activity...................... 23
Political Attitudes..................... 28
Summary.................................. 35
Political Participation and A ttitudes
35
Demographics............................. 35
Political Participation................. 42
Political Attitudes..................... 45
Summary.................................. 55
School Referenda............................ 55
Awareness/Knowledge..................... 56
Source of Knowledge..................... 57
Voting Correlates : Demographics........ 59
Voting Correlates: Alienation.......... 61
Summary.................................. 64
Hypotheses.................................. 65
Definitions................................. 72
Limitations................................. 72
Endnotes..................................... 74

CHAPTER III

METHODOLOGY................................... 91
Research D e s i g n ............................. 91
The Sample.................................. 93
Questionnaire Development.................. 94
Pretest .....................................101
Data Collection and Preparation.......... 102
Data Analysis.............................. 103
Endnotes....................................Ill

CHAPTER IV

RESULTS....................................... 114
Sample Characteristics.................... 114
Demographics
...................... 114
Attitudes............................... 116
Awareness and Information..............120
Media U s e ............................... 121
Source of Information.................. 123
V o t i n g .................................. 123
Sample Representativeness................. 126
Hypotheses................................. 128
The Derivative M o d e l ................... 128
Campaign Awareness and Information....142
Favorability............................ 149
Alienation.............................. 159
"Source of Information" T y p e s ............ 172
Source Profiles.........................172
Profiles of Source T y p e s ...............180
Unique Source Types.................... 184
Profiles of Source Category T y p e s
190
A Research U p d a t e..........................193
Endnotes....................................200

CHAPTER V

SUMMARY AND CONCLUSIONS..................... 208
Summary.....................................208
Conclusions................................ 217
Limitations................................ 221
Suggestions for Future Research.......... 223
Endnotes....................................226

APPENDIX 1:

FIRST WAVE QUESTIONNAIRE.................... 227

APPENDIX 2:

MAIN WAVE QUESTIONNAIRE..................... 236

APPENDIX 3:

CALL BACK QUESTIONNAIRE..................... 244

APPENDIX 4:

CODEBOOK...................................... 246

SELECTED BIBLIOGRAPHY...................................... 259

v

LIST OF TABLES

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.

Population/Sample Comparison.......................... 95
Political Efficacy.................................... 117
Newspaper Readership and Subscription............... 122
Radio Listenership.................................... 122
Times Voted in 1976-1978............................. 124
Sample Distribution on Trust and Efficacy........... 128
Awareness by Attitude G r o u p.......................... 131
Level of Correct Information......................... 134
Analysis of Variance--Level of Correct Information.134
Source of Information By G r o u p....................... 136
Awareness and Information By W e e k ....................143
Information Acquisition By School District..........146
Source By School District............................ 147
Favorability By Education............................ 150
Favorability By Income L e v e l ......................... 152
Favorability By Occupation........................... 153
Correlation M a t r i x .................................... 162
Betas For Regression Analysis........................ 163
Summary T a ble..........................................164
Regression Summary T a b l e ............................. 165
Means and Standard Deviations for Voting Groups ....168
Standardized Discriminant Function Coefficients....168
Discriminant Analysis Summary T a b l e ................. 169
Discriminant Analysis Prediction Res u l t s............169
Source of Information By Gro u p....................... 173
Awareness and Information By Source................. 174
Local Media Use By Sour c e............................ 175
Source By School District............................ 176
Source By Home Ownership, Vocational Education
and Children At H o m e .............................. 177
Source By Demographic Variables...................... 178
Source Type By Alienation G r o u p ...................... 185
Awareness and Information By Source................. 186
Source Type By Local Media U s e ....................... 186
Source Type By School District....................... 187
Source By Home Ownership, Vocational Education
and Children in H o m e .............................. 188
Source By Age and S e x ................................. 188
Source By Education, Employment and Income..........189

vi

LIST OF FIGURES

1.
2.
3.
4.

Political Participation Typology........................ 4
Information Typology
...............................5
Shiawassee County Derivative M o d e l ................... 141
Awareness and Information by W e e k .....................145

vii

CHAPTER I
INTRODUCTION
One of the underlying precepts of the United States1
Republican form of government is the concept of the informed
electorate, an electorate that bases its voting determina­
tions on accurate and complete information about the candi­
dates and/or issues in any given election campaign.

A

major premise of the First Amendment^ is to assure that the
electorate is informed, thereby supporting a democratic form
2
of government where the individual is sovereign.
Berelson,
et al point o u t :
The democratic citizen is expected to be interested
and to participate in political affairs.
His inter­
est and participation can take such various forms
as reading and listening to campaign materials,
working for the candidate or the party, arguing
politics, donating money, and voting. . . . The
democratic citizen is expected to be well informed
about political affairs.3
The role the mass media has assumed in this process
is one of providing the necessary information to the elec­
torate on which to base on "informed vote".

There are sit­

uations, however, where the mass media may not fulfill that
role.

Areas of the country exist where there are virtually

no local media.^

Suburban areas surrounding a city, and

within the metropolitan area, provide one example.

Such

communities as Farmington Hills, Michigan, and Sun City,

Arizona, have local weekly newspapers'* but daily information
comes from the media (i.e., radio, television, newspapers)
of Detroit and Phoenix, respectively, which may not provide
adequate coverage of the local issues facing these outlying
areas.^

Local election information, then, may not be

readily available in the media relied on daily as a source
of political information.

This local information must be

sought elsewhere.^
This lack of readily available local information is
g
even more critical in school millage elections.
Whereas,
in regular primary and general elections voters may receive
information from political parties on issues and candidates,
no such information vehicles exist in a school or other
millage election.

g

Voters must rely on mass media, quasi­

mass media such as flyers and brochures, or personal commun­
ication as sources of information.

The fact that this

information comes mainly from the school entity sponsoring
the election may further complicate matters as voters
may view the information as necessarily b i a s e d . ^

Voters

must make a concerted effort to acquire information con­
cerning the election, or even an awareness that an election
has been called.
Not all voters are willing to make this effort.

One

such type has come to be known as the "alienated v o t e r " .
According to Levin, the alienated voter is hostile to
politics and disenchanted with the political process, wary
of candidates who spend large sums of money during a

campaign,

skeptical of candidates and issues endorsed by

powerful public figures, tends to believe that campaign
promises and platforms are meaningless and tends not to
vote at a l l . ^

Nettler found that the alienate is one who

has been estranged from and made unfriendly toward his
society, its culture and its political process.

12

This type

of voter, or non-voter, is one with relatively deepseated
"feelings of estrangement, rejection, negativism and unhappiness with the political system or its salient parts".
Alienation may have a different impact in local
voting on ballot issues.

According to Templeton:

Unlike national politics, local political systems
maximize both the access and the potential impact
of interested individual citizens.
To the extent
that local political systems are democratic in
terms of the increased potential for individual
participation, they provide their alienated con­
stituents an avenue for expression. . . . Our data
suggest that the uses to which political partici­
pation is put are considerably more varied than
would be expected on the basis of classical demo­
cratic theory.
Indeed, there is reason to believe
that a democratic system of participation increases
the possibility for participation based upon the
individual’s hostility to and rejection of that
system.
Hamilton and Cohen posit similar behavior in
school referenda:
Alienation probably is endemic and to some extent
omnipresent in school elections, since referenda
furnish what our other political structures deny
alienates, a convenient, direct opportunity to
project their distrust, fears and discontents,
to strike back.at "they" and resist policy plans
of the elite.
Kowalski supports this view:
In most communities voters make no decisions

13

regarding taxes for the national military b u d g e t ,
for local roads, for state income tax, among
other issues.
Therefore . . . a school issue
for operating funds for new buildings is usually
the only tax issue in which the voter has a say
and he or she may vent frustrations over all high
taxes on school elections.
Alienation,

then, can be expressed in a number of

ways in a referendum or school financing election.

Pre­

dicting what action will be taken may depend, in part, on
a refinement of the political alienation measure.

Milbrath

and Goel, building on the work of F i n i f t e r ^ , have proposed
one such refinement:

that alienation is two-dimensional,

comprised of trust in government and political efficacy.

18

Their model (see Figure 1) deals with types of political
participation dependent upon high and low feelings of trust
in government and political efficacy.

They point out that

the various types of alienated action can be explained by
this dichotomization of political alienation.
Trust in Government
High
High
Political
Efficacy L q w

Figure 1:

Low

Active Allegiant
and Conventional
Participation

Radical Action
or Unconventional
Participation

Supportive,
Patriotic and
Ritualistic
Participation (e.g.
voting, enlisting)

Withdrawal
from Politics

Political Participation Typology

19

If the dichotomization holds for political partici­
pation, one should be able to extend the concept to

information and awareness as well, since the literature
shows that alienation and its two components are related to
media use in unique ways.

20

This study proposes and tests

such a derivative model (see Figure 2).

The model predicts

three measures of information (awareness, correctness of in­
formation,

source of information)

from two components of

political alienation (trust in government and political
efficacy).
Trust in Government
High

Low

High High Awareness
Correct Information
Information from
Media

Political
Efficacy

^ ow

^ ow Awareness
Correct Information
Information from
Media

Figure 2:

High Awareness
Incorrect Infor­
mation
Information from
Personal Sources
Awareness
Incorrect or No
Information
Information from
Personal Sources
l ow

Information Typology

This study is, in essence, a case study of one local
millage campaign in one county.

21

It will examine the

campaign from the vote r s ' point of view focusing on where
they get their information.

22

It will examine the concept

of alienation and its effects on information seeking and
holding.

It will also examine those demographic character­

istics which may affect this information-holding.

Shiawassee County Background
The study took place in Shiawassee County, Michigan.
The local election was a county-wide millage election held
by the Shiawassee County Intermediate School District on
June 11, 1979.

The issue under consideration was whether

the district should build and operate a centralized v o c a ­
tional education center for the county, a measure which had
failed twice in the previous ten years

(1969 and 1972).

Shiawassee County is located in the south central
23
portion of the state with an area of 540 square m i l e s .
The population in 1975 was 6 9 , 2 1 8 ^ with an estimated 1977
25
26
population of 69,500.
The county is 62.4% rural.
The
97
98
population is 99.67, white
and 49% male
. Slightly more
than one third (36.6%) of the population is under the age of
29
18 with 55.77. between 18 and 64, and 7.77>, 65 or older.
on
The per capita income is $5904.
The county has one AM-FM combination (WOAP) that
simulcasts during the daylight hours and one daily news­
paper (The Argus P r e s s , circulation:
in Owosso.

31

14,933) both based

There are also two weekly newspapers:

the

Durand Express (circulation: 2,386), published in Durand,
and the Shiawassee Journal (circulation: 6,035), published
in Perry.

The county has no local television station.

32

Various portions of the county receive Lansing,
Detroit and Flint/Saginaw radio stations, resulting in low
33
levels of listenership to WOAP (30% in a recent s t u d y ) .
Lansing, Flint and Detroit newspapers circulate in the

county with most respondents to an earlier study receiving
The Argus Press and one other newspaper.

34

To complicate media matters further, WOAP and The
Argus Press are cross-owned and Owosso is one of the seven­
teen markets where divestiture had been ordered because of
a potential media monopoly situation.

35

Because of these

factors, information concerning local elections may be
difficult to locate in the media used as a daily source of
information requiring a seeking-out of such information.
Justification of the Study
The impact of political alienation, as defined by
the Milbrath and Goel model, has not been applied to the
concepts of information-seeking and information-holding.
The application, however,

seems reasonable.

If alienated

voters experience frustration and anxiety whenever they
attempt to engage in political action and as a result tend
to withdraw from all such action as proposed by Templeton,

36

this withdrawal should apply to information-seeking as well
with the alienated voter not participating in such a b e ­
havior.

The consequences of such a withdrawal could have a

dramatic impact on the political structure of the United
States if that segment of the voting population increases
in size.

Candidates could be elected and issues decided by

a small segment of the population, with a portion of that
segment being under-informed,

thus calling into question the

concept of the rational, informed v o t e r .

The Milbrath and Goel model of political alienation,
with trust in government and political efficacy as its
component p a r t s , was developed to deal with the conflicting
results that had been noted in research on the alienated
voter.

Levin and Eden found that alienation actually en-

37

couraged certain types of voting b e h a v i o r .

Aberbach

found that it was positively related to traditional political activities, such as voting.

38

On the other hand, Erbe

found that the positive correlation between alienation and
political participation disappeared when socio-economic
status and involvement in non-political organizations were
held constant

39

and Olsen found that respondents who felt

incapable were less likely to act politically.

40

In a

study of school bond issues, Thompson and Horton found that
those who felt powerless and isolated in community affairs
voted in lower numbers, but were also more likely to cast
protest votes,

stating that a "no" vote in a bond issue was

a vote against the existing power structure.

41

The primary concern of the Milbrath and Goel model,
as originally conceived, was to explain differing types of
political participation.

In its derivative form in this

study it was used to examine information-seeking and infor­
mation-holding in a manner similar to studies done in infor­
mation diffusion.

The longitudinal nature of the study

allowed for a finer picture of the information-seeking pr o ­
cess.

Most information diffusion research, specifically

that dealing with a news event, has been cross-sectional

in nature and has dealt with critical events:

Eisen­

hower' s decision to seek the Presidency a second t i m e ; ^
the launching of an early satellite, Alaskan statehood and
President Eisenhower's illness;
dent;

a local racial inci-

44 the dropping of Eagleton as Vice-Presidential
45

candidate;
nedy.

43

and, the assassination of President Ken-

46 While these studies are valuable in assessing media

use for diffusion, the dramatic nature of the events studied
may have had an unaccounted-for impact on the rate and
nature of the diffusion.

Troldahl has pointed out that

more study is needed of localized, undramatic events and
that more attention should be paid to issues rather than
events . ^

The Shiawassee millage election provided an

opportunity to study one such localized, undramatic issueevent where interest was minimal.

49

The focus on a school referendum was intentional.

Numerous studies have been undertaken to examine the elec­
toral process for Presidential and Congressional races.
Referenda, particularly school referenda, have been vi r ­
tually ignored.

Hamilton and Cohen state the problem

succinctly:
Referenda campaigns are communication processes
about which little is known except what goes on
at the transmitting end.
Although the activities
and strategies of the campaigns are easily observed,
considerably less is known about the less visible
activities and tactics of opposition and there is a
dearth of anything beyond speculation and casual
observations about the effects of referenda cam­
paigns : How much of the campaign messages actually
gets through? What proportion of the voters
receive information and via which channels? What

10
is the extent and accuracy of the vote r s ' know­
ledge? How much are voters influenced and what
influences them? These aspects of school elections
have been almost totally neglected by the research
to date.51
This study was undertaken in an effort to respond to
a number of those issues:

awareness levels, proportion of

voters receiving information via what channels, extent and
accuracy of information, the impact of demographic charac­
teristics and the influence of political alienation.
The millage election situation provided a unique
opportunity to study one element of the United States polit­
ical system in process.

The study of interrelationships

between information disseminated and awareness and knowledge,
and the impact of political alienation on this relationship,
provided for an opportunity to examine how voters use media
in a local election campaign.
Organization of the Dissertation
The report is structured into six major divisions:
five chapters and an appendices section.

The background,

need and purpose for the study have already been discussed
in the first chapter.
Shiawassee County.

It also contained a description of

The second chapter presents a review of

pertinent literature and a presentation of hypotheses.

The

third chapter explores the methodology that was used in the
study including research design, sampling procedure, ques­
tionnaire development, data collection and analysis.

The

fourth chapter presents the results of the study in tabular

11
form with an explanation of them.

It also analyzes and

discusses the results of the study in light of the liter­
ature discussed in the second chapter.

The last chapter

includes the summary and conclusions of the findings, and
some suggestions for further research.

The material in

the appendices includes a copy of the questionnaire and a
copy of the codebook used in the data coding stage of the
methodology.

12
Endnotes
1The First Amendment to the U.S. Constitution
states:
Congress shall make no law respecting an establish­
ment of religion or prohibiting the free exercise thereof;
or abridging the freedom of speech, or of press, of the right
of the people peaceably to assemble, and to petition the
government for a redress of grievances.
U.S., Constitution,
amend. I .

2

See, for example, Thomas I. Emerson, Toward a G en­
eral Theory of the First Amendment (New Y o r k : Vintage B o o k s ,
1966), especially the discussion on pp. 8-11; Alexander
Mieklejohn, "The First Amendment is an Absolute" in Philip
B. Kurland, ed., Free Speech and Association:
The Supreme
Court and the First Amendment (Chicago: !rhe University of
Chicago "Press", 1975), pp. 1-22.
3

Bernard R. Berelson, Paul E. Lazarsfeld and William
N. McPhee, Voting (Chicago:
The University of Chicago
Press, 1954), ppT 307-8.
^Bogart and Orenstein examined one such situation.
See Leo Bogart and Frank E. Orenstein, "Mass Media and
Community Identity in an Interurban Setting," Journalism
Quarterly 42 (Spring 1965):179-88.
^Farmington H i l l s , Michigan is served by the Farm­
ington Observer and Sun City, Arizona by the Sun City
Citizen.
£
Conway studied a similar situation in the mid-1960s
in a city election in College Park, Maryland, a suburb of
Washington, D.C.
She found that subscribers to the local
weekly paper were more likely to vote than those who sub­
scribed only to a metropolitan daily or no paper.
M. M a r ­
garet Conway, "Voter Information in a Nonpartisan Local
Election," Western Political Quarterly 21 (March 1968):69-77.
Tichenor and Wackman, in a 1571 study set in a suburb
of Minneapolis found that knowledge of local public affairs
was highest among local paper readers who also had high
daily metropolitan usage.
Philip J. Tichenor and Daniel B.
Wackman, "Mass Media and Community Public Opinion," Ameri­
can Behavioral Science 16 (March-April 1973):593-607.
^This behavior may not always take p l a c e . Palmgreen
found that generally people tend to seek information about,
talk about, and learn about, only those issues which happen
to be in the media spotlight.
Philip Palmgreen, "Mass

13
Communication and Political Knowledge:
The Effects of
Political Level and Mass Media Coverage on Political Learn­
ing" (Ph.D. dissertation, University of Michigan, 1975).

g
A millage is taxation in mills per dollar of valu­
ation.
A mill is equal to one-tenth of a cent.
In the pre­
sent situation it refers to an increase in property taxes
to cover the cost of additional educational facilities.
9
Fleitas terms these situations minimal informa­
tion elections which are characterized by a dearth of
information as to election issues and partisan considera­
tions.
Daniel William Fleitas, "The Underdog Effect:
An
Experimental Study of Voting Behavior in a Minimal Infor­
mation Election" (Ph.D. dissertation, Florida State Univer­
sity, 1970) .
■^Hamilton and Cohen point out that in Austintown,
alienated voters did not believe the facts presented by the
school board, claiming that the information was biased in
order to get a favorable vote.
Howard D. Hamilton and Syl­
van H. Cohen, Policy Making by Plebiscite:
School Referenda (Lexington, Mass.:
d T c . Heath and Company, 1974),
pp. 140-42.
■^Murray B. Levin, The Alienated Voter:
Politics
in Boston (New York:
Holt, Rinehart and Winston, 1960).

12

Gwynn Nettler, "A Measure of Alienation," Ame r i ­
can Sociological Review 22 (December 1957):672-74.
^ L e s t e r W. Milbrath and M. L. Goel, Political Pa r ­
ticipation:
How and Why Do People Get Involved in Politics,
2d e d ., (Chicago: Rand McNally College Publishing Company,
1977), p. 62.
^ F r e d e r i c Templeton, "Alienation and Political Par­
ticipation:
Some Research Findings," Public Opinion Quart­
erly 30 (Summer 1966):261.
^ H a m i l t o n and Cohen, Policy M a k i n g , p. 205.
"^Joan P. Sullivan Kowalski, Voter Behavior and
Campaign Strategies in School Finance Elections (Arlington,
V a .: Educational Research Service, I n c ., 1977), p. 18.
"^Ada W. Finifter, "Dimensions of Political Alien­
ation," American Political Science Review 64 (June 1970):

14
407.
1ft
Milbrath and Goel, Political Participation, p. 70.

19ib±d.
20

This is developed in detail in the Review of
Literature in Chapter II.

21

The study is part of a larger effort undertaken by
a team of researchers from Michigan State University using
a metric multi-dimensional scaling technique known as Gali­
leo.
The larger study was designed to track attitudes of
registered voters in the county concerning vocational edu­
cation across time and to provide the funding agency with
information on the optimal campaign messages to assure
passage of the measure.

22
Anderson and Meyer point out the importance of
the study of source of information:
"Finally we desperately
need field research which will document the behaviors of
individuals vis a vis their communication outlets. We need
scientists who will systematically observe the rich variety
of behaviors related to information flow of which infor­
mation diffusion research is at least a beginning".
James
A. Anderson and Timothy P. Meyer, "Functionalism and the
Mass Media," Journal of Broadcasting 19 (Winter 1975):21.
23

Michigan Department of Commerce, Office of Econ­
omic Expansion, Economic Profile of Shiawassee County
(Lansing, Michigan, September 1966).
24

U.S., Department of Commerce, Bureau of the Cen­
sus, 1977 City County Data Book: A Statistical Abstract
Supplement (Washington, D.C., 1977).
25

David I. Verway, Michigan Statistical Abstracts,
13th ed. (East Lansing, Michigan:
Michigan State University Graduate School of Business Division of Research,
1978), p. 42.
26I b i d ., p. 47.
27Ibid., p. 69.
28I b i d .

15
29Ibid.
30Ibid., p. 383.
31Ibid, p. 925.
32

The Detroit media include two daily newspapers
(The Detroit Free Press and The Detroit N e w s ) , twentyeight radio stations ^twenty FM and eight AM) and eight
television stations
(four VHF, one of which is from Canada,
and four U H F ) . The Lansing media include one daily news­
paper (The Lansing State Journal) , eleven radio stations
(three FM and three AM from Lansing and three FM and two
AM from East Lansing) and three television stations, one
of which is UHF.
The Flint media include one daily news­
paper (The Flint Journal) , eleven radio stations (five FM
and six AM) and five television stations, three of which
are UHF.
33

These results are from the baseline study under­
taken prior to the introduction of the overall message
campaign.
Martin Block and Joseph Woelfel, Area Vocational
Education Campaign Strategies and Techniques Project, Phase
I: Campaign Strategies (East Lansing, Michigan;
College
of Communication A r t s a n d Sciences, Michigan State Univer­
sity, September 1978), p. 19.
/

This result is from the baseline study.
35

Federal Communications Commission, Amendment of
Multiple Ownership Rules (Newspapers), Second Reoort and
Order, 50 g.C.C. 2d 1046', 32 RR Sd 954, 40 Fed. feeg'.' 6449
(1975); On reconsideration 53 F.C.C. 2d 589, 33 R.R. 2d
1063 (1975). The FCC decision was appealed to the Court
of Appeals (D.C. Circuit) where the decision was overturned.
Upon appeal to the U.S. Supreme Court, the FCC decision
was upheld. National Citizen's Committee for Broadcasting v.
F.C.C., 555 F .2d 938 (D.C. CIR. 1977), 39 R.R. 2d 1463;
FCC v. National Citizen's Committee for Broadcasting, 436
U.S. 775, 98 S. Ct. 2096 (1978).
3^Templeton,
pation," p. 256.
37

"Alienation and Political Partici­

Murray B. Levin and Murray Eden, "Political
Strategy for the Alienated Voter," Public Opinion Quarterly
26 (Spring 1962):47-63.

16
OQ

Joel Aberbach, "Alienation and Political Behav­
ior," American Political Science Review 63 (March 1969):
86- 9 9 .
QQ

William Erbe, "Social Involvement and Political
Activity:
A Replication and Elaboration," American Socio­
logical Review 29 (April 1964):198-215.
^ M a r v i n E. Olsen, "Two Categories of Political
Alienation," Social Forces 47 (March 1969):289-99 .
^Htfayne Thompson and John E. Korton, "Political
Alienation as a Force in Political Action," Social Forces
38 (March 1960):190-95.
/O
Wayne A. Danielson, "Eisenhower's February Deci­
sion:
A Study of News Impact," Journalism Quarterly 33
(Fall 1956):431-41.
/Q
Paul J. Deutschmann and Wayne A. Danielson, "Diffu­
sion of Knowledge of a Major News Story," J ournalism Quart­
erly 37 (Summer 1960):345-55.
^ D o n a l d I. Warren, "Mass Media and Racial Crisis: A
Study of the New Bethel Church Incident in Detroit," Journal
of Social Issues 28 (1972):111-32.
^ K e n n e t h G. Sheinkopf and M. Timothy O'Keefe, "The
Eagleton Affair:
A study of News Diffusion," paper pre­
sented to the Mass Communication Division of the Interna­
tional Communication Association, Montreal, Canada, 25-28
April 1973, cited in Sidney Kraus and Dennis Davis, The
Effects of Mass Communication on Political Behavior (Univer­
sity Park, P a . : The Pennsylvania State University P r e s s ,
1977), pp. 167-68.
^ H a r o l d Mendelsohn, "Broadcast vs. Personal
Sources of Information in an Emergent Public Crisis:
The
Presidential Assassination," Journal of Broadcasting 8
(Spring 1964):147-56.
^ V e r l i n g C . Troldahl, "Studies of Consumption of
Mass Media Content," Journalism Quarterly 42 (Autumn 1965);
596-606.

48
voters.

Turnout for the election was 12% of all registered

17
49

Stone points to the need for expanding the alien­
ated voter concept to include the study of alienation in the
context of the local political process.
Clarence E. Stone,
Local Referendums: A n Alternative to the Alienated Voter
Model," Public Opinion Quarterly 29 (Summer 1965):215.
■*®For information on voting in Presidential elec­
tions see,, for example, Angus Campbell, Philip E. Converse,
Warren E. Miller and Donald E. Stokes, The American Voter
(Chicago:
The University of Chicago P r e s s , i960);
Angus
Campbell, Gerald Gurin and Warren E. Miller, The Voter
Decides (Evanston, 111.:
Row, Peterson and Company, T954);
Warren E . Miller and Teresa E. Levetin, Leadership and
Change:
The New Politics and the American Electorate
(Cambridge, M a s s .: Winthrop Publishers, I n c ., 1976);
Norman H. Nie, Sidney Verba and John R. Petrocik, The Chang­
ing American Voter (Cambridge:
Harvard University Press,
19/6); and, Rebecca Colwell Quarles, "Mass Media and Voting
Behavior:
The Accuracy of Political Perceptions Among First
Time and Experienced V o t e r s ," Communication Research 64
(October 1979):407-36.
For information on Congressional
elections see, for example, Angus Campbell, "Surge and
Decline:
A Study of Electoral Change," Public Opinion
Quarterly 24 (Fall 1960):397-418;
Andrew T . Cowart, ^Elec­
toral Choice in the American States:
Incumbency Effects,
Partisan Forces and Divergent Partisan Majorities," American
Political Science Review 67 (September 1973):835-52;
Barbara Hinckley, C . Richard Hofstetter and John H. Kessel,
"Information and the Vote:
A Comparative Election Study,"
in Richard G. Niemi and Herbert F. Weisberg, e d s ., Contro­
versies in American Voting Behavior (San Francisco:
W.H.
Ereeman Company, 1976), pp. 274-96;
Gerald H. Kramer,
"Short-Term Fluctiiations in U.S. Voting Behavior," American
Political Science Review 65 (March 1971):131-43;
Donald E.
Stokes and Warren E. Miller, "Party Government and the
Saliency of Congress," Public Opinion Quarterly 26 (Winter
1962):531-46; and, Edward R. T u l t e , "Determinants of the
Outcome of Midterm Congressional Elections," American Polit­
ical Science Review 69 (September 1975) :812-25"!
■^Hamilton and Cohen, Policy M a k i n g , p. 137.

CHAPTER II
REVIEW OF LITERATURE
Any study that crosses academic disciplines must, of
necessity,

touch upon a large body of literature.

The pr e ­

sent study took a political science model and applied it to
a number of media variables in an educational setting.

This

necessitated an exploration of the literature from the areas
of media use, political participation research and school
referenda to present the theoretical underpinnings of the
derivative m o d e l .
Media Use
Demographics
The earliest research into media use focused almost
entirely on the demographic characteristics of users of v a r ­
ious media.

For example,

in 1950 Star and Hughes reported

that the college educated were much more likely to expose
themselves to informational or public affairs mass communi­
cation.^
the media.

Another area of research explored time spent with
Ross and Bostain, in a three-day diary study

with 523 Wisconsin farm families, found that television was
the leading medium in average time use with radio second and
2
reading third.
Men spent one-fifth of their waking time in
contact with mass media while women spent one-fourth.
18

19
The research into demographic differences proved to
be more prolific.

Westley and Severin, in a study of the

adult population in Wisconsin in 1961-62, were able to de­
velop a demographic profile of the daily newspaper no n ­
reader as someone:
. . . low on the scale of occupations, low in ed­
ucation, low in income and likely to regard h i m ­
self as "working class" rather than "middle class"
. . . much more likely to be a farmer than a city
dweller and, among the non-farm population, some­
what more likely to be a resident of a place under
10,000 population than a larger . . . no more likely
to be of one sex or the other . . . tends to be
either in his 20's or in his 70's or older.3
In addition they found the non-reader to be likely to belong
to few formal organizations, an infrequent church attender
and visitor of friends

and relatives. On political vari­

ables,^ the non-reader tended to have no political identifi­
cation or leaning and to be a non-voter.
A study undertaken in 1959 allows for a similar time
period comparison of the non-television household.

Westley

and M o b i u s , in a survey in Madison, Wisconsin, found that the
non-television household was characterized by certain highstatus and certain low-status occupations, by a chief wage
earner with less than a high school education or who has
gone beyond college, and by a household with two adults and
more than one child between the ages of five anf fifteen."*
Samuelson, Carter and Paiggels, in a study with 203
adult males in Redwood City, California,

found that the

more educated persons do not, on the average, use the
broadcast media as much as the less educated.

They relate

20
the difference to available time:
Clearly, however, the time demands of role activ­
ities reduce the amount of time that the more edu­
cated man might otherwise be inclined to spend view­
ing television and, to a lesser degree, listening
to radio and reading newspapers.
Westley and Severin,. again relying on data col­
lected in their 1961-62 Wisconsin study, found demographic
differences in groups who assign high credibility (opera­
tionalized as responses to the question "As between tele­
vision, radio and newspapers, which one do you feel gives
the most accurate news?") to the various media.^

Those who

rated newspapers most credible tended to be persons of high
socio-economic status with the education factor in this
measure positively related to high newspaper credibility
and negatively related to trust in television as a source
of n e w s .

The authors point out that "the relationship of

education to media credibility is one of the strongest and
Q

most consistent in the entire study".
variables,

On the political

the person who sees newspapers as most credible

is most likely to be an independent.

Those who voted in

the last general election at the time of the study had a
slight tendency to rate the newspaper as higher and tele­
vision lower than those who did not vote.
Greenberg uncovered demographic differences between
those relying on television and those relying on newspapers

9

as the principal source of n e w s .

Newspaper readers tended

to be male with some college education.

Slightly more men

over forty fell into this group than men under forty

21
(71% compared with 65%).

Women with some college and over

the age of forty tended to rely on newspapers, while all
other categories of women relied more heavily on television.
Males with less than thirteen years of school relied
slightly more on newspapers (under forty. 52%; over forty,
56%).

On media credibility measures,

those rating news­

papers as more credible tended to be older males of high
education levels, a confirmation of earlier findings re­
garding education.
A t h e n s , Ohio was the setting for a study by Sargent
and S t e m p e l . ^

To explore the relationship of media use,

poverty and alienation
respondents:

11

, the researchers used two sets of

39 families in the poverty group and 114 in

the general population.

They report that the general popu­

lation had a high, although not significant, use of news­
papers and a significantly lower use of radio and tele­
vision.

The general population also had a significantly

lower use of radio and television as news media.
Wade and Schramm found that television is more
likely to be used as a source of political information by
those with little education, females, nonwhites,
blue-collar workers.

12

farm and

Newspapers, on the other hand, were

more likely to be used by highly educated groups, whites,
males, professional, managerial and white-collar workers,
and high income groups.

Television was used by approxi­

mately 557o of the population while newspapers were used by
25%.

22
Kline, using path analysis to discover the extent
to which media use in contemporary society is related to
lifestyle or ecological demographic location, found that
older families and those higher on the occupational ladder
are more likely to read newspapers.

13

He also discovered

that the better educated spent less time with the newspaper,
but felt this could be a result of education leading to an
improved capability for selective and/or faster reading.
Clark and Ruggels also found an education-newspaper
link.^

They report, in their survey of 1250 male and

female heads of households in Seattle, that education was
positively related to a preference for newspapers as a
source of public affairs information.

Becker, Sobowale

and Casey also found a newspaper-education link with both
education groups reporting newspaper dependency for national
and local news but

with the better educated group being

more dependent on newspapers than television.

The younger

members of their sample were more television dependent,
particularly for local n e w s .

Vinyard and Sigel report that

reliance on a diversity of media increases as education
increases.^

Barnett and Hughes note that increased media

use produced an increase in political knowledge that was
most pronounced for those with less than a high school
education and almost non-existent for college g r a d u a t e s . ^
This trend to examine media use demographically
continues.

In a recent study of approximately 1200 adults,

Poindexter developed four categories of television news

23
usage:

non-news viewers

(21%), non-network news viewers

(7%), non-local news viewers
(67%).

(5%), and, news viewers

She found that the younger the age group, the more

pronounced the tendency not to view television news;

that

those with high school and technical degrees were least
likely to view; and, that those with large incomes were also
least likely to view.

Most non-viewers, however, did read

daily newspapers.
The research clearly points to media use differences
based on demographics.

Newspaper reliance is strongly asso­

ciated with higher education levels, higher occupational
status and higher incomes.

Additionally, newspaper reading

is associated with being older, being male and being white.
Conversely, television viewing is associated with lower
levels of education, income and occupational status as well
as being young, being female and being non-white.
Political Activity
A second major area of research into media use
focuses on its relationship to various forms of political
activity.

This focus is based upon the assumption, as

Burstein states:
. . . that contact with national political centers,
directly or indirectly, through contacts with edu­
cational institutions, government bureaucracies,
mass media, formal organizations, or informal
groups, produces an inclination to participate
actively in political l i f e . 19
Some of the earliest voting behavior studies con­
sidered media use variables.

Lazarsfeld, Berelson and

24
Gaudet, in their study of the 1940 Presidential election in
Erie County, discovered that about half of the population
ignored front page news stories and political broadcasts at
the peak of the campaign.

20

They also found that those

highly exposed to one medium tended to be highly exposed to
other media as well.

Berelson, Lazarsfeld and McPhee, in a

study of voters in Elmira, New York during the 1948 Presi­
dential election found that of the total populace only 36%
in June and 38%, in October reported paying a great deal of
attention to news about the election in the mass media.

21

As in the Erie County study, they found that those highly
exposed to one medium tended to be highly exposed to others
as well.

In addition, demographic differences were found

22

relating to high and low exposure levels.

Glaser examined television's effect on voting turnout as compared with other mass media.

23

He found that the

differences existed between those who do and those who do
not vote:

television viewers have a higher turnout than

non-viewers, and newspaper readers have a higher turnout
than non-readers.

He also found that the association

between newspapers and turnout was higher than that between
television and turnout, and that those differences remained
when controlled on demographic variables.
Media usage and political activity were compared by

2/
Becker and Preston.

Media usage was operationalized as

the number of media used by the respondent while political
activity was determined through psychological involvement

25
in the campaign, political efficacy, voting tendencies and
actual involvement (other than voti n g ) .
examined effects of education and income.
file patterns were discovered.

The study also
No clear pro­

Media usage was found to

be more highly associated with concern about the election
outcome, interest in the campaign, voting in 1964, and
attempts at personal persuasion.

Education was associated

with political efficacy, voting in previous elections, time
of voting decision, and tendency to write letters.
Prisuta examined the relationship between media
exposure and political variables dealing with voting turnout,
time of voting decision, information level, interest level
and political involvement using data from the 1968 Presidential election.

25

He found that newspapers were the only

medium to correlate significantly with voter turnout; that
there was no relationship between media exposure and time of
voting decision; that information level correlated strongly
with media usage (as exposure rises so does information
level) with newspapers exhibiting the strongest relation­
ship; and, that the same tendency was apparent in interest
level but that this did not carry over into the political
activity area.

He concluded that lack of involvement in the

political system may be a result of low media exposure but
that "it is also at times a cause, which involves the delib­
erate avoidance of media information due to apathy, sociali,,26
zation or cross-pressure .
Chaffee and McLeod report what they believe to be a

26
consistent information-seeking habit by the most informed.

27

In a study of gubenatorial and Congressional races in M a d i ­
son, Wisconsin, they found that those subjects with the most
knowledge about the campaigns and those who expected to dis­
cuss the campaigns in the future were the groups most likely
to request informational pamphlets about the candidates.
O'Keefe and Mendelsohn report that the non-voter
was less likely to attend to news and public affairs media
content and that television was the prime medium of political
28
information for this g r o u p .

Exposure to newspapers and

television n e w s , and attention to news content generally,
increased with age, education and political interest.
Dreyer utilized data from five Presidential elec­
tions (1952, 1956, 1960, 1964, 1968) to examine the extent
to which the American electorate employs the mass media to
follow the progress of Presidential campaigns.

29

He re ­

ported that the larger intake of current political infor­
mation that results from a more active monitoring of cam­
paign activities through the mass media did not have the
same impact as the smaller intake among those less actively
exposed.

The latter group was far more susceptible to the

"slightest dose of current information, since their under­
standing and 'mass' of stored political knowledge is so
impoverished".
Newspapers, of all media types, seem to have the
greatest impact on political knowledge and activity.

Al e x ­

ander, using data collected in 1968, found that in local

27
elections a clear understanding of candidates and issues
was garnered through newspapers

(40%), followed by tele31
vision (26%) , radio (67o) and magazines (1%) .
Twenty-three
percent reported that their information came from other

people.

Barnett and Hughes,

in a preliminary investigation

using thirty-five students in a Communications class at
Rensselaer Polytechnic Institute,

found that newspaper

reading was a better predictor of "level of certainty" than
was exposure to any other media,

32

while in a much earlier

study, Nafziger and his colleagues found that avid readers
of newspapers tended to be better informed than light
readers.

33

O'Keefe, Mendelsohn and Lui, using a panel of

voters in Summit County, Ohio, found a shift in media reliance between 1972 and 1974.

In 1972, television ranked

above newspapers as the dominant source of political infor­
mation.

In 1974, newspapers ranked above television and

fully 43% of those who had named television in 1972 named
newspapers in 1974.

35

Newspaper readership also appears to be directly
related to voting.

Buchanan reported in 1956 that voters

were significantly more likely to be subscribers to a local
weekly newspaper and to a nearby daily and that newspaper
subscription related to voting even when education was held
O

constant.

f.

In a study of the 1965 city election in College

Park, Maryland, Conway found that newspapers were the major
source of information about the campaign.

37

Additionally,

she found that subscribers to the local weekly papers were

28
much more likely to vote than those who subscribed to no
paper or only to the metropolitan dailies.

McLeod and his

colleagues also found a newspaper-vote relationship with the
frequent newspaper readers being more likely than other
young people to turn out to vote.

38

Quarles, similarly,

in

an analysis of 1972 Presidential election voting data,
suggested that first time voters were more likely to respond
to campaign stimuli and to learn about candidates and issues
39
through newspapers.
Media use, then is related to political activity in
two ways.

First, increased media use leads to more concern

about election outcomes, more voting, more attempts at p er­
sonal persuasion, greater turnout, increased information
levels and an increased likelihood to request more infor­
mation.

Second, political behavior is related to type of

media used with newspaper reliance associated with greater
turnout, a clearer understanding of candidates and issues,
a greater level of certainty, voting and being better in­
formed.

Television, on the other hand,

is seen as the

prime political information medium for the non-voter.
Political Attitudes
A third area of related media use research is the
exploration of its relationship to political attitudes.
It is held by some that media use is not directly linked to
political action, as proposed in the previous section, but
rather is a contributor to a process in which various types

29
of political predispositions are formed which ultimately
influence political a c t i o n . ^

In other words, the influence

that media presentations will have on people will depend on
what people bring to the media:
Their interests, their forms of
difference or opposition, their
naivete', and their established
tions set the way in which they
the presentations.41

receptiveness, in­
sophistication or
schemes of defini­
initially receive

The research falls into several areas.

A number of

researchers have examined the relationship of media to
political interest.

42

Clark and Ruggels, in a sample of

Seattle voters, found that while television was less pre­
ferred as a source of political information than newspapers,
the broadcast media were the major sources of information
for those having average or below average knowledge of and
interest in public affairs.

43

Similarly, Atkin, Galloway

and Nayman report that the amount of political reading in
newspapers is the strongest correlate of both political interest and political knowledge.

44

Education and social

class were both found to be moderately related to print
media but unrelated to electronic media use, and that when
these two demographic variables were controlled the print
media correlations with political interest and knowledge
were only slightly reduced.

McCombs and Mullins found that

while level of education showed a strong relationship with
media exposure to public affairs, political interest was
consistently as strong a correlation.

45

In their sample of

college students, the "extra" media use associated with high

30
political interest consisted "disproportionately of

ex­

posure to national and international news, analyses of social
issues and editorials" . ^
The relationship of alienation and media use has
captured the attention of a variety of authors.

Pietila

points o u t :
The concept of alienation is important in massmedia research precisely because the so-called
"information explosion" may indeed be one factor
contributing to a feeling of alienation.
The
amount of information available is so enormous
that it may be difficult to form a crystallized
and articulate conception of events and their
causes; under such conditions the individual may
easily feel helplessly caught up in the train of
events.^7
In a study of Finnish voters, Pietila found that alienated
individuals use newspapers,

television and radio for vicar­

ious reasons more than the non-alienated who use media more
for informational reasons.

48

McLeod, Ward and Tancil found

that the more alienated respondent was relatively uninter­
ested in nonsensational headlines devoted to political and
49
economic affairs.

They concluded that the alienated

should have little interest in government news.

Katz and

Foulkes investigated the escape function of the media and
found that alienation produces a desire to escape.'*®

They

note that, according to popular culture, people who are
alienated turn to the entertainment fare of the mass media
as a substitute gratification, which in turn brings about a
further withdrawal from political and social a c t i o n . ^
Alienation and the desire to escape have been

31
operationalized as avoidance of specific media content.
Blumler and McQuail,

in their study of British electoral

behavior, identified nine reasons for avoiding party broadcasts.

52

Previous vote decision (37%), distrust of poli­

ticians' messages

(35%) and lack of interest in politics

(267o) received the most mentions.

McLeod and Becker, using

substantially the same list of reasons, uncovered three
underlying dimensions:
ation.

53

partisanship, relaxation and alien-

They found that heavy television users tended to

be high on all three avoidance dimensions and that enter­
tainment fans were more likely to prefer to relax when
watching rather than watching political material.
sanship was the best predictor of avoidance.

Parti­

Those avoiding

the campaign media were less interested in politics, exposed
themselves less often to general public affairs media and
discussed politics less.
Semlac and Williams investigated avoidance in the
1976 election campaign and found that while those high on
avoidance tended to view more television daily, they viewed
less network news.

54

They suggest that study of the concept

of alienation could offer some insight into the question of
what might be inherent in the context of political media
that forces people to avoid both political messages from
the media and the entire political decision-making process.
Political efficacy and trust in government have also
been related to media use.

Eldersveld found that personal

contact brought out voters who felt they were politically

32
55
efficacious.

This group also has a greater tendency to

be exposed to other media during the campaign, especially
newspaper coverage.

Eulau and Schneider report that the de­

gree of "relatedness" a person feels to the political p r o ­
cess is associated with media u s a g e . ^

Defining relatedness

as a combination of the degree to which a person has inter­
nalized political expectations and the degree to which that
person is efficacious, they found that highly related indi­
viduals expose themselves to the major media of political
communication with much greater frequency.

In a more r e ­

cent study of white, black and Mexican-American adults, Tan
found that white adults who felt politically efficacious
were more likely to use the mass media for public affairs
information than those who felt inefficacious.^^
lationship did not hold for the other two g r o u p s .

The r e ­
The

study also found that political participation significantly
predicted mass media use for public affairs information
among all three groups and that politically active respon­
dents were more likely to use newspapers and television for
public affairs information than those not actively involved.
Becker, McLeod and Ziemke, in a multiple regression
analysis of the relationships between demographic variables,
social-psychological variables, several political orienta­
tions , and uses and avoidances of the political content of
the media,

found that while the demographic and social-

psychological variables showed little relationship to media
use or avoidance orientations, political orientations were

33
quite strongly related to political media use orientations.

58

Persons scoring highly on the efficacy scale were found to
be frequent users of the media for vote guidance and sur­
veillance purposes while persons with high levels of trust
in government tended to be the highest excitement seekers.
O'Keefe, approching the same problem from a different per­
spective, found that high reliance on the mass media for
political purposes was associated with more positive or
integrative citizen orientations toward the political
system.

59

He reported that the greater the reliance on tele­

vision, the more likely the respondent would believe that
politicians are altruistic and that their vote is effica­
cious and the less likely they would feel powerless and
alienated.

Newspaper reliance was even more strongly asso­

ciated with these ide a s .
In their examination of the relationship between
media use and voting, Lang and Lang point out that "persons
with above-average political sophistication (and therefore
less subject to a 'chronic distrust' of politics) are in the
habit of checking one source of information against an ­
other".^®

For those who feel debarred from any channels of

political influence (the inefficacious) the authors wonder
if they would distrust all mass media sources and believe
only what their friends tell them.

They also noted that

television has an especially strong appeal for the chron­
ically distrustful.
Robinson supports this final contention with data

34
from an examination of television users on three attitudinal
61
variables.
He found that of those relying only on tele­
vision as a source of information, 91%, felt they couldn't
understand politics,

34% thought that government leaders

were crooked, and 68% thought Congressmen lose touch with
their constituents after the election.

Comparable figures

for those who do not rely on television were 63%,, 21%, and
47%, respectively.
In a study in Syracuse, New York in 1977 with 460
heads of households, Becker, Sobowale and Casey found
further support for the relationship between media use and
62
trust.
Newspaper dependency at the local level was found
to be positively related to trust of local officials.

Tele­

vision dependent persons tended to be less favorable towards
and trusting in local leaders than those not dependent on
that medium.
Political attitudes are related to media use such
that the less the alienation, distrust and inefficacious
feeling, the more the use of media for public affairs infor­
mation.

Newspaper readership is associated with high levels

of both trust in government and political efficacy.

The

amount of reading is also the strongest correlate of both
political interest and political knowledge.

Television, on

the other hand, appeals to the inefficacious and distrust­
ful, those high on alienation and avoidance.

35
Summary
A distinct pattern emerges from the literature.
Newspaper readership is clearly related to feelings of
attachment to the political system and behaviors that demon­
strate that attachment.

These include feelings of trust and

efficacy, a high level of political interest and political
knowledge, a clearer understanding of candidates and issues
and voting.

Television viewing is associated with aliena­

tion, distrust, inefficacy, a limited amount of information
and non-voting.

Additionally demographic differences can be

seen with newspaper readership being associated with higher
socio-economic status.

It is also inportant to note that an

increase in media use for political information leads to an
increase in interest, knowledge, trust, sense of efficacy
and v o t i n g .
Political Participation and Attitudes
Demographics
A variety of demographic factors are related to
political participation:

age, sex, race, social class, in­

come, education and socio-economic status (a combination of
the previous three).

In an early study, Connelly and Field,

focusing in the non-voter in the 1940 Presidential elec63
tion, developed a profile of the non-voter.
They deter­
mined that races other than white had a lower percentage of
voters.

Women and the young were also less likely to vote

as were service workers

(e.g., domestics, firemen, policemen,

36
waitresses, etc.) and manual workers, whether skilled or
unskilled.

The relationship between income, or, in their

terms, economic security, and education with non-voting is
less clear cut.

Whereas non-voters were located primarily

at the lower end of both scales, when combined, economic
security appeared to be more influential.

A larger pro­

portion of non-voters were found among those with maximum
education and minimum economic security than among those
with minimum education and maximum economic security.
larly, Lang and Lang

Simi­

found that non-voters in California

contained a high concentration of older people with little
education (non-high school graduates who are over 55). ^
This group also exhibited, generally,

less involvement in

the outcome of the election and in the campaign that led
up to i t .
A national sample of non-voters undertaken in July,
1976 by O'Keefe and Mendelsohn showed that this picture did
not change substantially in thirty-six y e a r s . ^

The n o n ­

voter was found to be female (52%) more often than male
(48%), young (55% of the non-voters were between eighteen
and thirty-four), and at the lower end of the economic
scale (51% had incomes of less than $10,000;
comes between $10,000 and $14,999).

23% had in­

The non-voters'

educational level was also low with 21% having an eighth
grade educational attainment or less and another 61% having
an educational attainment of the ninth to the twelfth grade.
Since the 1940 study offers no percentages as to race, a

37
comparison between the two studies cannot be made.

However,

O'Keefe and Mendelsohn report that of their non-voters,737,
were white, 17% were black, 17> Oriental, 8% Chicano and 17.
other.

The authors point out that traditionally the n o n ­

voter has been characterized as "likelier to be

young, less

educated, less affluent, single, female, geographically
mobile",

66

and their data appears to add new support to that

contention.
Eulau and Schneider, using a national area probabil­
ity sample, examined the relationship of political related­
ness, a combination of political efficacy and responsibility,
and d e m o graphics.^

They found that persons scoring highly

on political relatedness were more often among those with
higher incomes, those of higher occupational status, urban­
ites and the better educated.

Low scorers were more often

found at the opposite end of these demographic groupings.
Thompson and Horton reached a similar conclusion.

68

In

their Berkeley, California sample, they found a high inci­
dence of non-voting and disinterest among those with lower
socio-economic status.
Erbe reports that those with higher social status
are more likely to register, to vote, to be interested in
politics, to discuss politics,

to belong to a politically

relevant organization and to attempt to influence the
political views of others while for those with a lower
social status, the reverse is true.

49

Bennett and Klecka,

utilizing national election data for 1964, 1966 and 1968,

38
found education to be the best predictor of political actions
when compared with occupation and income.^
Milbrath and Goel, citing a wide variety of studies,
report that those with a higher social class standing, higher
income and higher level of educational attainment are more
likely to p a r t i c i p a t e . ^

Men are more likely to participate

than women, and whites are more likely than blacks.

The

relationship between age and participation was found to be
curvilinear with participation increasing steadily until the
middle years after which it begins to decline as age
advances.
The relationship between social status and partici­
pation takes an unexpected turn on a community l e v e l .

Minar

analyzed aggregate voting data on elections and referenda in
forty-eight suburban elementary school districts in Cooke
County,

Illinois.

72

He found that communities with higher

aggregate status levels tended to show lower participation
and lower levels of dissent.
Demographic characteristics have also been found to
be related to political attitudes, particularly political
alienation and its two components, political efficacy and
trust in government.

Thompson and Horton and Templeton

both found that alienation was closely related to those in
the lower socio-economic strata.

73

While the former

suggests that political alienation may be a mediating factor
between socio-economic status and political participation
which provides an alternative to "subcultural orientations

39
or self-interest explanations of political b e h a v i o r " ^
among that group, the latter claims that alienation is
manifested in a withdrawal from political life.

Templeton

found in his 1960 Berkeley, California sample that the alien­
ated were found disproportionately among bla c k s , manual
workers,

those with less than a college education, those

who identify with working or lower classes and those in the
lower two socio-economic quartiles (as determined by the
interviewer).

Levin found similar results in Boston:

feel­

ings of powerlessness were greatest among elementary or
high school graduates,

those employed in blue or white

collar jobs and those in lower income groups, although the
data only implies the relationship of income and power-

i
lessness.75
Dean,

in a study in Columbus, Ohio, found that

there was a negative correlation between social status,
defined as a combination of occupation, education and in­
income, and alienation; a positive correlation between
advancing age and alienation;

and, a negative correlation

between a rural background and alienation.^
tions were statistically significant but low.

The correla­
Dean con­

cludes that, in general, alienation decreases as status in­
creases.

He points out, however,

be a "personality

that alienation may not

'trait', but a situation-relevant

variable".^
Middleton examined the relationship between race,
education and alienation in a central Florida sample in the

40
summer of 1962.

78

He found that the percentage of blacks

who felt alienated was far higher than the percentage for
whites for every type of alienation except cultural estrangement.

79

For both racial groups, education was inversely

related to alienation, such that as education level in­
creased, the degree of alienation decreased.
similar results

Olsen found

in a study of race and alienation in

Detroit in the late 1950's.

80

Blacks were more alienated

than whites even when occupation was controlled.

For occu­

pation, as status increased, alienation decreased, an asso­
ciation that was evident for both r a c e s .

Education and in­

come were both significantly and negatively related to
alienation among both blacks and whites, however, the rela­
tionship almost disappeared when occupation was controlled.
Finally, he found that age was not related to alienation and
that men were slightly more alienated than women.
The relationship between alienation and perception
of political strength in a school tax levy election was
81
examined by Eckhardt and Hendershot.

They found that

alienation was associated with social position in the power
structure.

The alienated were largely found among those of

a lower socio-economic status whether measured by occupation
or educational attainment.

Additionally,

the older members

of the sample were more likely to be alienated as were those
who had resided in the community for a longer period of
time.
Bennett and Klecka noted that a strong sense of

41
political efficacy was dependent upon knowledge about poli­
tics and the capacity to use information properly, a result
of the educational process.
elections (1964, 1966,

82

Using data from three national

1968) they found that political effi­

cacy was related to education, income and occupation and that
of the three, educational attainment was the strongest p re­
dictor of efficacy.

Similarly, Vinyard and Sigel found that

social class was related to efficacy with middle class r e ­
spondents generally rating higher in the sense of efficacy
83
than working class respondents.
Mendelsohn and O'Keefe,

in a study of the 1972 elec­

tion in Summit County, Ohio, found that 40% of their respon­
dents rated high on political c y n i c i s m . ^

They note that

"the more detached, alienated and remote the voters were
from the mainstreams of social, economic and political
activities, the more likely they were to express a high de85
gree of political cynicism".

Those respondents who were

apt to score high on political cynicism were "low socio­
economic status voters,

the least affluent, the downwardly

mobile with regard to social class, and voters undergoing
the greatest degree of economic stress".

86

Low socio-economic status (low income, occupation
status and education level), then, is related to alienation
from the political system which is manifested in a wit h ­
drawal from politics

(nonparticipation and non-voting).

Alienation decreases as status increases.

The higher levels

of socio-economic status are associated with participation,

42
voting, trust in government and a high sense of political
efficacy.
Political Participation
The research into the effects of certain political
attitudes on participation is broad, having been undertaken
by

researchers from the fields of psychology,

political science and communication.

sociology,

The one attitude that

has attracted the most attention has been that of alienation.
For example, in the 1959 mayoral campaign in Boston, Levin
found an attitude of alienation in a large number of v o t e r s :
Our analysis of this post-election survey has shown
that a large proportion of the electorate feels polit­
ically powerless because it believes that the commun­
ity is controlled by a small group of politically
powerful and selfish individuals who use public
office for personal gain. Many voters assume that
this power elite is irresponsible and unaffected by
the outcome of elections. Those who embrace this
view feel that voting is meaningless because they
see the candidates as undesireable and the elec­
toral process as a sham. We suggest the term
"political alienation" to refer to these attitudes.®'
The research seems to point to the fact that the
alienated withdraw from political life.

88

Nettler found

this type of behavior in an in-depth study of thirty-seven
"alienates" from across all demographic groups.

The com­

mon ground beneath them was "a consistent maintenance of
unpopular and averse attitudes toward familism, the mass
media and mass taste, current events, popular education,
conventional religion and the teleic view of life, nationalism and the voting process".

90

He discovered that, as a

group, they were politically disenchanted, regarded national

43
voting as a type of "non-rational circus" and showed only
a slight interest in current events.

Levin and Eden paint

an even grimmer picture of the alienated voter as one who is
hostile to politicians, disenchanted with the political
process, tends to believe that campaign promises are "empty
verbiage", and casts votes against rather than for anything
when actually voting, which is rare.

91

Campbell found that

alienation was high among those most withdrawn from party
politics and that this is an active, rather than a passive,
rejection of political matters.

92

Further, their orienta­

tion towards the world of politics is one of suspicion,
distrust and cynicism, viewing the whole political process
as a fraud and a betrayal of public trust.
In support of this withdrawal hypothesis, Kraus and
Davis claim:
Alienated persons are those who cannot understand
social
institutions, and in particular, their own
relationships to those institutions.
Consequently,
they experience frustration and anxiety whenever
they attempt to engage in political action.
These
persons should tend to withdraw from all political
action and develop attitudes which rationalize or
reinforce this w i t h d r a w a l . 93
This withdrawal has been noted in a variety of
studies.

Dean found that alienation was related to inter­

est apathy (a lack of personal involvement in political
interests), influence apathy (lack of interest in influ­
encing others) and behavior apathy, but not related to
voting apathy.

94

Erbe reports a negative relationship b e ­

tween alienation and political participation as does Clark

44
who found the same negative relationship between alienation
and participation in the affairs of an agricultural cooperative.

95

McLeod and Becker, in their study of avoidance,

found that the alienation dimension associated highly with
a lack of interest in the campaign and less attention to
96
political advertising.
The most visible and most basic form of political
participation is voting.

Alienation has been found to influ­

ence behavior in this area in two
voting at all or protest voting.

distinct w a y s :

not

McDill and Ridley found

that the politically alienated were less likely to have
voted but if they voted, were less likely to vote in
favor.

97

They point out:

Political alienation involves not only apathy as a
response to political powerlessness but also a general
distrust of political leaders who are the wielders
of this power.
Consequently, if feelings of polit­
ical alienation find expression in voting behavior,
the vote is likely to be a vote of resentment against
the political powers that be.98
Key, noting that turnout rates do provide a measure,
albeit a rough one, of the extent of political involvement
within the population, points out that non-voting may be
"the product of disillusionment and lack of faith in the
democratic process".

99

Hamilton also examined turnout rates

in a city elections and found that those with a low sense
of political efficacy voted at a much lower rate (27%) than
those with a high sense of political efficacy ( 4 1 % ) . ^ ^
O'Keefe and Mendelsohn, in a 1976 sample of n on­
voters,

found that the most important reasons for non-voting

45
were a distruct towards the candidates, a lack of concern
with politics and a lack of information about the candi­
dates.

Pomper, noting that less than 55% of those 18 and

over voted in 1976, found that there was a widespread
feeling of powerlessness:
An increasing proportion of Americans believe them­
selves unable to control the government, which they
see as dominated by special interests and uncaring
leaders. These feelings are the root cause of no n ­
voting.
Greater turnout will follow not from re­
forms in registration forms, but from actions that
increase the voters' sense of political e f f i c a c y . 1 0 2
Feelings of alienation, then, have an impact on
political behavior.

If alienation is present there appears

to be a withdrawal from politics.
disenchanted,
cynical.

suspicious,

The alienate is hostile,

distrustful, apathetic and

He demonstrates a lack of interest in political

campaigns,

does not pay attention to political advertising

and, if he votes at all, casts protest votes.
Political Attitudes
Political alienation is not a simple, one-dimentional
concept.

Cohen found that alienation combined factors of

distrust of authorities, ethnocentrism, resistance to
change and an orientation toward populism.

103

Kraus and

Davis note an inconsistency in the findings on alienation:
that it has been found to be negatively correlated with
traditional institutionalized forms of political action but
also positively correlated with involvement in non-tradi104
tional social move m e n t s .

They point out that some forms

46
of alienation may increase rather than decrease the likeli­
hood that certain types of political action will occur in
certain election situations.
Any number of researchers have attempted to define
the concept.

Seeman identified five basic ways alienation,

as a term, is used:

powerlessness, which has to do with an

individual's sense of influence over socio-political events;
meaninglessness, or the individual's sense of understanding
the events in which he is engaged; normlessness, or a belief
that socially unapproved behavior may be required to achieve
given goals; isolation, or the detachment of the intellectual
from the popular culture standards; and self-estrangement, or
a loss of intrinsic meaning or pride in w o r k . ^ ^

Dean

posits that alienation can be defined as a combination of
powerlessness, normlessness and social isolation.
Finifter identifies four ways alienation toward the
political system may be expressed:

political powerlessness,

political meaninglessness, perceived political normlessness
and political isolation.

She proposes a provisional

typology of political participation based on the interaction
of political powerlessness and perceived political normlessness.

108

Education was the most powerful predictor of

powerlessness followed by occupation and income.

For norm­

lessness, the predictors were race, then income and educa­
tion.

Males exhibit more normlessness, while females, more

powerlessness.
Olsen defines two categories of alienation:

47
incapability, operationalized as political efficacy, and
discontentment, or political cynicism (distrust of govern109
ment).

Incapability was inversely and moderately strongly

correlated with education, occupation and income; was pre­
valent in those over the age of sixty; and in women more
than men.

Discontentment was also inversely correlated with

education and income, but not as strongly as in the case of
incapability; and was more common among women.

Applying

the scheme to political participation, he found that non­
voting was slightly correlated with political incapability
but not with discontentment.

Overall, attitudes of incapa­

bility towards politics were moderately associated with a
lack of participation in political activities, but that
attitudes of political discontentment had very little effect
on political activities.
Other researchers have noted the relationship between
efficacy and trust in government as two components of politi­
cal alienation.

Thompson and Horton state:

The suggestion is that political inefficacy may
result in political alienation which involves not
only apathy or indifference as a response to aware­
ness of powerlessness, but also diffuse displeasure
at being powerless and mistrust of those who do wield
power.
Given the opportunity for expression, polit­
ical alienation would be expected to be translated
into either an undirected vote of resentment or an
organized vote of opposition.H O
Stokes, using 1958 national election data, found a
correlation between political efficacy and political
t r u s t . H e

hypothesized that under certain conditions one

could find subjective powerlessness linked with a positive

48
attitude toward government, but that generally "feelings
of powerlessness toward public authority tend to create
feelings of hostility toward the authority."

112

Litt pr o ­

poses a close link between cynicism about politics and feel113
ings of ineffectiveness in politics.

He found a rela­

tionship between personal trust and political cynicism.

He

concludes that "a high level of political cynicism may exist
independent of the belief that one may exert some influence
upon these politicians who are regarded as hacks unconcerned
about satisfying any notion of community interest".
'Aberbach uses trust and efficacy as two operational
measures of alienation.

115

He points out that "the funda­

mental reason for defining alienation as a combination of
distrust and powerlessness is the expectation that the joint
occurence of the two attitudes produces a unique behavioral
tendency".

116

Using data from the 1964 national election,

he found that the powerless vote less often than the power­
ful no matter what the level of political trust.

Miller

notes that political cynicism is related to feelings of
political inefficacy and that both have been viewed as com­
ponents of the more general concept of political aliena­
tion.

Accordingly, one would expect that "the relation­

ship between efficacy and political trust would be a criti­
cal determinant of whether one acts politically and the
direction that action takes".

118

In the 1972 election, for

example, 897. of those who felt both efficacious and trusting
turned out to vote whereas only 627, of the inefficacious and

49
distrustful did likewise.
These two dimensions of political alienation,
political efficacy and political trust, have been investi­
gated separately.
government.

Two early studies dealt with trust in

Connelly and Field noted that non-voters seemed

to be more trusting of government.

120

This group did not see

anything particularly wrong with the conduct or caliber of
politicians and were satisfied with the politicians in their
state.

A greater proportion

of non-voters than voters dis­

agreed that politics makes men dishonest, but, the r e ­
searchers note, fewer non-voters had opinions on the issue
at all.

In the other study, Lang and Lang, after an examin­

ation of media use and distrust, report that prior research
indicated that television has a strong appeal to the chroni121
cally distrustful.
Agger, Goldstein and Pearl found that age, income
and education were all related to political cynicism.
Those with less education,
tended to be more cynical.

122

a lower income and who were older
Relating cynicism to political

potency, they report that the sense of political impotence
goes hand in hand with a cynical view of politics and poli­
ticians .

Aberbach and Walker warn that resentment and dis­

trust are elements of dissatisfaction which can lead to
resistance, noting that "the existence of distrustful citi­
zens who are convinced that the government serves the inter­
ests of a few rather than the interests of all is a barrier
to the realization of the democratic ideal".

123

50
Miller makes a similar point in noting the increase
in political cynicism among both blacks and whites since
1966:
. . . widespread discontent prevalent in the United
States today arises, in part, out of dissatisfaction
with the policy alternatives that have been offered
as solutions to contemporary problems.124
Citrin points to the lack of homogeneity of low
scorers on the trust in government scale.

125

They include

'"ritualistic cynics' and partisans of the 'outs' as well as
respondents who see no viable alternative to the incumbent
126
authorities
and reject the constitutional order".
His
date refute the popular proposition that low trust in govern­
ment brings about a withdrawal from conventional political
participation.

On the other hand, Milbrath and Goel report

that negative evaluations of the political system are
associated with low levels of political activity.

127

Two

studies by Mendelsohn and O'Keefe also provide supportive
128
data for the low trust-withdrawal hypothesis.
Political efficacy has been defined a s :
. . . the feeling that individual political action
does have or can have an impact upon the political
process, i.e., that it is worthwhile to perform
one's civic duties.
It is the feeling that social
change is possible and that the individual citizen
can play a part in bringing about this change.
Its negative component includes a sense of futility,
insignificance, powerlessness and incompetence.

130

This

sense of inefficaciousness has been found to be a determ­
inant of lack of participation in the political system.
early as 1951, Rosenberg identified powerlessness

As

51
(inefficacy) as one of the factors fostering apathy and
inactivity:
The giant economic and power blocs typified by giant
corporations and unions, thrust the individual about
with pressures too powerful to resist.
As a conse­
quence the individual is likely to feel overwhelmed
and powerless.
Given this feeling, the idea that his
puny strength can match the giants is absurd, and
he feels that a lonely individual can do nothing to
change the way society is run.
Raising his weak
voice against the massive roar of the mass media
and political giants is futile.
For this reason
many people with political conviction do no more
than vote, convinced that they can have no substan­
tial effect in any event,
In a later work delineating the determinants of political
apathy, Rosenberg identified a sense of the futility of
political activity.

This was based on a sense of personal

inadequacy and unmanageability of political forces--a
classical example of the sense of political inefficacy.
Seeman defines alienation as powerlessness,

132

the

"expectancy or probability held by the individual that his
own behavior cannot determine the occurence of the outcome he seeks".

133

Tying powerlessness to knolwedge

acquisition, he found that those high in powerlessness were
less interested in political activities and had a lower
level of knowledge acquisition.

134

Wolfinger and his

colleagues, in an examination of the radical right, found
that the alienated do not necessarily become social iso135
lates.

The alienated group they term "crusaders" are not

possessed by feelings of powerlessness, but rather have a
strong sense of political efficacy and a high level of p o ­
litical participation in political campaigns.

Iyengar notes

52
that the sense of subjective political efficacy does not
appear to be closely intertwined with evaluations of the
incumbent government.

136

It is not "a fleeting response to

current political realities but is, instead, a more firmly
embedded attitude concerning the responsiveness of the
• „ 137
regime .

Milbrath and Goel, citing numerous studies, report
that those who feel efficacious are more likely to parti­
cipate in the political process, and that higher levels of
efficacy are associated with those of upper socio-economic
status.

138

Eulau and Schneider found that efficacy alone

was not a sufficient condition for involvement.
of responsibility was also necessary.

139

A sense

They determined that

an individual may be a regular voter even if low on efficacy,
if that person rates high on civic duty or responsibility.
Persons who feel a duty to participate in politics
are much more likely to do s o .

Demographically, those

who are in the upper socio-economic strata, especially those
with high education levels, are more likely to develop this
sense of citizen duty.

Nie, Powell and Prewitt found that

the direct impact of socio-economic characteristics on
political participation, including voting, was weakened by
the presence of five attitude sets toward the political
system.These

included a sense of civic duty, political

efficacy, political knowledge, perceived importance of polit­
ical outcomes, and attentiveness to mass media political
content.

53
Feelings of alienation,

distrust and inefficacy have

been found to influence political behavior, as well as
attitudes and opinions on political issues and problems.

142

Horton and Thompson propose that:
. . . voting down local issues may be in part a
type of mass protest, a convergence of the individual
assessments and actions of the powerless who have
turned politics into a "phobic" sector by projecting
into available political symbols the fears and
suspicions growing out of the alienated conditions of
their existence.1^+3
Erbe found that alienation can be expressed in two
distinct w a y s .

Retreatist alienation leads to disen­

chantment , despair and the feeling that the world promises
nothing in the way of comfort or support.
ation, on the other hand,

Rebellious alien­

leads to a "desire to strike back

at or to somehow violently manipulate the world from which
the individual has become estranged," the kind of "energy
that activates the 'true believer'".

145

Templeton deter­

mined that "alienation is reflected in hostility toward the
agencies of government and is not significant in the formu­
lation of policy judgements''.^^
Finifter suggested that the varying levels of
powerlessness combined with varying levels of perceived
political normlessness would be associated with different
types of political p a r t i c i p a t i o n . T h o s e high on both
would be characterized by a complete withdrawal from polit­
ical participation, while those low on both would conform to
the standard modes of participation.

Those high on norm­

lessness and low on powerlessness would have a reform

54
orientation, working to change the system from within,
while those high on powerlessness and low on normlessness
would exhibit apathy and very little political activity.

In

confirmation, Paige reports that mistrust of local govern­
ment, when combined with a strong sense of political
efficacy, fostered participation in riot activity in Newark
in 1967, while Gamson points out that "a combination of a
high sense of political efficacy and low political trust is
the optimum combination for mobilization - a belief that
influence is both possible and necessary".
Seeman notes that normlessness threatens the develop­
ment of trust and that powerlessness (inefficacy) is related
to protest voting.

He concludes:

Thus, if the studies of voting in local referenda
and of ghetto violence (and possibly campaign
activity as well) tie activism to powerlessness,
while powerlessness is also associated with n on­
participation, retreat from knowledge and unwill­
ingness to act, then clearly some distinctions
have to be made concerning kinds of knowledge, and
the like.1*9
Milbrath and Goel propose a relationship between
efficacy, trust and participation:
The low efficacy-low trust combination likely leads
to withdrawal from participation, for why get in­
volved when o n e 's actions have such a low probability
of influence.
Those who have both efficacious and
trustful orientations are likely to choose conven­
tional modes of participation because they are per­
ceived as producing the desired outcomes.
In con­
trast, the efficacious mistrustful persons may be
inclined toward radical action because such action
is perceived as both necessary and possible.
Those
who trust political leadership but doubt their own
capability for political influence most likely
engage in supportive and ritualistic activities
(voting, flying the flag, enlisting, and so on).1^0

55
A number of authors support the contention that
alienation is composed of trust in government and political
efficacy.

Each dimension affects political behavior.

For

example, distrust leads to discontent, low participation and
non-voting.
low interest,

A low sense of political efficacy is related to
low level of knowledge acquisition,

of participation and non-voting.
bring about unique behaviors.

low levels

Together they interact to

Low trust and high efficacy

leads to increased participation but of a rebellious or pro­
test sort while low efficacy and high trust brings about
participation limited almost exclusively to voting.

Persons

high on both are participants while persons low on both
tend to withdraw totally.
Summary
Alienation's impact on political behavior is clear.
It is most closely associated with groups of lower socio­
economic status.

When present it can lead to a withdrawal

from politics as seen in a lack of interest,
ledge,

lack of know­

lack of participation and lack of voting.

When de­

fined as the interaction of trust in government and political
efficacy, allowance can be made for the fact that in some
instances alienation leads to protest voting.
School Referenda
The literature regarding school referenda, as con­
cerns the present study, falls into three general areas of
concern:

Awareness/knowledge, source of information and

56
voting correlates, including both demographics and
alienation.
Awareness/Knowledge
Research on awareness levels in school referenda
shows mixed results at best.

Hamilton and Cohen cite three

studies where awareness levels were determined.

151

A study

conducted in Bowling Green, Ohio, the week after the election
which asked voters the purpose of the four fiscal propo­
sitions on the ballot found that levels of correct responses
were 95% for two propositions,
fourth.

80% for one and 60% for the

Eighty percent of the voters knew of the proposed

cutbacks if the school levy failed and 70% were aware that
teachers' salaries were below the regional average.

In

Austintown, only 40%, had heard the school-closing warning
and 33%. knew teachers' salaries were lower than those in the
surrounding communities.

A survey of a bond election in

DeKalb County, Georgia, found few voters who had acquired in­
formation on the ten measures on the ballot and few who were
aware of the activities and positions of civic and profes­
sional groups.
Awareness and knowledge seem to be related to p a r ­
ticipation.

Kowalski reports a positive relationship b e ­

tween knowing the names of neighborhood principals and
152
school board members and voting.
Carter found that the
extent of knowledge about the school was slight but that the
more the knowledge, the more participatory the voter.

153

57
A study undertaken by the Washington State Legislature hypo­
thesized that knowledge of school elections would be dis­
tinctly greater among yes v o t e r s .
eighteen communities,

After surveying

they report that "don't know" re­

sponses on a battery of factual items was two to three times
greater among no v o t e r s .
School referenda,

then, are generally characterized

by low awareness levels and low levels of knowledge.

Aware­

ness and knowledge are related to voting and to participation
such that as awareness and knowledge increase, so do parti­
cipation and voting.

Knowledge is also related to vote cast

with correct information characterizing yes voters and
little or no information, no voters.
Source of Information
An early campaign manual lists twenty-six types of
media and materials a school entity can utilize in a school
election including open letters from the board of education
to the electorate, endorsements of the proposals, a manual
of information, meetings, radio, material in newspapers,
material in private house organs, material in school publi­
cations, personal letters, posters, billboards, tags,
stickers, and school reports.

155

While all of these outlets

are used to communicate with the electorate during a cam­
paign, research to date shows that newspapers are the pre­
dominant source of information regarding school issues,
including school referenda.

Carter found that while

58
two-thirds of the voters said they would turn to a school
official for information about the schools, when queried
about actual channels of information the majority of the
respondents identified newspapers as their primary
source.

156

Hamilton and Cohen confirmed this finding in

their Bowling Green study in the m i d - 1 9 6 0 1s .
Rowe came to a similar conclusion in a study of
voters in a bond election in San Diego, California.

158

He

also found that newspapers were not viewed as part of the
school district's public information campaign.

This may

partially explain Rarick's findings that 42.27o of the
respondents said they would take the advice of the news­
paper in a local bond election compared to only 32.8% in
Presidential elections.

159

Krischak, in a study of an

unsuccessful rural Michigan community college millage elec­
tion, found that more voters relied on the newspapers than
any other source of information; that almost all voters were
exposed to information during the election campaign; that
radio did not appear to affect voter decision; and, that
there was an indication that voters who got their informa­
tion from a secondary source were more inclined to vote
i
160
against the millage request.
Greenberg found that in a school bond election, those
who thought their side would win exposed themselves to more
sources of information than those who thought their side
would lose and that those who held consistent cognitions
relied more heavily on print sources such as bulletins and

59
leaflets.

161

Generally, however, personal communication

with friends, students and school employees follows news­
papers in usefulness while pamphlets, leaflets and school
-I£0
bulletins rank third.
The pre-eminence of newspapers as
information sources is the product of three givens in
school elections:
Most voters are not parents of school children,
most people do not attend meetings where the school
proposal is discussed, and the press is inclined to
be generous in reporting school news and publishing
campaign press releases.
Sigel points out, however, that the importance of
the press relative to other communication channels is the
function of the size of the school district with more
reliance placed on the press in larger districts while in
smaller districts the emphasis shifts to a doorbell campaign utilizing

164
leaflets and pamphlets.

Newspapers, then, are the dominant source of infor­
mation regarding school referenda.

Personal communication

and other printed materials follow in importance with radio
a distant fourth.

The research also indicates that voters

who get their information from a secondary source, gen­
erally in the form of personal communication, are more in­
clined to vote against the issue.
Voting Correlates:

Demographics

Certain demographic variables have been found to be
closely related to positive voting in school referenda.
Carter found that the most favorable voter was young (under

60
fifty-one) with children in school, a recent resident of
the community with an occupation as a skilled worker or in
clerical and sales, while the most unfavorable voter was
also young but without children in school, a long-time
resident of the community with a professional or technical
165
occupation.

Hamilton and Cohen found that the highly educated
are more supportive, that endorement of levies by voters in
the top income bracket was twice the rate of those in the
bottom bracket and that professionals were distinctly more
favorably inclined than businessmen and far more supportive
than those in manual occupations.

They also found an

inverse relationship between age and support with support
decreasing as age increased.

They also report that parents

were more supportive as were women, that renters voted less
often but not differently than own e r s , and that religious
affiliation was a situational variable coming into play only
when the percentage of parochial school students was high.
Kowalski reports similar results with some refine1 67
ment.

able:
farmer.

Occupations were ranked from most to least favor­

professional, managerial, white collar, manual,
As for parental status, those with children in the

public schools, specially grades ten and under, were more
positive than non-parents or those with no children in school
either because they have not entered yet or have already
completed their schooling.
The relationship between income and voting was

61
further examined by Wilson and Banfield who point out that
middle income homeowners often vote against proposals
approved by the very poor (renters) and the very well-to-do
168
(owners).

Alexander and Bass found that the highest in­

come class and managerial and professional occupations are
associated with greater election support.

169

Kowalski also examined the relationship between past
and present voting and found that a person's past voting re­
cord was a prime determinant of a future v o t e .

She

states that the chances are three out of four that a person
who previously voted in favor of a school proposal will vote
positively the next time.
Support for school referenda is generally found among
those with children in school, those with higher levels of
education and income, those with very low incomes (renters),
those with high occupation levels and women.

Those who have

voted for a referendum in the past are also most likely to
be supportive.
Voting Correlates:

Alienation

Alienation has also played a part in school monetary
elections.

In a study of school bond issues, Thompson and

Horton found that those who felt powerless and isolated in
community affairs voted in lower numbers, but these people
were also more likely to cast protest votes, stating that a
no vote on a bond issue was a vote against the existing
power structure.

171

Piele and Hall point out that the

62
alienated group,

individuals who are extremely distant from

the center of the community, generally avoid participation
in the decision-making process (i.e., voting) but occa­
sionally, when certain political situations arise, like a
school referendum, will register a protest against school
government by voting negatively.

172

Reeder also points out

that a lack of confidence in the board of education, in the
superintendent of schools or in some phase of the management
of the school, constitutes one of the major reasons for a
no vote.

173

Templeton found that on a series of items de­

signed to test the respondent's knowledge of the propo­
sitions to be included on the ballot, "alienated respondents
were consistently less likely to give correct responses than
nonalienated respondents".

This finding is substan­

tiated by Horton and Thompson in their research into the
relationship between powerlessness and local referenda
«-• •- 175
activity.
Eckhardt and Hendershot, in an examination of the
relationship between alienation and the perception of polit­
ical strength in a school tax levy election, found that the
alienated were worse assessors of political opinion at both
the precinct and the community l e v e l . T h e

alienated,

they report:
. . . are more likely to overestimate their polit­
ical strength in the community in comparison to the
unalienated, and this perception of political
strength may in turn induce a feeling of "power­
fulness" among the alienated which accounts for
their political participation at the community
level.i77

63
Hamilton and Cohen found that no voting was an ex­
pression of general distrust on the part of the politically
alienated.

178

In studies undertaken in Corning and Ithaca,

New York, there was striking evidence of alienation and
negative voting, a relationship which remained even when
socio-economic status was controlled.

In their Austintown

study, one-fifth of the actual voters were high in alien­
ation.

Four-fifths of those alienated voters voted no com­

pared with only one-fifth of those scoring low in alien­
ation.

Alienation was one of the underlying factors found

in Youngstown, Ohio using a Q sort factor methodology.

This

factor contained "nearly all the elements of alienation:
parochialism, ethnocentrism, local chauvinism, aggressive­
ness, distrust of authority, suspiciousness, and a persei .. .179
cution complex

Carter investigated the sense of efficacy, a feeling
that the voter's participation would be meaningful, and one
of the dimensions of political alienation under study.

180

He found that one-third of the voters felt that they had
little to say about what the schools did; almost one-half
felt that the only voice voters had was the act of voting;
that one-fourth felt that school officials did not care what
the average voter thought;

and, that two-fifths thought that

educational policy was too complicated for persons like them­
selves to understand.

He concluded that the presence of low

efficacy almost assures a negative view of the schools.

Wirt

and Kirst found that the efficacious voter was younger, more

64
educated and felt the pressure to be involved in school
policies.

181

Giles, Gatlin and Cataldo determined that

trust and efficacy were positively related to support for
u i taxes.
182
school
Hamilton and Cohen propose that trust in government
and efficacy be examined as two components of political
alienation that may affect the outcome of school referenda;
Perhaps both popularity and vagueness stem from
the proliferation of dimensions that have been added
to the phenomena:
thus it may be popular because
alienation can be used to explain so much in the
way of behavior and yet it may be vague because the
domain has become unmanageable as a single concept.
To the perceived lack of power in community affairs
and distrust of those who hold power, there has
been added interpersonal distrust (Do you trust
people?) and personal inefficacy (Do you feel pe r ­
sonally ineffective?). The additional sources of
alienation seem to be especially significant for
referenda, where the so-called alienated voter can
be politically efficacious, and who was for six
consecutive elections in Y o u n g s t o w n . -*-83
The alienated, then, are less likely to vote but
more likely to cast a protest vote in school elections.
no vote is seen as an expression of general distrust.

The
The

alienated are also less likely to have correct information
about school referenda.

Low efficacy is associated with a

negative view of the schools while high levels of trust and
efficacy are positively related to support for school taxes.
Summary
School referenda are low awareness, low knowledge,
low participation elections.

They are characterized by the

participation of alienates protesting the current system

65
authorities.

Support is generally found among those most

closely associated with education, either because of school
age children or because of high levels of educational
attainment.
Hypotheses
This study was designed primarily to examine the
relationships proposed in the derivative model.

The model

posits differences in level of awareness and correctness
and source of information based on varying levels of trust
in government and political efficacy.
supports this typology.

The literature

Those persons high on trust and

efficacy are generally found among those with higher levels
of socio-economic s t a t u s . B o t h high levels of trust and
efficacy and higher socio-economic status are related to the
185
use of newspapers as a source of political information,
186
and to yes voting in school referenda.
Support for such

issues, finally, is related to level of correct information.

187

Those scoring high on both trust in government

and political efficacy should be aware of the election, hold
correct information regarding the issue and receive the
majority of their information from the media, in this case,
newspapers.
A strong sense of political efficacy and a distrust
of government has often been associated with protest voting
in referenda.
formation

189

188

No voting is related to having little in-

and with getting information from secondary

66

sources.

190

Low trust is associated with a reliance on

television as a source of political information.

191

In

this case television is not available as a source of polit­
ical information.
highly aware as
but to

Therefore this group is expected to be
a result of strong feelings of efficacy,

192

have little or no information as a result of no

local television.

Information, then, should come from p e r ­

sonal sources.
Those scoring low on efficacy but high on trust have
been found to be supportive of the system.
is related to apathy and inactivity

194

193

Low efficacy

but this is tempered

by high trust which is related to voting.

195

Those high on

trust are more likely to use media as a source of information,
voting.

196
197

while newspaper readership is related to
Given these findings, those low on efficacy and

high on trust should have little awareness of the election
but those who are aware should have correct information
and should acquire that information from the media, in the
present instance, newspapers.
The final group, those low on both trust and effi­
cacy, is the easiest to type.
from all political activity.

198

This group should withdraw
The alienated have been

found to avoid political information,
media generally,
newspapers.

201

199

to distrust the

and to utilize television rather than

Those low in trust and low in efficacy

should, therefore, have little awareness of the campaign,
have little or no information about the issue and to have

67
gained that information from personal sources.
Three major hypotheses and six sub-hypotheses were
used to test these differences in the groups as proposed in
the derivative model.

The first set of hypotheses dealt

with the level of awareness:
:

There will be significantly different levels of
awareness among the four groups.
Hx:

(A)GWA)GII^(A)GIII^(A)GIV

Hq :

(A)GI=(A)GII=(A)GIII=(A)GIV202

Where (A) equals the number aware and GI to GIV
203
the groups.

:

The proportion of those aware in Group I will be
significantly greater than the proportion aware
in Group I I .
Hla:

(PA)GI>(PA)GII

Hq :

(PA) GIjfc-(PA) GII

Where (PA) equals the proportion aware.
H lb:

Hi

:

T^ e ProP ortion °f those aware in Group III will
be significantly greater than the proportion
aware in Group IV.
H lb:

(PA)GIII>(PA)GIV

Hq :

(PA)GIIIi(PA)GIV

The proportion of those aware in Groups I and II
combined will be significantly greater than the
proportion aware in Groups III and IV combined.
R lc:

(PA)GI+II>(PA)GIII+IV

Hq :

(PA)GI+IIi(PA)GIII+IV

The second set of hypotheses examined the differ­
ences in correctness of information:
H^:

There will be significantly differential levels of
correct information among the four groups.

H2:

MGI^MGII^MGIII^MGIV

Hq :

MGI=MGII=MGIII=MGIV

Where M equals the mean level of correct infor­
mation held by the group.
H 2a :

H2b:

H 2c ’
•

The mean level of correct information of Group I
will be significantly greater than the mean level
of correct information of Group I I I .
H 2a:

MGI>MGIII

Hq:

MGLsMGIII

'r^ie mean level °f correct information of Group II
will be significantly greater than the mean level
of correct information in Group IV.
H2b:

MGII>MGIV

Hq :

MGII^MGIV

The average of the mean levels of correct infor­
mation of Groups I and II will be significantly
greater than the average of the mean levels of
correct information of Groups III and IV.
H 2c:

MGI+MGII>MGIII+MGIV

Hq :

MG I4MGI LiMG 11 I+MG IV

The third major hypothesis focused on source of in­
formation:
Ho:

Groups I and III will report media sources in a
significantly greater proportion than Groups
II and IV.
H3:

(PM)GI+III>(PM)GII+IV

Hq:

(PM)GI+IIIi(PM)GII+IV

Where (PM) equals the proportion in the combined
groups reporting relying on newspapers and radio
as their sources of information.
In addition to testing the derivative model,

69
hypotheses were tested dealing with the major areas dis­
cussed in the review of literature.

The fourth hypothesis,

then, examined awareness levels in the sample throughout
the campaign:
H^:

There will be significant increases in the pr o ­
portion aware of the election as the campaign
progresses.
H/ :
*

(PA)W1 <(PA)W2< (PA)W3<(PA)W4< (PA)W5< (PA)W6 <
(PA)W7<(PA)W8<(PA)W9<(PA)W10

Hn :

(PA) WL* (PA) W2 ^ (PA) W3 ^(PA) W4^ (PA) W5 * (PA) W 6 £(PA)W7^(PA)W8*.(PA)W9^(PA)W10

Where (PA)Wn equals the proportion of the sample
called that week that was aware that a special
election had been called.
Hypothesis five dealt with correctness of infor­
mation:
The proportion of those holding correct informa­
tion about the issue will increase significantly
across the campaign.
He:
3

(PC)Wl<(PC)W2<(PC)W3<(PC)W4<(PC)W5<(PC)W6 <
(PC)W7 < (PC)W8 <(PC)W9 < (PC)WIO

Hn :
u

(PC)WU(PC)W2^(PC)W3^(PC)W4*(PC)W5^-(PC)W6-*(PC)W7-(PC)W8-(PC)W9— (PC)WIO

Where (PC)Wn equals the proportion of the sample
contacted during the week holding correct infor­
mation about the election.
Hypothesis six examined source of information:
Hg:

A significantly higher proportion of the respon­
dents in Owosso-Corunna will receive their infor­
mation from the newspaper than respondents in the
outlying districts.
H6:

(P)NPO-C>(P)NPOTHERS

70
H0 :

(P)NPO-Ci(P)NPOTHERS

Where (P)NP equals the proportion of the respon­
dents who reported receiving their information
about the election from the newspaper; 0-C repre­
sents the combined school districts of Owosso and
Corunna; and OTHERS, the remaining six school
districts.
Hypotheses seven through twelve examined various
factors influencing favorability to the issue of an area
vocational education center.
Hy:

Favorability toward the issue will decrease sig­
nificantly as age increases.

H7 ’

<1>x age',< « X £.vor

V

a > X age=<I)XfaVor

Where (I)Xn represents an increase in the mean of
of the variable indicated.
Hg:

Parents with children at home will be significantly
more favorable towards the issue than those without
children at h o m e .
Hg:

X l >X l I

Hq :

X i * X][][

Where X represents the mean level of favorability,
I represents the group with children at home and
II, the group of parents without children at home.
Hg:

Favorability will increase significantly as edu­
cation level increases.

Hg:

X^<X2 <'X2<'X^<X^^ Xg <'Xy<'Xg

Hq :

Xl>X2^X3iX4>X5>X6>Xy^Xg

71
Where X represents the mean favorability score
na /

for the subscripted level of education.
H-^q :

Favorability will increase as income level in­
creases .
H 1 0 :

Hq :

X 1 < X 2 < X 3 < X 4

X 12X 2^ X 3>X4

Where Xn represents the mean level of favorability
for the subscripted income level.
H-,:

205

There will be significant increases in favora­
bility as occupational status increases.
H X1:

X 1<X2<X3<X4<X5

Hq:

X 1>X2>X3^X4^X5

Where Xn represents the mean level of favorability for each occupation status level.
Hn2

206

Those reporting a yes vote in the past will be sig­
nificantly more favorable than those reporting
a
past negative vote in a school millage election.
h

12:

H

q

:

X I>XII
x i * x n

Where X represents the mean favorability level; I,
the group voting yes in the past; and II, the
group voting n o .
Hypotheses thirteen and fourteen examined the influ­
ence of alienation on favorability and negative voting.
H n3 :

Alienation will be a significant independent pre­
dictor of favorability.
H^3 :

F=a+b-^X^

Hq:

F?t a + b 1X 1

Where F represents level of favorability and

72
X-p the level of alienation.
H^:

Alienation will be a significant independent pre­
dictor of negative voting.
H 14:

V=a+b1X 1

H0 :

Vfa+b-^

Where V represents the self-report of the person's
vote on the issue after the election and X^, the
level of alienation.
Definitions
"Awareness" is defined as knowing, either aided or
unaided, that a speical county-wide election has been
called.

Level of "correct information" is defined as the

number of correct responses to a series of questions con­
cerning who had called the election, what the election was
about, and the amount of millage requested.

"Source of in­

formation" is defined as newspaper, radio, a friend, a flyer
or brochure, a poster, and a meeting.

"Favorability" is

defined as a self-report of the respondent's likelihood of
voting in favor of the millage with 0 meaning no chance and
100 meaning definitely in favor.

"Alienation" is defined as

a combination of the scores on the trust in government and
sense of political efficacy scales such that the higher the
score the greater the sense of alienation.
Limitations
The model is being tested in a unique situation:
school millage election in a minimal local media county.

a

73
The results may not be generalizable to a partisan election,
or a media-rich area or both.
The data collected were an added feature to a larger
project.

The researcher therefore had limited control over

the other portions of the study.

The length of the inter­

view, inexperience of the interviewers and difficulty of the
main instrument may have biased the results in ways that
would not have occured in a different research setting.

Endnotes
Shirley A. Star and Helen MacGill Hughes, "Report
on an Educational Campaign:
The Cincinnati Plan for the
United Nations," American Journal of Sociology 55 (January
1950):389-401.

2
John E. Ross and Lloyd R. Bostain, "Communication
Activities of Wisconsin Farm Families in Wintertime,"
Journal of Broadcasting 2 (Fall 1958) :319-28.
3

Bruce H. Westley and Werner Severin, "A Profile
of the Daily Newspaper Non-Reader," Journalism Quarterly
41 (Summer 1964):156.
4

Obviously some cross over in subject matter takes
place in these studies.
The entire results of the study, u n ­
less otherwise noted, will be reported in that section
dealing with the primary topic of the article.
^Bruc'e H. Westley and Joseph B. M o b i u s , "A Closer
Look at the Non-Television Household," Journal of Broad­
casting 4 (Spring 1960):164-73.
^Merrill Samuelson, Richard F. Carter and Lee
Ruggels, "Education, Available Time and Use of the Hass
Media," Journalism Quarterly 40 (Autumn 1 9 6 3 ) :496.
^Bruce H. Westley and Werner S. Severin, "Some
Correlates of Media Credibility," Journalism Quarterly 41
(Summer 1964):325-35.
8I b i d ., p. 327.
Q
Bradley S. Greenberg, "Media Use and Believability:
Some Multiple Correlates," Journalism Quarterly 43 (Winter
1966) -.665-70, 732.
^ L e s l i e W. Sargent and Guido Stemple III, "Poverty,
Alienation and Mass Media Use," Journalism Quarterly 45
(Summer 1968):324-26.
■^Alienation results are reported elsewhere.

12

Serena Wade and Wilbur Schramm, "The Mass Media as
Sources of Public Affairs, Science and Health Knowledge,"
Public Opinion Quarterly 32 (Summer 1969):197-209.
74

75
13

F. Gerald Kline, "Media Time Budgeting as a Func­
tion of Demographics," Journalism Quarterly 48 (Summer
-------1971):211-21.
■^Peter Clarke and Lee Ruggels, "Preferences Among
News Media Coverage of Public Affairs," Journalism
Quarterly 47 (Autumn 1970):464-71.
15

Lee B. Becker, Idowu A. Sobowale and William E.
Casey, Jr., "Newspaper and Television Dependencies:
Their
Effect on Evaluation of Government Leaders," paper presented
to the International Communication Association, Chicago,
1978.
■^Dale Vinyard and Roberta S . Sigel, "Newspapers and
Urban Voters," Journalism Quarterly 48 (Autumn 1971):486-93.
^ G e o r g e A. Barnett and Carol E. Hughes, "The Effects
of Media Exposure on Uncertainty Reduction and Voting:
A
Preliminary Analysis," paper presented to the International
Communication Association, Chicago, 1978, p. 3.
18

Paula M. Poindexter, "Non-News Viewers," Journal of
Communication 30 (Autumn 1980):58-65.
‘IQ

Paul Burstein, "Social Status and Individual Polit­
ical Participation in Five Countries," American Journal of
Sociology 72 (May 1972):1096.

20

Paul F. Lazarsfeld, Bernard Berelson and Hazel
Gaudet, The People's Choice (New York: Columbia University
Press, 1964).

21

Bernard R. Berelson, Paul F. Lazarsfeld and William
N. McPhee, Voting (Chicago: The University of Chicago Press,
1954).

22

Both of these studies were undertaken prior to the
widespread penetration and usage of television by the elec­
torate and candidates.
A replication of either study today
might elicit different results.
However, since Shiawassee
County is devoid of any local television station, the situa­
tion in the present instance may more closely approximate
that of the Elmira and Erie County studies than more recent
studies where television plays a more integral part.
^ W i l l i a m A. Glaser, "Television and Voting

76
Turnout," Public Opinion Quarterly 29 (Spring 1965):71-86.
24

Jerome D. Becker and Ivan L. Preston, "Media Use
and Political Activity," Journalism Quarterly 46 (Spring
1969):129-34.
------------ -------25

Robert H. Prisuta, "Mass Media Exposure and Polit­
ical
Behavior,"
Educational Broadcast Review 7 (June 1973):
167_73
-------------------------26I b i d ., p. 172.
2^Steven H. Chaffee and Jack M. McLeod, "Individual
vs. Social Predictors of Information Seeking," Journalism
Quarterly 50 (Summer 1973):237-45.
23Garrett J. O'Keefe and Harold Mendelsohn, "Non­
voting: The Media's Role" in Charles Winick, e d . , Deviance
and Mass Media (Beverly Hills, Calif.: Sage Publications.

vmr.--29

Edward C. Dreyer,."Media Use and Electoral Choices;
Some Political Consequences of Information Exposure," Public
Opinion Quarterly 35 (Winter 1971-72):544-53.
30Ib i d ., p. 548.
31

Herbert E. Alexander, "Communications and Politics;
The Media and the Message," Law and Contemporary Problems
34 (Spring 1969):255-77.
32

Barnett and Hug h e s , "The Effects of Media Exposure
on Uncertainty"
33

Ralph 0. Nafziger, Warren C. Engstron and Malcolm
S. MacLean, Jr., "The Mass Media and an Informed Public,"
Public Opinion Quarterly 15 (Spring 1951):105-14.
Q/

Garrett J. O'Keefe, Harold Mendelsohn and Jenny
L u i , "Voter Decision Making: 1972 and 1974," Public Opinion
Quarterly 40 (Fall 1976):320-30.
35

It must be noted that the 1972 election was a
Presidential election and 1974, only a Congressional elec­
tion.
The switch from television to newspapers may be the
result of this fact rather than a massive overall shift from
television to newspapers as the primary source of all

77
public affairs information.
36

William Buchanan, "An Inquiry into Purposive
Voting," Journal of Politics 18 (May 1956):281-96.
37

M. Margaret Conway, "Voter Information in a No n ­
partisan Local Election," Western Political Quarterly 21
(March 1968):69-77.
38

The study overrepresented persons less than twentyseven years old and focused on that group.
Jack M. McLeod,
Carl R. Bybee, William D. Luetscher and Gina Garramore, "Mass
Communication and Voter Volatility," paper presented to the
Association of Educators in Journalism, Seattle, 1978.
39

Rebecca Colwell Quarles, "Mass Media Use and
Voting Behavior:
The Accuracy of Political Perceptions Among
First Time and Experienced Voters," Communication Research
6 (October 1969):407-36.
40

Sidney Kraus and Dennis D a v i s , The Effects of
Mass Communication on Political Behavior (University Park:
The Pennsylvania State University Press, 1976).
^ H e r b e r t Blumler, "Suggestions for the Study of
Mass Media Effects," in E. Burdick and A.J. Brodbeck, e d s .,
American Voting Behavior (Glencoe, 111.: The Free Press,

1959), pp. 197-208.----4^See also, Becker and Preston, "Media Use"; Prisuta,
"Mass Media Exposure"; Chaffee and McLeod, "Individual vs.
Social Predictors"; and, O'Keefe and Mendelsohn, "Non­
voting" .
/Q
Clarke and Rugg e l s , "Preferences Among News
Coverage".
44Charles K. Atkin, John Galloway and Oguz B.
Nayman, "News Media Exposure, Political Knowledge and Cam­
paign Interest," Journalism Quarterly 53 (Summer 1976):
231-37.
4"*M.E. McCombs and L.E. Mullins, "Consequences of
Education, Media Exposure, Political Interst and Infor­
mation-Seeking Orientations," Mass Communications Review
1 (August 1973):27-31.
46Ibid., p. 29.

78
4 ^Viekko Pietila, "Alienation and the Use of Mass
Media," ACTA Sociologicia 13-14 (1970):238.
48Ibid., p. 241.
49

Jack McLeod, Scott Ward and Karen Tancil, "Alien­
ation and Uses of the Mass Media," Public Opinion Quarterly
29 (Winter 1965):584-94.
^^Elihu Katz and David Foulkes, "On the Use of the
Mass Media as 'Escape': Clarification of a Concept,"
Public Opinion Quarterly 26 (Fall 1962):377-88.
■^This view is supported by van den Haag who sees
the mass media as "alienating people from personal exper­
iences and, though appearing to offset it, intensifies their
moral isolation from each other, from reality and from them­
selves". E. van den Haag, "Of Happiness and of Despair We
Have No Measure," in B. Rosenberg and D.M. White, e d s .,
Mass Culture (Glencoe, 111.: The Free Press, 1957), p. 529.
52

The nine reasons were: 1) Because my mind is al­
ready made up; 2) Because you can't always trust what poli­
ticians tell you on television; 3) Because I'm not much
interested in politics; 4) Because they hardly ever have any­
thing new to say; 5) Because I prefer to relax when watching
TV; 6) Because some speakers talk over one's head; 7) Be­
cause some speakers talk down to the audience; 8) Because
I dislike being "got at" by politicians; and, 9) Because pol­
itics should not intrude into the home and family affairs.
Jay G. Blumler and Denis McQuail, Television in Politics:
Its Use and Influence (London:
Faber and Faber LTD, 1968),
jT 56.
53

The items related to these dimensions were:
Parti­
sanship--!) Because I'm not much interested in politics; and,
2) Because my mind is already made up.
Relaxation--1) B e ­
cause I prefer to relax when watching television.
Alien­
ation--!) Because you can't always trust what politicians
tell you on television; 2) Because some candidates talk down
to the audience; 3) Because some candidates talk over one's
head; and, 4) Because they hardly ever have anything to say.
Jack M. McLeod and Lee B. Becker, "Testing the Validity of
Gratification Measures Through Political Effects Analysis,"
in Jay G. Blumler and Elihu Katz, eds., The Use of Mass
Communications: Current Perspectives on~Gratifications
Research (Beverly Hills, Calif.: Sage Publications, .1974),
p p . 137-64.

79
54W illiam Semlac and Wenmouth Williams, Jr., "Uses,
Gratifications and Avoidance: Voting Decisions in the 1976
Presidential Debates," paper presented to the International
Communication Association, Chicago, 1978.
■^Samuel J. Eldersveld, "Experimental Propaganda
Techniques and Voting Behavior," American Political Science
Review 50 (March 1956):154-65.
^^Heinz Eulau and Peter Schneider, "Dimensions in
Political Involvement," Public Opinion Quarterly 20 (Spring
---------------1956)=128-42.
■^Alexis S. Tan, "Political Participation, Diffuse
Support and Perceptions of Political Efficacy as Predictors
of Mass Media Use," Communication Monographs 48 (June 1981):
133-45.
58

L.B. Becker, J.M. McLeod and D. Ziemke, "Corre­
lates of Media Gratifications," paper presented to the Amer­
ican Association for Public Opinion Research, Ashville, N . C . ,
1976, cited in Kim A. Smith and Warren Bechtolt, Jr., "Ado­
lescent's Use of the Content of the 1976 Presidential De­
bates," paper presented to the International Communication
Association, Chicago, 1978.
"^Garrett J. O'Keefe, "Political Malaise and Reliance
on Media," Journalism Quarterly 57 (Spring, 1980):122-28.
^ K u r t Lang and Gladys Engel Lang, "The Mass Media
and Voting," in Burdick and Brodbeck, American Voting
Behavior, p. 231.
^ M i c h a e l J. Robinson, "American Political Legit­
imacy in an Era of Electronic Journalism: Reflections on
the Evening News," in Douglass Cater and Richard Adler,
e d s ., Television as a Social Force: New Approaches to TV
Criticism (New York: Praeger Publishers, 1975), p p . 97-139.
62

Becker, Sobowale and Casey, "Newspaper and Tele­
vision Dependencies".
^ G o r d o n M. Connelly and Harry H. Field, "The NonVoter--Who He is, What He Thinks," Public Opinion Quarterly
8 (Spring 1944):175-87.
gA
Kurt Lang and Gladys Engel Lang, Voting and NonVoting (Waltham, M a s s .: Blaisdale Publishing^ 19o8), p . 8T.

80
^ 0 ' Keefe and Mendelsohn, "Nonvoting1’.
66I b i d ., p. 264.
^ E u l a u and Schneider, "Political Involvement".
68

Wayne Thompson and John E. Horton, "Political
Alienation as a Force in Political Action," Social Forces
38 (March 1960):190-95.
^ W i l l i a m Erbe, "Social Involvement and Political
Activity: A Replication and Elaboration," American Socio­
logical Review 29 (April 1964):198-215.
^ S t e p h e n E. Bennett and William R. Klecka, "Social
Status and Political Participation: A Multivariate Analysis
of Predictive Power," Midwest Journal of Political Science
14 (August 1970):355-82T
^ F o r a listing of these studies, the reader is
referred to Milbrath and Goel: social class, pp. 90-96;
income, pp. 96-98; education, pp. 98-102; sex, pp. 116-118;
age, pp. 114-116; race, pp. 119-122.
The complete citations
for these studies can be found at pp. 161-214.
Lester W.
Milbrath and M.L. Goel, Political Participation:
How and
Why Do People Get Involved in Politics, 2d ed., (Chicago:
Rand McNally College Publishing Company, 1977).
77

David W. Minar, "The Community Basis of Conflict
in School System Politics," American Sociological Review
31 (December 1966):822-35.
73

Thompson and Horton, "Political Alienation".
Frederic Templeton, "Alienation and Political Participation:
Some Research Findings," Public Opinion Quarterly 30 (Summer
1966):249-61.
^ T h o m p s o n and Horton, "Political Alienation,"
p. 191.
^ M u r r a y B. Levin, The Alienated Voter:
Politics
in Boston (New York:
Holt, Rinehart and Winston, 1960),
p. 65:—
^ D w i g h t G. Dean, "Alienation:
It's Meaning and
Measurement," American Sociological Review 26 (October 1961):
753-58.

81
77Ib i d ., p. 758.
78Russell Middleton, "Alienation, Race and Ed u ­
cation," American Sociological Review 28 (December 1963):
973-77.
79

Six areas were used by Middleton to define alien­
ation:
powerlessness or efficacy, meaninglessness, normlessness, cultural estrangement, social estrangement and
estrangement from work.
80

Marvin E. Olsen, "Alienation and Political
Opinions," Public Opinion Quarterly 29 (Summer 1965):200-12.
81

Kenneth Eckhardt and Gerry Hendershot, "Trans­
formation of Alienation into Public Opinion," Sociological
Quarterly 8 (August 1967):459-67.
82

Bennett and Klecka,

83

"Social Status".

Vinyard and Sigel, "Newspapers and Voters".

84

Political cynicism was measured by responses to
four items: 1) People like me don't have any say about what
government d o e s ; 2) Politicials never tell us what they
really think; 3) I don't think public officials care much
what people like me think; and, 4) Sometimes politics and
government seem so complicated that a person like me can't
really understand what's going on.
Items one, three and
four are very similar to three items used to measure trust
in government in the present study.
Harold Mendelsohn and
Garrett O'Keefe, The People Choose a President:
Influences
on Voter Decision Making (New York: Praeger Publishers, I n c .

1TT6).-----

----------

85Ibid., p. 45.
86Ibid.
87

88
89

Levin, The Alienated V o t e r , p. 58.
Templeton,

"Alienation and Participation".

Gwynn Nettler, "A Measure of Alienation," American
Sociological Review 22 (December 1957):670-77.

82
90Ibid., p. 674.
91
Murray B. Levin and Murray Eden, "Political
Strategy for the Alienated Voter," Public Opinion Quarterly
26 (Spring 1962):47-63.
92

Angus Campbell, "The Passive Citizen," in Edward
C. Dreyer and Walter A. Rosenbaum, eds., Political Opinion
and Electoral Behavior (Belmont, Calif.: Wadsworth Publishing
Company, 1966), p p . T75-88.
93

Kraus and Davis, Effects of Mass Communication,

p. 181.
q

a

Dwight G. Dean, "Alienation and Political Apathy,"
Social Forces 38 (March 1960):185-95.
QK
Erbe, "Social Involvement";
John P. Clark,
"Measuring Alienation Within a Social System," American
Sociological Review 24 (December 1959):849-52.
96

McLeod and Becker, "Testing the Validity of Grati­
fication Measures".
97

Edward L. McDill and Jeanne Clare Ridley, "Status,
Anomia, Political Alienation and Political Participation,"
American Journal of Sociology 68 (September 1962):205-13.
98Ibid., pp. 206-7.
QQ

V.O. Key, Jr., "Varieties of Participation and the
Political System," in Dreyer and Rosenbaum, Political Opinion,
pp. 51-56.
------------- -----■^^Howard D. Hamilton, "The Municipal Voter:
Voting
and Nonvoting in City Elections," American Political Science
Review 65 (December 1971):1135-40.
■^■^O'Keefe and Mendelsohn, "Nonvoting".
109

Gerald M. Pomper, "The Presidential Election," in
Marlene M. Pomper, e d . , The Election of 1976, Reports and
Interpretations (New York"! David McKay Company, I n c . , 1977) ,
p. 72.
103

Sylvan H. Cohen, "Voting Behavior in School

83
Referenda:
An Investigation of Attitudes and Other Deter­
minants by Q Technique and Survey Research" (Ph.D. disser­
tation, Kent State University, 1971).
See also, Howard D.
Hamilton and Sylvan H. Cohen, Policy Making by Plebiscite:
School Referenda (Lexington, M a s s .: D .C . Heath and Company,
1974).
^®4Kraus and Davis, Effects of Mass Communication,
p. 181.
^■^Melvin Seeman, "On the Meaning of Alienation,"
American Sociological Review 24 (December 1959):783-91.
For
a further refinement, see Melvin Seeman, "Alienation and
Engagement," in Angus Campbell and Philip E. Converse, eds.,
The Human Meaning of Social Change (New York:
Russeli
Sage Foundation, 1972) , pp. 467-527.
^®^Dean, "Alienation and Apathy".
107Ada W. Finifter, "Dimensions of Political Alien­
ation," American Political Science Review 64 (June 1970):
389-410.
^ ^ Ibid., p. 407.
Finifter related political normlessness to political cynicism as per the work of Robert
E. Agger, Marshall N. Goldstein and Stanley A. Pearl,
"Political Cynicism:
Measurement and Meaning," Journal of
Politics 23 (August 1961):479.
■^^Marvin E. Olsen, "Two Categories of Political
Alienation," Social Forces 47 (March 1969):288-99.
llOThompson and Horton, "Political Alienation,"
p. 109.
^■^Donald E. Stokes, "Popular Evaluations of Govern­
ment:
An Empirical Assessment," in Harlan Cleveland and
Harold D. Lasswell, eds., Ethics and Business:
Scientific
Academic, Religious, Political and Military (New York:
H a r p e r , 1962), p p . 61-73.
112I b i d ., p. 67.
^ ^ E d g a r L i t t , "Political Cynicism and Political
Futility," Journal of Politics 25 (May 1963):312-23.
114Ibid., p. 319.

84
'^'’joel Aberbach, "Alienation and Political Be­
havior," American Political Science Review 63 (March 1969):
86-99.
116Ibid., p. 89.
^ ^ A r t h u r H. Miller, "Political Issues and Trust in
Government 1964-1970," American Political' Science Review
68 (September 1974):951^77:
118

Arthur H. Miller, "Rejoinder to 'Comment' by
Jack Citrin:
Political Discontent or Ritualism?" American
Political Science Review 68 (September 1974) :989-10(31.
119ibid.
1 90
l 91

Connelly and Field, "The Non-Voter".
Lang and Lang,

122

"Media and Voting".

Agger, Goldstein and Pearl, "Political Cynicism".

123

Joel Aberbach and Jack Walker, "Political Trust
and Radical Ideology," American Political Science Review
64 (December 1970):1199.
1 M M iller, "Political Issues," p. 970.
125

Jack Citrin, "Comment:
The Political Relevance
of Trust in Government," American Political Science Review
68 (September 1974) :973-85*!
126Ibid., p. 978.
1 27

Milbrath and Goel, Political Participation.

128

Mendelsohn and O'Keefe, The People Choose a
President; and, O'Keefe and Mendelsohn, "Nonvoting".
129

Angus Campbell, Gerald Gurin and Warren E. Miller,
The Voter Decides (Evanston, 111.: Row, Peterson and Company,
1954), p. 187.
130

Eulau and Schneider, "Dimensions in Involvement".

85
1 31

Morris Rosenberg, "The Meaning of Politics in
Mass Society," Public Opinion Quarterly 15 (Spring 1951):9.
132

Morris Rosenberg, "Some Determinants of Political
Apathy," Public Opinion Quarterly 18 (Winter 1954-55):350.
133

Melvin Seeman, "Alienation, Membership and Polit­
ical Knowledge: A Comparative Study," Public Opinion
Quarterly 30 (Fall 1966):355.
^3^This finding holds in hospital and reformatory
settings as well.
See Melvin Seeman and John Evans, "Alien­
ation and Learning is a Hospital Setting," American Socio­
logical Review 27 (December 1962):772-78; and, Melvin Seeman,
"Alienation and Social Learning in a Reformaroty," American
Journal of Sociology 69 (November 1963):270-84.
135

Raymond Wolfinger, Barbara K. Wolfinger, Kenneth
Prewitt and Sheila Rosenhack, "America;s Radical Right:
Politics and Ideology," in Dreyer and Rosenbaum, eds.,
Political Opinion, pp. 189-226.
^3^Shanto Iyengar, "Subjective Political Efficacy
as a Measure of Diffuse Support," Public Opinion Quarterly
44 (Summer 1980):249-56.
137I b i d ., p. 255.
138

The reader is referred to Milbrath and Goel,
Political Participation, pp. 57-61 for a listing of these
studies and p p . 1&1-214 for the complete citations. See
also, Pomper, "The Presidential Election".
139

Eulau and Schneider, "Dimensions in Involvement".

^ ^ M i l b r a t h and Goel, Political Participation.
Nie, G.B. Powell and K. Prewitt, "Social
Structure and Political Participation:
Developmental
Relationships," American Political Science Review 63 (June
and September 1969):361-78, 808-32.
■^^Olsen, "Alienation and Opinions".
^ 3John E. Horton and Wayne E. Thompson, "Powerless­
ness and Political Negativism:
A Study of Defeated Local

86
Referendums ," American Journal of Sociology 67 (March
1962):485-93.
^^Erbe,

"Social Involvement," p. 206.

145ibid.
■^^Templeton, "Alienation and Participation," p. 255.
■^^Finifter, "Dimensions of Alienation."
148

William A. Gamson, Power and Discontent (Home­
wood, 111.: Dorsey Press, 1968), p. 48; and Jeffery Paige,
"Political Orientation and Riot Participation," American
Sociological Review 36 (October 1971):810-20.
149

Seeman, "Alienation and Engagement," p. 524.

■^^Milbrath and Goel, Political Participation, p. 69.
'^'''Hamilton and Cohen, Policy M a k i n g , pp. 143-44.
152

Joan P. Sullivan Kowalski, Voter Behavior and
Campaign Strategies in School Finance Elections (Arlington,
Virginia: Educational Research Service, I n c ., T97 7 ) .
1 S3

Richard F. Carter, Voters and Their Schools:
Technical Report from the Institute for Communication
Research (Stanford, C a l i f .: Stanford University Press,
30 June 1960), p. 7.

A

154

Temporary Special Levy Commission, Summary Report
and Research Reports (Olympia, Washington: Washington
Legislature, 1 9 /l) cited in Hamilton and Cohen, Policy
M a king, pp. 144-45.
^■*Ward G. Reeder, Campaigns for School T a x e s :
Manual for Conducting Such Campaigns (Hew York:
The
MacMillan Company, 1946), p p . 43-106.
1 5fi

A

Carter, Voters and Their Schools, p. 14.
Hamilton and Cohen, Policy M a k i n g , pp. 137-38.

158

Ronald Roger Rowe, "Selective Factors Influencing

87
Voter Response to a School Bond Issue" (Ph.D. dissertation,
U.S. International University, 1971).
159

Galen R. Rarick, "Political Persuasion:
The News­
paper and the Sexes," Journalism Quarterly 47 (Summer 1970):
360-64.
------------ -------■^^Donald Eugene Krischak, "An Analysis of an U n ­
successful Rural Community College Millage Election to Deter­
mine How Selected Communication Techniques and Demography
Interacted to Influence Voter Decision" (Ph.D. dissertation,
Michigan State University, 1971).
161

Bradley S. Greenberg, "Voting Intentions, Election
Expectations, and Exposure to Campaign Information," Journal
of Communication 15 (September 1965):149-60.
162

This is confirmed by Carter, Voters and Their
Schools, pp. 204-206; Hamilton and Cohen, Policy MakingV
p p . 137-38; and, Rowe, "Factors Influencing Voter Response".
16^

Hamilton and Cohen, Policy M a k i n g , p. 138.

^ ^ R o b e r t a S. Sigel, "Election with an Issue: Voting
Behavior of a Metropolitan Community in a School Fiscal
Election," 1960, cited in Hamilton and Cohen, Policy M a k i n g ,
p. 138.
^■^Carter, Voters and Their Schools, p. 67.
"^^Hamilton and Cohen, Policy M a k i n g , pp. 78-84.
■^^Kowalski, Voter Behavior, pp. 28-30.
James Q. Wilson and Edward C. Banfield, "PublicRegardingness as a Value Premise in Voting Behavior,"
American Political Science Review 58 (December 1964):876-87.
■^^Arthur J. Alexander and Gail V. Bass, School
Taxes and Voter Behavior: An Analysis of School District
Property Tax Elections (Santa Monica, C a l i f .: The Rand Cor­
poration , 1974), p . 55.
^^Kowalski, Voter Behavior, p. 32.
■^^"Thompson and Horton, "Political Alienation".

88
^7^Philip K. Piele and John Stuart Hall, Budgets,
Bonds and Ballots: Voting Behavior in School Financial
Elections (Lexington, Mass.: D.C. Heath and Company, T973),
p. 129.
173

Reeder, Campaigns for School T a x e s .

^ T e m p l e t o n , "Alienation and Participation," p. 256.
^ " ’Horton and Thompson, "Powerlessness".
^■^Eckhardt and Hendershot,
Alienation".

"Transformation of

177I b i d ., p. 467.
^7^Hamilton and Cohen, Policy M a k i n g , p. 203.
179I b i d ., p. 230.
1 ftfi
Carter, Voters and Their Schools, pp. 132-36.
^"Frederick M. Wirt and Michael W. Kirst, The
Political Web of American Schools (Boston: Little, Brown
and Company, 1972) , p . l O l .
1 QO

Michael W. Giles, Douglas S. Gatlin and Everett
F. Cataldo, "Parental Support for School Referenda," Journal
of Politics 38 (May 1976):442-51.
^ ^ H a m i l t o n and Cohen, Policy M a k i n g , p. 229.
^®^See Bennett and Klecka, "Social Status"; Mendel­
sohn and O'Keefe, The People Choose a President; Agger,
Goldstein and Pearl, "Political Cynicism"; and, Milbrath
and Goel, Political Participation.
IQ C

See Greenberg, "Media Use"; Wade and Schramm,
"Media and Sources of Public Affairs"; Kline, "Media Time
Budgeting"; Clarke and R u g g e l s , "Preferences Among News
Coverage"; Eldersveld, "Experimental Propaganda Techniques";
Eulau and Schneider, "Dimensions in Involvement"; Tan,
"Political Participation"; and, O'Keefe, "Political
M a laise".

89
1 86

See Hamilton and Cohen, Policy M a k i n g ; Kowalski,
Voter Behavior; Alexander and B a s s ,~~5chool, Taxes and Voter
Behavior; a n d ,' Giles, Gatlin and Cataldo, "Parental
Support".
187

See Temporary Special Levy Commission, Summary

Report.
188

See Thompson and Horton, "Political Alienation";
Horton and Thompson, "Powerlessness"; Erbe, "Social Involve­
ment"; Seeman, "Alienation and Engagement"; Milbrath and
Goel, Political Participation; and, Piele and Hall, Budgets,
Bonds and Ballots.
1 89

See Templeton, "Alienation and Participation";
and, Temporary Special Levy Commission, Summary R e p o r t .
190

See Krischak, "Analysis of an Unsuccessful Millage

Election".
191

See Lang and Lang, "Mass Media and Voting"; and,
Becker, Sobowale and Casey, "Newspaper and Television
Dependencies".
192

See Wolfinger, Wolfinger, Prewitt and Rosenhack,
"America's Radical Right".
i q i

See Milbrath and Goel, Political Participation;
and, Finifter, "Dimensions of Alienation".
194

See Rosenberg, "Meaning of Politics"; Seeman,
"Alienation and Membership"; and, Milbrath and Goel, Polit­
ical Participation.
195

See O'Keefe and Mendelsohn, "Nonvoting", and,
Citrin, "Comment".
196

See Becker, Sobowale and Casey, "Newspaper and
Television Dependencies".
197

See Buchanan, "Purposive Voting"; Conway, "Voter
Information", and McLeod, Bybee, Luetscher and Garramore,
"Voter Volatility".
198

See Levin, The Alienated V o t e r ; Nettler,
of Alienation"; Kraus and Davis, Effects of Mass

"Measures

90
Communication; Dean, "Alienation and Apathy"; Erbe, "Social
Involvement"; and Clark, "Measuring Alienation".
1 QQ

See Katz and Foulkes, "On the Use of Mass Media";
Blumler and McQuail, Television and Politics; McLeod and
Becker, "Validity of Gratifications Measures"; and, Semlac
and Williams, "Uses, Gratifications and Avoidance".
^^^See Rosenberg, "Determinants of Apathy"; and,
van den Haag, "Of Happiness and Despair".
^ ^ S e e Lang and Lang, "Media and Voting"; O'Keefe,
"Political Malaise ; and, Eldersveld, "Experimental Propa­
ganda Techniques".

202

The symbol Hn will be used throughout to indicate
an hypothesis; the symbol H q its null.
203

Group I includes those who are high in efficacy
and high in trust; Group II, those high in efficacy and
low in trust; Group III, those low in efficacy and high
in trust; and, Group IV, those low in efficacy and low in
trust.
See Chapter III for information on assignment of
respondents to g r oups.
20^i=8th grade or less; 2=some high school; 3=high
school diploma; 4=trade school after high school; 5=some
college; 6=college degree; 7=some graduate school; and,
8=graduate degree.
^■*l=less than $1 0 ,0 0 0 ; 2=more than $10,000 but less
than $20,000 (also includes those reporting $10,000); 3=more
than $20,000 but less than $30,000 (also includes those
reporting $20,000); and, 4=$30,000 or more.
206

l=farming; 2=blue collar; 3=white collar; 4=manager; and, 5=professional.

CHAPTER III
METHODOLOGY
Research Design
The study is part of a larger, twelve-month study
undertaken a Michigan State University using a metric
multidimensional scaling technique known as Galileo.^"

The

larger study was designed to track attitudes of registered
voters in Shiawassee County concerning vocational education
across time and to provide the Michigan State Department of
Education and the Shiawassee County Intermediate School Di s ­
trict with information on the design of the best campaign
messages to assure passage of the millage for the Area Voca­
tional Technical Education Center.

This study focuses on

that portion of the data collected after the actual millage
election had been called.
The main study began in June 1978 with a series of
indepth interviews with both a scientifically selected sample
of Shiawassee County residents and a sample of community
leaders.

Based on the results of these interviews the main-

wave metric multidimensional questionnaire was developed
and administered to approximately one hundred residents of
the Shiawassee County Intermediate School District beginning
the week of July 17, 1978, and continuing through August
91

92
1978.

These data provided the baseline information for

the development of the message campaign.

Interviewing

resumed on January 2, 1979, and continued through June 10,
1979.

Interviews were conducted daily and interim reports

on changes in attitude structure were provided on February
28, April 15, April 29, May 17 and June 4.

The data for

this study are drawn from the interviews conducted between
April 12, the day the election was formally announced, and
June 10, 1979, the day before the election was held.
A telephone survey was utilized as the primary
method to collect data for the larger study.

This method

was selected for the original study for a number of reasons.
First, the study required that attitudes be tracked daily to
measure possible changes brought about by the message cam­
paign being presented by the Intermediate School District.
This meant that the data be collected using either a tele­
phone survey or a personal interview rather than a mail sur­
vey where it would be impossible to make daily contact with
potential respondents.

Second, the complicated nature of

the data collection instrument necessitated personal con­
tact between the respondent and interviewer, once again
pointing to the need for either a personal or telephone
interview.

Of the two methods considered, personal inter­

view or telephone survey,

the latter was chosen due to the

third factor, the geographic dispersion of potential respondents.

2

The cost of data collection was also lessened con­

siderably by adoption of the telephone as the instrument of

93
data collection as personal interviews involve travel
which is costly in both time and money.

3

Telephone surveys have the advantage of a reasonable
4
cost, rapid completion and a relatively high response rate.
The sampling technique generally involves minimum expense
and call backs are simple and economical.

Additionally,
fi
interviewers can clarify any misunderstandings , a partic­
ularly important factor in a study utilizing a new type of
questionnaire.

Finally, it is simple to train, supervise

and monitor the performance of interviewers.^
The major disadvantage of telephone surveys is the
fact that a percentage of homes do not have telephones, par­
ticularly in rural areas such as Shiawassee County or in
inner city areas

8

, or have unlisted numbers.

9

It is not

known if some crucial differences exist among homes with
listed telephones, unlisted telephones and no telephones.
The Sample
The sample of 1108 potential respondents for the
main study was drawn from the 1978 Shiawassee County Voter
Registration lists using a systematic random sampling tech­
nique.

It had been determined that, to complete several

calls a day between January 2 and June 10, a sample of
eleven hundred would be needed.

Based on this figure and

42,129, the number of registered voters in Shiawassee County,
a skip interval of thirty-eight was calculated.

From a

random starting point every thirth-eighth name was selected

94
resulting in a sample of 1108, 2.67, of all registered
voters in Shiawassee County.

As the voter registration

lists did not contain telephone numbers, the appropriate
telephone directories and directory assistances were con­
sulted to finalize the sample.
Interviews were completed with 257 respondents for
the main survey, a 23.27, completion rate.

Of these, 148

(57.6%) were accomplished between January 2 and April 11
while 109 (42.4%) were completed between April 12 and
June 10.

It is upon these latter respondents that this

study focuses.

It should be noted that the low completion

rate was due in part to the length of the interview (thirtyfive to forty-five minutes on average) and in part to the
fact that telephone numbers were, at times, unavailable.
The in-tab sample for the main study was not truly
representative of the school districts as Laingsburg was not
represented at all while, for the April to June segment, New
Lothrop was not represented as well.

Care must therefore

be taken in generalizing to the county as a whole.

Table 1

shows the 1978 voter registration data from Shiawassee County
by school district and the sample data.
Questionnaire Development
The questionnaire for the overall study collected
information in six general areas:

attitudes toward cen­

tralized vocational education using a metric multidimensional
scaling design, political efficacy and trust in government

TABLE 1
POPULATION/SAMPLE COMPARISON

District

Population

'o

Sample

°i

January-April

°L

April-June

%

Byron

2,461

6

6

2

5

3

1

1

Corunna

6,392

15

41

16

24

16

17

16

Durand

6,930

16

41

16

12

8

29

27

Laingsburg

2,026

5

—

--

—

--

--

—

New Lothrop

2,491

5

3

1

3

2

Morrice

1,730

4

13

5

6

4

7

6

Owosso

15,672

37

115

45

78

53

37

34

Perry

4,427

11

24

9

9

6

15

14

TOTAL

42,129

243b

94c

137d

92e

106f

99a

a
Total does not equal 100% due to rounding.
«
Fourteen cases were not identified as
to school district.
A

Percent is based on 257 cases.
^Eleven cases were not identified as
to school district.
ePercent is based on 148 cases.

—

97s

f
Three cases were not identified as
to school district.
^Percent is based on 109 cases.

96
scales, awareness of the upcoming election, media use and
information sources, voting, arid

demographics.

The

follow-up questionnaire collected data regarding respon­
dent's information about the election, the source of that
information, contact prior to the election, whether or not
the respondent had voted and how they had voted.
The primary methodology employed in the examination
of the first item was the Galileo system, a series of p ro­
cedures for making a "mental map" of the audience for a
persuasive campaign,

in this instance a campaign to convince

voters of the need to construct a centralized Area Voca­
tional Technical Education Center to serve the needs of high
school students in Shiawassee County.

The system identified

the main concepts registered voters in the county used to
understand and define centralized vocational education and
measured the beliefs and attitudes about those concepts.
This was accomplished through twenty-five indepth personal
interviews with registered voters in Shiawassee County.

The

unstructured interviews, lasting on average about twentyfive minutes, focused on the respondent's feelings and ideas
about vocational education and elicited approximately thirtyfour concepts about this topic.

Of these, nine concepts

receiving at least twenty mentions were chosen for
inclusion:

tax increase, expensive, practical training,

work skills, needed, job opportunities,
of programs and discipline problems.

college, duplication

Four other concepts

(your vote for, your vote against, area skills center,

97
Intermediate School District) were incorporated into the
final instrument.

Each concept was then paired with each

of the others and respondents were asked to estimate the
distance between two concepts based on a unit difference
of ten, in the present case, the distance between practical
training and work skills being ten units apart.

How this is

accomplished can be seen by examining the instructions and
questions 1 through 78 on the instrument included in Appen­
dix 1.

As stated earlier, data from this first section of

the questionnaire was analyzed e l s e w h e r e . ^

The remaining

items are of concern h e r e .
The model developed for this study proposed that two
dimensions of political alienation, political efficacy and
trust in government, can be used as predictors of infor­
mation u s e . ^

Respondents were asked, therefore, to indi­

cate agreement or disagreement to two series of attitude
statements.

The first six items dealt with the respondents'

sense of political efficacy.

The items used were drawn from

the University of Michigan Survey Research Center's Political Efficacy Scale
Walker

13

12

, however, as suggested by Aberbach and

, the items were localized so as to deal specifi­

cally with the School Board.

For example, item one read

"People like me don't have any say about what the School
Board does" rather than the standard " . . .
ment does.".

what the govern­

Respondents may feel inefficacious about an

amorphorous "government" but highly efficacious when that
government is something quite local, like a school board.

98
Since the study dealt with a truly local event, a millage, a
localized version of the scale was developed and used.

The

remaining five items of the scale, then, were as follows:
--Voting is the only way people like me can have
a say about how the School Board is run.
--Sometimes School politics seems so complicated
that a person like me can't really understand
w h a t 's going o n .
--I don't think school officials care much what
people like me
think.
--Generally, those people we elect to the School
Board lose touch with the people pretty quickly.
--School Board members are only interested in
people's votes and not in their opinions.
Five items comprised the trust in government scale,
drawn from the University of Michigan Survey Research
.
14
Center s Trust in Government Scale.

As with the previous

scale, this was also localized to maintain continuity in
thought and to make sure that the two scales dealt with
attitudes toward the same government entity.

As the two

scales were to be combined to yield a single measure of
political alienation,
same

concept.

the school board,

it was necessary that both measure the

Itis possible to trust "government" but not
or

vice versa. Combining a sense

of

political efficacy toward the school board with trust in an
amorphorous "government" would be the attitudinal equivalent
of mixing apples and oranges.

The items, then, used in the

scale were:
--First of all, do you think that people on the School
Board waste a lot of money
we pay in taxes, waste
some of it, or don't waste very much at all?

99
--How much of the time do you think you can trust
the School Board to do what is right--just about
always, most of the time, or only some of it?
--Would you say the School Board is pretty much run
by a few big interests looking out for themselves
or is it run for the benefit of all the people?
--Do you feel almost all of the people running the
School Board are smart people or do you think
quite a few of them don't seem to know what they're
doing?
--Do you think quite a few of the people running the
School Board are crooked, not very many are crooked,
or do you think hardly any of them are crooked?
The third section of the instrument dealt with
awareness of the election.

Respondents were first asked if

they were aware of any upcoming special, county-wide
elections, a measure of unaided awareness.

If the respondent

indicated he was aware of such, he was queried as to who had
called the election (County Attorney,

Intermediate School

District, State Department of Education, or Someone Else),
the issue of the election (millage for a new grade school,
area vocational technical education, or special education)
and the amount of the millage (less than one mill, between
one and two mills, more than two mills).

Those respondents

who answered negatively to the unaided recall item were
asked if the Shiawassee County Intermediate School District
had called any special elections in the near future, an aided
awareness item.

If awareness was indicated, they were asked

the specific questions of issue and amount of millage.
The fourth section of the questionnaire dealt with
the respondent's media use and source of information

100
regarding the election.

Respondents were asked to indicate

if they regularly read a newspaper, which newspaper(s) they
read and the newspaper(s)

to which they subscribed.

The

second set of media items dealt with radio listenership:
hours of listening per day, to which stations they listened,
and if it was the local radio outlet whether they listened
to AM, FM, or both.

As Shiawassee County has no local tele­

vision outlet, no questions were asked regarding television
v i ewing.
To determine source of information, respondents were
asked if they had heard anything about Area Vocational Tech­
nical Education "since last Monday" and, if they had, to
identify the source:

newspaper, radio, friend, flyer or

brochure, poster, m e e t i n g . ^
The fifth section of the instrument dealt with the
respondent's past voting behavior in both general and school
elections in the past two years, when they had last voted in
a school millage election and how they had voted in that
election.

To measure expected voting behavior, respondents

were first given the particulars of the election under study
and then asked if they thought they might vote in it and if
they would, the chances they would vote in favor of the
millage.
Demographic items made up the final section of the
questionnaire.
graphic

Most of the items used were standard demo­

questions found in most surveys:

home ownership,

education, presence of children in the home, age of

101
respondents, employment (full or part-time), occupation,
income and sex of respondent.

Several additional study-

specific items were included.

To measure sample repre­

sentativeness, respondents were asked in which school dis­
trict they resided.

To determine if participation in voca­

tional education could bias attitudes toward it, respondents
were asked whether they or any of their children had ever
taken a vocational education class in high school.

Finally,

respondents were asked whether they worked in the county or
somewhere else to see if favorability toward the issue was
higher among those working within the county.
The follow-up questionnaire included seven ques­
tions.

Respondents were asked to identify the ballot issue

and the millage amount.
of unaided awareness.

These two items provided a measure
Respondents were then asked if they

had "heard" anything about Area Vocational Technical Edu­
cation, and if so, the source, including personal contact
by the Intermediate School District or Citizen's Committee.
Finally, respondents were asked if and how they had"voted
in the election.
Pretest
The majority of the survey instrument had been uti­
lized in the field prior to the election being called (See
Appendix 1).

This included the metric multidimensional

scaling portion and certain of the demographic items.

102
Problem areas in these sections had been eliminated prior
to this phase of the study.
The final questionnaire (see Appendix 2) was pre­
tested in the field under actual survey conditions.

Ten

potential respondents were randomly selected from the voter
registration lists and were contacted by telephone.

Upon

completion of the instrument, they were asked if there had
been any problem areas with the questionnaire.

Based on the

results of the pretest, it was not necessary to revise the
instrument further.
Data Collection and Preparation
Data for the study were collected continuously from
April 12, 1979, the date of the announcement that the elec­
tion had been called by the Intermediate School District,
through June 10, 1979, the day before the election.

Inter­

views were conducted throughout this period.
Telephone calls were placed Sunday through Friday
evenings between 6:30 p.m. and 9:00 p.m.

In addition, calls

were made during the afternoon hours on Saturday and Sunday.
Four attempts, at different times and on different days,
were made to reach each sample member.

Interviewers were

instructed to make an appointment if the individual was
unable to participate at that time.
For the follow-up data collection, an attempt was
made to contact all original respondents to the entire
study.

Calls were placed between June 12 and June 19, 1979,

103
the week immediately following the election on the same
calling schedule as before:

Sunday through Friday between

6:30 p.m. and 9:00 p.m. and Saturday and Sunday afternoon.
Respondents were not told that they were drawn from those
previously contacted.

Rather, they were told that the

caller was from Michigan State University and was doing a
study of voting behavior in Shiawassee County.
The questionnaire utilized for the study was of the
pre-coded type.

The data were coded on the instrument

itself, then keypunched and verified.

The data were com­

puter analyzed using the Statistical Package for the Social
Sciences (SPSS)
University.

on the CDC computer at Michigan State

Additional data analysis was performed by hand

where the needed tests were not available from SPSS.
Data Analysis
A variety of statistical tests were employed to test
the hypotheses outlined in Chapter II.

The first hypothesis

stated that there would be different levels of awareness
among the four groups.

Sub-hypotheses specified that the

proportion aware in group one would be significantly greater
than the proportion aware in group two; the proportion aware
in group three would be significantly greater than the p ro­
portion aware in group four; and, that the proportion aware
in groups one and two combined would be significantly
greater than the proportion aware in groups three and four
combined.

To test this hypothesis, respondents were

assigned into one of the four groups on the basis of their
responses to the political efficacy and trust in government
scales.
Each efficacy item had an efficacious and an ineffi
cacious response.

Based on responses to each item, a total

score was computed for each respondent .^

Those respondent

scoring a six to nine were considered high in political
efficacy; those ten to twelve, low in political efficacy.
Likewise, each trust in government item had a trustful and
a distrustful response.
had a middle response.

Additionally,

three of the items

For example, when asked about the

School Board wasting tax money respondents could answer
"waste a lot", "waste some" or "don't waste very much at
all".
dent.

Again, total scores were computed for each respon18

Those respondents scoring a five to nine were

considered high on trust, those ten to fifteen, low on
trust.

Assignment to each of the four groups was made on

the following basis:
Group I:

High in Efficacy (6 to 9), and
High in Trust (5 to 9);

Group II:

High in Efficacy (6 to 9), and
Low in Trust (10 to 15);

Group III:

Low in Efficacy (10 to 12), and
High in Trust (5 to 9);

Group IV:

Low in Efficacy (10 to 12), and
Low in Trust (10 to 15).

The first hypothesis was tested using a Chi-Square
test to determine if the distribution among the four groups
of those aware differed significantly from chance.

The

105
three sub-hypotheses were tested using a difference of
proportions Z-test.

The Z-test was chosen for the sub­

hypotheses because interest was in the proportion aware,
not in the number aware.

It was also recognized that the

four groups might have different numbers of respondents.
Examination of the proportions would equalize the groups
in case this did occur.

Rejection of the null hypothesis

took place at an alpha level of .05.

19

The second hypothesis stated that there would be
significantly differential levels of information among the
four groups.

The sub-hypotheses specified that the mean

level of information would be significantly greater for
group one than for group three;
for

group four;

greater for group two than

and, the average of means for groups one

and two greater than the average of means for groups three
and four.

To test the hypothesis, a correctness of infor­

mation variable was created.

Respondents deemed aware of

the election without aid were asked who had called the
election, the issue and the millage amount;
aid, the issue and the amount.

those aware with

Each respondent received a

score of zero to three depending on the number of correct
responses.

The hypothesis was tested using a one-way

analysis of variance.

The three sub-hypotheses were tested

using planned comparisons and a t-test difference of means.
The third hypothesis explored media use differences
proposing that groups one and three would rely on media
sources to a significantly greater extent than groups two

106
and four.

The first step in testing this hypothesis was

the creation of a source of information variable.

The

questionnaire had included a series of six items for those
respondents who had heard anything about the issue "since
last Monday".

If the respondent answered affirmatively for

an item, a score of one was assigned; a negative response,
a zero.
media;

Two sources, newspaper and radio, were considered
the remaining four (friend, flyer or brochure,

poster, meeting) were deemed to be non-media sources.

The

proportion of media sources used was calculated for each
g r oup.

Groups one and three were combined as were groups

two and four.

A difference of proportions Z-test was pe r ­

formed to test the hypothesis that groups one and three
would report media sources in a significantly greater pro­
portion that groups two and four.

Again, the difference of

proportions test was used due to the unequal number of
respondents in each g r o u p .
The fourth hypotheses stated that there would be
significant increases in the proportion of those aware of
the election as the campaign progressed.

To test this hyp o ­

thesis , the proportion of those aware each week was calcu­
lated.

The proportion aware for each week was compared,

using a difference of proportions test, with the proportion
aware the previous week to determine if significant in­
creases were taking place.

The difference of proportions

Z-test was specifically selected to allow for directional
comparisons and, because the number contacted each week

107
varied, to provide a means of equalizing the groups across
ti m e .
Hypothesis five is similar to four in that it pro­
posed that the proportion of those holding correct infor­
mation would increase significantly across the campaign.
The proportion of those responding correctly to all three
information items concerning the election and thus scoring
a three on the correctness of information variable created
for hypothesis two was compared across the weeks of the cam­
paign.

Nine differences of proportions tests were performed

to test this hypothesis, as per the tests done for hypo­
thesis four.
The sixth hypothesis stated that a significantly
greater proportion of respondents in Owosso-Corunna would
receive their information from newspapers than respondents
in the outlying school districts.

To test this hypothesis,

respondents were sorted into two groups:

those residing in

the Owosso-Corunna school districts and those residing
in the remaining six districts.

The proportion of those

who had read of the issue in the newspaper was calculated
for each group.

A difference of proportions Z-test was

performed to test the statistical significance of the
hypothesis.
The seventh hypothesis examined the relationship
between favorability towards the issue and age and predicted
that as age increased, favorability would decrease.

Favor­

ability towards the issue was based on the respondent's

108
self-reported likelihood of voting in favor of the issue.
As both age and favorability were measured at at least the
interval level, a Pearson Product-Moment Correlation was
performed between the two vairables with a Student's t
statistic utilized to determine statistical s i g n ificance.^
The effect of the presence or absence of children
in the home on favorability towards the issue was tested in
hypothesis eight.

The respondents were divided into two

groups based on whether or not they had children currently
living at home.

The mean favorability level was computed

for each group and a difference of means t-test was p e r ­
formed to determine if the difference was statistically
significant.
Hypothesis nine predicted that favorability towards
the issue would increase significantly as education level
increased.

To test the hypothesis, a mean favorability

level for each educational level was calculated.

A dif­

ference of means t-test was performed between each education
level and the next highest one to determine if support for
the hypothesis was present in the data.

This method of

statistical testing was selected because education was not
measured at the interval level, hence a correlational
analysis was not possible, and because it would allow for
directional testing.
The relationship between income and favorability,
such that as one increased the other would as well, was
the subject of hypothesis ten.

To test the prediction, a

109
mean favorability level was calculated for each income
level.

As income was not measured at the interval level,

it was not possible to test the hypothesis using corre­
lations.

Therefore, a difference of means t-test was p er­

formed between each income level and the next highest.
The eleventh hypothesis stated that there would be
a significant increase in favorability towards the issue as
occupational status increased such that the professional
would be more favorable than, in decreasing order, the
managerial, the white collar,
farmer.

the blue collar and the

To test the hypothesis mean favorability levels

were calculated

and a difference of means t-test was per­

formed between each occupational level and the next lowest.
Hypothesis twelve examined past voting and favor­
ability proposing that those who had cast a "yes" vote on a
school millage in the past would be more favorable towards
the current issue.

The respondents were sorted into two

groups based on their self-report of past voting behavior.
A mean favorability level was calculated for each group and
a difference of means t-test performed between the two
groups to test the hypothesis.
The thirteenth hypothesis predicted that alienation
would be a significant independent predictor of favorability.

21

To test the hypothesis an alienation variable

was created by combining (summing) a respondent's scores
on the political efficacy and trust in government variables.
Those respondents receiving large scores on the created

110
variable were said to be more alienated than those receiving
smaller scores.

A multiple regression analysis was then

performed on the data with favorability as the dependent
variable and alienation, presence of children in the home,
someone in the household taken vocational education, home
owner, work in Shiawassee County, age, income and sex as the
independent variables.

All independent variables, save

alienation, age and income, were coded as dichotomous dummy
variables.

A variable was considered an independent pre­

dictor if the probability associated with the individual F
statistic was less than or equal to .05.
The final hypothesis proposed that alienation would
be a significant independent predictor of negative voting.
To test hypothesis fourteen, a discriminant analysis was
performed as the dependent variable was dichotomous.

The

dependent variable was drawn from the self-report of voting
behavior during the follow-up portion of the survey.

The

independent variables were identical to those used in hypo­
thesis thirteen.

As above, a variable was considered an

independent predictor if the probability associated with
the individual F statistic was less than or equal to .05.
In addition to the testing of the fourteen hypo­
theses in the study, a demographic profile of each source
type was developed.

A demographic profile for the sample as

a whole was also created to check sample representativeness
against other published data regarding the demographic
composition of the county.

Ill
Endnotes
■^For a discussion of Galileo, see Joseph Woelfel
and Edward L. Fink, The Measurement of Communication
Processes:
Galileo Theory and Method (New York; Academic
t r e s s , 1980); Joseph Woelfel, “Foundations of Cognitive
Theory: A Multidimensional Model of the Message-AttitudeBehavior Relationship" in Donald P. Cushman and Robert D.
M c P h e e , e d s ., Message-Attitude-Behavior Relationship:
Theo r y , Methodology and Application (New York: Academic
P r e s s , 1980); and, Joseph woelfel and Jeffrey E . D a n e s ,
"Multidimensional Scaling Models for Communication Research"
in Peter R. Monge and Joseph N. Cappella, eds., Multi­
variate Techniques in Human Communication Research (New
York: Academic P r e s s , 1980).
9

Selltiz, et al point out that telephone interviews
are particularly advantageous in this type of situation.
Clair Selltiz, Lawrence S. Wrightsman and Stuart W. Cook,
Research Methods in Social Sciences 3rd ed. (New York:
holt, Rinehart and Winston,
1976), p. 299.
3I b i d .
^ S . Sudman, Reducing the Cost of Surveys (Chicago:
Aldine-Atherton, 1967); Roger D. Wimmer and Joseph P.
Dominick, Mass Media Resea r c h : An Introduction (Belmont,
Calif.: Wadsworth Publishing Company, 1983), p . 130.
3Wimmer and Dominick, Mass Media Research, p. 130.
6Ibid.
^Seltiz, Wrightsman and Cook, Research Methods,
p. 299.
8I b i d ., p. 298.

g

Wimmer and Dominick, Mass Media Research, p. 130.

^ M a r t i n Block and Joseph Woelfel, Area Vocational
Education Campaign Strategies and Techniques Project, Phase
I: Communication Strategies (East Lansing, Michigan: College
of Communication Arts and Sciences, Michigan State Univer­
sity, September, 1978); Martin Block and Joseph Woelfel,
Area Vocational Education Campaign Strategies and Tech­
niques Project, Phase II: Campaign Development, Phase til:
Campaign Evaluation (East Lansing, Michigan: College of

112
Communication Arts and Sciences, Michigan State University,
June 29, 1979).
"^For a complete explanation of these items, see
the Review of Literature in Chapter II.

12

John P. Robinson, Jerald G. Rusk and Kendra B.
Head, Measurement of Political Attitudes (Ann Arbor,
Michigan: Institute for Social Research, The University of
Michigan, September 1966).
13

The authors state:
"However, items in many scales designed to measure
political distrust often involve simple cliches
about the quality of politics and politicians with
little or no indication as to the governments or
figures involved . . . it would seem vitally impor­
tant that the subject be stimulated to think about
some more focused symbol than 'politicians'."
Joel Aberbach and Jack Walker, "Political Trust and Radical
Ideology," American Political Science Review 64 (December
1970):1201.
Other authors have also implied the need for
the "localization" of scales.
Citrin suggests many polit­
ical cynics focus their dissatisfaction on incumbent author­
ities rather than system values.
Dean stated, "It might be
profitable to develop - instead of a general alienation
scale - scales to be specifically applied to various insti­
tutional areas of social l i f e " . McClosky points out that
"It is impossible in the present context to determine the
extent to which scores contained in these tables signify
genuine frustration and political disillusionment and the
extent to which they represent familiar and largely ritual­
istic responses".
Jack Citrin, "Comment: Political Rele­
vance of Trust in Government," American Political Science
Review 68 (September 1974):975; Dwight G. Dean, "Alienation
and Political Apathy," Social Forces 38 (March 1960):189;
and, Herbert McClosky, "Consensus and Ideologies in American
Politics," American Political Science Review 58 (June
1964):370.
1/

Robinson, Rusk and Head, Measurement of
Political Attitudes.
^ A s there was a dearth of media in Shiawassee
County and as the School Board itself was relying on lessthan-mass media and interpersonal methods of distribution,
it was necessary to include a wide range of sources of
information.
The model tested herein also presumes that
certain types of people will rely on varying types of
information sources.

113
16

Norman H. Nie, C. Hadlai Hull, Jean G. Jenkins,
Karin Steinbrenner and Dale H. Bent, Statistical Package
for the Social Sciences 2d ed. (New York: McGraw-Hill Book
Company, 1975).
^ T h e efficacious response was assigned a one, the
inefficacious response, a two.
18

The most trustful response was scored a one, the
middle response, a two and the most distrustful response,
a three.
19

This alpha level is used throughout the study for
rejection of the null hypothesis.
As such, it will not be
mentioned with each hypothesis.
The reader is to assume
that the level of rejection is .05.

20

Student's t is the statistic reported in the
PEARSON CORR procedure of S P S S . See N i e , et a l , Statistical
Package, p. 281.

21
For a complete discussion of alienation as used
in this study, the reader is referred to the Review of
Literature and Definition sections of Chapter I I .

CHAPTER IV
RESULTS
Sample Characteristics
Demographics
The greatest majority of respondents to the survey
own their own homes (n=8 6 ; 86.9%) while 8.1% (n=8) are
renters.

More than half are employed full-time (n=56;

56.6%.) while 8.1% (n=8) work part-time and 35.47> (n=35)
do not work outside the home.

Of those that are employed,

thirty-eight (59.5%) work in Shiawassee County and twentysix (40.67o) are employed elsewhere.

The blue collar cate­

gory was the largest occupation group with 41% (n=25) of the
respondents reporting occupations classified as such.

The

second largest category was white collar (n=19; 31.2%,)
followed by professional (n=10; 16.4%) and managerial (n=6 ;
9.8%,).

Only one respondent (1.6%) reported farming as an

occupation.
The overall education level of the sample as a whole
was low.

Fully 63.4%, (n=64) of the respondents have only a

high school diploma or have attended trade school after high
school.

Twenty-nine respondents (28.7%,) have not completed

high school.

Twenty-three respondents

(22.8%,) have some

college and nine (8.9%,) have earned a college degree.
114

Five

115
respondents (4.9%) had

attended graduate school with three

of them (3.0%) having earned a graduate degree.

Few respon­

dents (n=1 0 ; 19.2%) had ever taken a vocational education
class in high school or had children who had taken such
courses (n=21; 24.7%).
Income levels were fairly low with 17.7% (n=17) re ­
porting a total household income of less than $10,000 a
year; 25%, (n=25) , an income of between $10,000 and $20,000 a
year; and, 32.37. (n=31) , an income of between $20,000 and
$30,000 a year.

Only 18.7% (n=18) report a total household

income of greater than $30,000 a year.
Almost 63%, (62.4%,; n=63) of the sample had childern
living at home.

The sample was fairly evenly distributed

across all age groups:

20 to 29: n = 1 9 , 17.8%; 30 to 39:

n = 2 3 , 21.5%; 40 to 49: n=22, 20.6%; 50 to 59: n=19, 17.8%;
60 to 69: n=12,

11.2%; and, 70+: n = 6 , 5.6%.

Thirteen

respondents (15.5%,) were 65 years of age and older.
the sample was

Finally

38.1% male (n=40) and 61.9%» female (n=65) .

The characteristics of the sample of respondents
utilized in this study point to the kind of area Shiawassee
County is.

It is a relatively stable area with a majority

of home owners and few renters.

It is also a "bedroom

community" of sorts with a substantial minority of those
who work being employed outside the county (40.6%,).

It is

also primarily a blue collar area with many residents being
employed in the automotive factories in the surrounding
counties.

116
The education and income levels are reflective of
the occupational standings in the county.

Few respondents

have attended any schooling past high school.

These respon­

dents, for the most part, make up the blue collar and white
collar occupational rankings.

The fact that 57% of the

sample reports incomes of between $10,000 and $30,000 a year
further supports the assumption that the county is primarily
a middle class areas.

One interesting, and somewhat unex­

pected, finding was the fact that less than one-fourth of the
respondents had taken, or had someone in the family take, a
vocational education class in high school.

Given the

employment picture in the county one would expect this
figure to be quite a bit higher.
Attitudes
Six items were used to measure a sense of political
efficacy as it relates to the School Board and school poli­
tics .

On two of the ite m s , the sample registers feelings of

inefficacy while on the remaining four, the majority have
efficacious feelings

(see Table 2).

The sample is more

likely to agree that voting is the only way they can have
a say about how the School Board runs things and that some­
times school politics seems so complicated that they can't
really understand what's going on.

Respondents are more

likely to disagree that people like themselves don't have
any say about what the School Board does, that they don't
think school officials care much what people like themselves

117
TABLE 2
POLITICAL EFFICACY
Item

Agree

People like me have no say
about what the School Board
does.

34.3%
(n=34)

62.6%
(n=62)

3.0%
(n= 3)

Voting is the only way to
have a s a y .

61.0%
(n=61)

34.0%
(n=34)

5.0% 100
(n= 5)

School politics seems to
complicated

53.5%
(n=53)

42.4%
(n=42)

4.0%
(n= 5)

99

School officials don't care
what people think

34.0%
(n=33)

54.6%
(n=53)

11.3%
(n=ll)

97

People on the School Board
lose touch.

41.1%
(n=39)

50.5%
(n=48)

8.4%
(n= 8)

95

School Board members are
only interested in votes.

28.9%
(n=28)

60.8%
(n=59)

1 0 .3%
(n=10)

97

Disagree

Depends

N
99

think, that people on the School Board lose touch prety
quickly after being elected, and, that School Board members
are only interested in people's votes and not their opinions.
The attitude statements point to differences between
the Shiawassee sample and other studies investigating the
presence or absence of political efficacy in school refer­
enda.

Carter had found that one-third of the voters felt

they had little say in what the School Board did.^
Shiawassee sample agreed (34.37„) with this finding.

The
Where

only one-half of Carter's sample believed that the only
voice voters had was the act of voting, 61% of the Shia­
wassee sample agreed with this statement.

Two-fifths of

Carter's sample felt that educational policy was too

118
complicated for persons like themselves to understand.
Using a similar item, more than half of the Shiawassee
sample believed school politics was too complicated for
them to understand (53.5%).

Carter found that one-fourth

felt that school officials did not care what the average
voter thought.
wassee study.

Two items measured this concept in the Shia­
In both instances the present sample is less

efficacious, more likely to agree that school officials
don't care what people think (34%) and that School Board
members are only interested in votes

(28.9%).

Carter con­

cluded that the presence of low efficacy would almost
assure a negative view of the schools.
the case, the negative view

If that is indeed

would be stronger in Shiawassee

County.
Five items comprised the trust in government, in
this case the School Board, scale.

An examination of the

items shows that the sample is generally trustful of the
School Board.

Only 18.1%, (n=17) of the respondents believe

that the School Board wastes a lot of money paid in taxes
while 59.6% (n=56) feel they waste some of it and 22.3%
(n=21) feel they don't waste very much at all.

When asked

how often they thought they could trust the School Board to
do what is right, 63.3%, (n=63) reported that they could be
trusted most of the time; 8 .1% (n=8) , just about always;
and 28.3%, (n=28) , only some of the time.

The majority of

the respondents (n=54; 61.4%) felt that the School Board
was run for the benefit of all the people while only 38.6%,

119
(n=34) felt it was run by a few big interests looking out
for themselves.

Almost three fourths of the respondents

(73.9%; n=65) felt that the people running the School Board
are smart people while 26.1% (n=23) felt that quite a few of
them don't seem to know what they're doing.

Finally, 70.8%

(n=68) believe that hardly any of the people running the
School Board are crooked while 25% (n=24) believe not very
many are crooked and 4.2% (n=4), that quite a few of them
are crooked.
While the trust in government items point to a
sample that is generally trusting of the School Board, a
closer examination of the items points up some interesting
trends.

On two of the items, wasting tax dollars and being

trusted, the majority of the sample selected the middle
response, that the School Board wastes some of the money
paid in taxes (59.6%) and can be trusted most of the time
(63.67©).

These two responses show a less than total trust

of the School Board.

Further 23% of the sample thought that

some of the people running the School Board were crooked,
again a less than total trust response.
with no middle response,

On the two items

substantial minorities are found to

be distrustful of the School Board:

38.67© believe the

School Board is run by a few big interests looking out for
themselves and 26.17© believe that quite a few of the the
people on the School Board don't seem to know what they are
doing.

It would appear, then, that the Shiawassee sample is

not totally enamoured of the School Board.

This level of

120
distrust did not bode well for the issue under study.
Awareness and Information
Less than half of the respondents were aware of any
upcoming special county-wide elections (43.6%; n = 4 4 ) .

The

directed inquiry as to whether the Shiawassee County Inter­
mediate School District had called any special elections
netted three respondents who answered in the affirmative.
Only fifteen respondents

(34.17» of those aware unaided) knew

that the election had been called by the Intermediate School
District while nine (20.5%,) thought it had been called by
the County Attorney and one (2.3%,) by someone else.

Of

those aware that an election had been called, 59.6%, (n=28)
knew that the election concerned a millage for Area Voca­
tional Technical Education while 6.4% (n=3) thought it was
a millage for special education and 8.5%, (n=4) each for a
nursing home or something else.

A large percent also cor­

rectly identified the amount of millage as being between
one and two mills (44.7%; n=21) while six respondents
(1 2 .8%) thought it was more than two mills and three respon­
dents (6.4%,), less than one mill.
One of the most disconcerting findings of the study
was this low level of overall awareness that an election had
been called (43.6%).

Coupled to this was the low level of

correct information held by those respondents who were aware
of the election.

Only twelve respondents could correctly

identify who had called the election, the issue being voted

121
upon and the amount of millage requested.
11% of the overall sample.

This represents

The turnout for the election

was 127o so perhaps only those holding correct information
voted.

This is unlikely.

That means that voters were going

to the polls to vote without complete information, a
not-too-pleasant thought in a republican democracy.

2

Media Use
The majority of respondents to the survey (n=83;
76.2%) report reading a newspaper regularly.

Many of these

subscribe to and read the local newspaper, the Owosso Argus
Pre s s .

Second in both readership and subscription was the

Detroit Free Press followed by the Flint Journal, the
Lansing State Journal and the Detroit N e w s . There was
little readership and subscription to the local weekly
newspapers (see Table 3).
Most respondents spend time each day listening to
the radio (see Table 4).

However, little of that listening

involves WOAP, the local radio station.

Only nineteen

respondents (2 1 .8% of radio listeners) report that they
usually listen to WOAP.

Twenty-three respondents (26.4%,)

listen to stations from Lansing, nineteen (21.8%,) to Detroit
stations, sixteen (18.4%0) to Flint-Saginaw stations and six
(6.9%) to other Michigan stations.

Listenership to WOAP is

fairly evenly distributed between AM (n=5; 26.3%) and FM
(n=7; 36.8%,) with five respondents (26.3%,) listening to
both.

122
TABLE 3
NEWSPAPER READERSHIP AND SUBSCRIPTION

Newspaper

Readers

Subscribers

Owosso Argus Press

58

54

Detroit Free Press

28

16

Flint Journal

14

11

Lansing State Journal

10

9

Detroit News

7

5

Durand Express

5

2

Shiawassee Journal

5

2

Out-of-State Papers

3

1

Jackson Citizen Patriot

1

0

TABLE 4
RADIO LISTENERSHIP

Hours

Number

Percent

None

22

2 0 .8%

One

24

2 2 .6%

Two

13

12.3%

8

7.6%

Four

13

1 2 .3%

Five

6

5.7%

Six

7

6 .6%

13

12.3%

106

1 0 0 .2%

Three

Seven+
Total

It was not surprising that little more than half of
the respondents read the Owosso Argus Press and 17.4%
listened to the local radio outlet, WOAP.

Shiawassee

123
County is not media poor in its truest sense.

However,

so

much media from other markets reaches into the area that
there is little need to attend to the local media.

Add i ­

tionally, given that a sizeable portion of the employed
work outside the county, there may be little sense of com­
munity associated with the county.

If you work in Lansing,

you read the Lansing newspaper and listen to Lansing radio
stations.
Source of Information
Thirty-five respondents (32.17„) reported having
heard something about Area Vocational Technical Education
"since last Monday".

Many of these had received information

form a variety of sources rather than a single source.
Twenty-three respondents had read about Area Vocational
Technical Education in the newspaper while seven had heard
of it on the radio.

Of the non-media sources, eleven had

received information from a friend, nine from a flyer or
brochure, seven at a meeting and two from a poster.
These results follow from the previous o n e s .

With

so little attention paid to local Shiawassee County media,
it is not surprising that so few had heard anything about
Area Vocational Technical Education.

This is also in line

with the low levels of awareness noted in the County.
Voting
Few respondents (n=14; 12.8%) indicated having not
voted in a general election in the past two years (1976-

124
1978) while forty-five respondents (41.370) voted in all
general elections

(see Table 5).

Turnout among respondents

for school elections was poorer with twenty-eight respon­
dents (25.7%) voting in none in the past two years and only
thirty-seven (33.9%) voting in all of them.

It should be

noted that there had been one general election and one
school election during the time period under study which
raises the vote levels of 63 (57.8%) for the general
election and 53 (48.6%) for the school election.
TABLE 5
TIMES VOTED IN 1976-1978

Number of Times

General

School

N

Percent

N

Percent

None

14

1 2 .8%

28

25.7%

One

18

16.5%

16

14.7%

Two

13

11.9%

9

8 .0%

Three

1

.9%

1

.9%

One or Two

0

---

5

4.6%

Two or Three

3

2 .8%

0

---

Most All

3

2 .8%

0

---

45

41.3%

37

All

Twenty respondents

33.9%

(18.3%) indicated they had never

voted in a school millage while sixty respondents (55%) had
3
voted in one within the past year.
Six respondents (5.5%)

125
had voted in a school millage two years ago and three
respondents

(2 .8%) each, three years and more than three

years ago.

Of those respondents who would reveal their

previous vote in a millage election, forty-three (6 6 .2%)
had voted in favor and twenty-two (33.9%), against.
The majority of respondents indicated they would
vote in the upcoming election (n=82; 75.2%).

Seven (6.4%)

said they would not vote and ten (9.2%) were not sure.
Of those who reported they would vote in the election,
41.5% (n=34) indicated they would definitely vote in favor
of the millage.

Twenty respondents (23.4%) said the chances

were greater than fifty but less than one hundred while
nine (11%) said the chances were fifty and eight (9.8%) ,
less than fifty,

that they would cast a favorable vote.

The sample reports a very high level of voting in
past elections.

Only 12.87. indicated they did not vote in

a general election in the previous two y e a r s .

Overall

national turnout for the 1976 Presidential election was only
A
54% of all eligible citizens.
This would mean that voters
in Shiawassee County are much more likely to vote than the
general population.

This is not necessarily the case.^

What is more likely is that respondents were providing so ­
cially accepted responses.

This is even more clear when one

considers the fact that 75% of the respondents indicated they
would vote in the upcoming millage election.

Either the

sample voted at a much higher rate than the general popula­
tion, as overall turnout was only 12%, or, socially accepted

126
responses were being provided.
Sample Representativeness
Shiawassee County is made up of eight school dis­
tricts, only six of which were represented in the survey
(see Table 1).

No completions were achieved in either

Laingsburg or New Lothrop.

Of the remaining six districts,

none is present in the sample proportionate to its size in
the population.

The districts of Byron and Owosso are

under-represented while Corunna, Durand, Morrice and Perry
are over-represented.

Even though the two districts not

included are relatively small, the results may not be generalizable to the county as a whole because of their absence.
The survey respondents do cover a broad range of
demographic groups.
accurately, however.

Several groups are not represented
Census data reports that the county

is 62.4% rural^ while of the respondents, only one reports
farming as an occupation.

While it is entirely possible

that many of the respondents live in rural settings and are
employed in urban are a s , it is unlikely that this would
explain a difference of 60.7%.

The sample, therefore, under­

represents the rural community in Shiawassee County, posing
potential problems for the generalizability of the results.
The sample over-represents those age 65 and older.
Census data reports that 7.7% of the population is 65 years
of age and older^ while 12.27. of the sample is in that age
group.

There are several possible explanations for this

127
discrepancy.

First, the sample was drawn from voter regis­

tration lists.

It has been found that younger members of

the population vote in fewer numbers than those who are
g
older.
This lower voting turnout may also be reflected in
a lower rate of voter registration.

Thus, the sampling

frame may have contained a larger share of older voters than
is found in the population as a whole.

It is also possible

that the older respondents were more willing to complete the
questionnaire so that it was not the sample that was unrep­
resentative but rather the in-tab sample.

Again, this dis­

crepancy could bias the results.
The final group that was under-represented was the
male portion of the population.

Comprising 49% of the

9

county population , males make up only 38.1% of the in-tab
sample.

The reasons for this are unclear.

One plausible

explanation for this is that males were less likely to
be willing to complete the survey than females, as males and
females were fairly equally represented in the sample.
Finally, the sample is more likely to subscribe to
the Owosso Argus Press (53.2%) than the population as a
whole (21.5%).

It is less likely to report listening to

WOAP (21.8%,) than has been found in previous studies (30%,).
The sample, then, is not truly representative of the
population as a whole.

While most demographic groups are

represented in the sample, the few discrepancies may result
in unknown biases in the results.

128
Hypotheses
The Derivative Model
The study was undertaken primarily to test the
derivative model described in Chapter I.

This model pr o ­

posed that differences would exist as to awareness, level of
information and information sources based on differing
levels of trust in government and political efficacy.

To

test the model, trust in government and political efficacy
variables were created and then the sample was distributed
into the cells of the model based on their scores on the two
variables

(see Table 6).
TABLE 6
SAMPLE DISTRIBUTION ON TRUST AND EFFICACY

Political Efficacy

Trust in Government
High

Low

56
*
(51.4%)

12
(11.0%)

68
(62.4%)

Low

11
(10.1%)

30
(27.5%)

41
(37.6%)

Total

67
(61.5%)

42
(38.5%)

109

High

Total

X 2= 3 3 .2928; df=1; p<.001.
Percentages are percent of total (109).
Slightly more than half (51.4%) of the respondents
were found to be high in both trust in government and a
sense of political efficacy while slightly more than onefourth (27.57o) scored low on both variables.

An almost

129
equal number of respondents were high in efficacy but low
in trust or vice versa.

It is also interesting to note

that an almost equal number of respondents were high on
efficacy (n=68) or high on trust (n=67).

The relationship

between the two variables, as represented by the distri­
bution, is highly significant.
The statistical significance of the overall distri­
bution of respondents on trust in government and political
efficacy, addressed specifically to the School Board and
school politics, coincides with the work of F i n i f t e r ^ and
Milbrath and Goel

12

(see Table 6).

The two dimensions

do differentiate between respondents with the distribution
being significantly different from chance.

As expected, the

greatest number of respondents scored high on both var­
iables .

It had been hoped that the high trust-low efficacy

group would be second.
pative population.

This would have indicated a partici­

The low trust-low efficacy group, ho w ­

ever, was second largest.

The stating of the trust and

efficacy items to relate specifically to the School Board
and school politics may have been the cause of the fact
that more than 25% of the respondents feel both distrustful
and inefficacious.

As Carter

13

and Hamilton and Cohen

14

discovered, these feelings are often present in school
referenda and can account for low turnouts.

The size of

the protest vote would be expected to be small as the low
trust-high efficacy gorup was quite small.

Given the sub­

stantial size of the high trust-high efficacy group, the

130
turnout would have been expected to be large.

The fact that

it was not points to the fact that other factors were at
work in this particular election that disrupted the normal
or expected impact of alienation.

In this case, the media-

poor environment may have been a mediating factor.

The

results of the study seem to point that way.
The first set of hypotheses examined the level of
awareness among each of the four groups.

The main hypo­

thesis predicted that there would be significantly different
levels of awareness among the four groups.
the case (see Table 7).

This was indeed

The greatest level of awareness was

found in group three, the group scoring high in trust and
low in efficacy where nearly three-fourths of the respon­
dents (72.7%) were aware that an election had been called.
Half of those in group two, those low in trust but high in
efficacy, were aware while 41.7% of group four, those low in
both, were aware.

The lowest level of awareness was found

in group one, the group high in both trust and efficacy.
The three sub-hypotheses predicted significant dif­
ferences among the four groups when compared individually.
No support was found for these three hypotheses.

The first

sub-hypothesis had predicted that the proportion aware in
group one would be greater than the proportion aware in
group two.

The data show that the opposite is true.

The

proportion aware in group two was larger, though not signif­
icantly larger, than the proportion aware in group one.

The

second sub-hypothesis had predicted that the proportion aware

131
TABLE 7
AWARENESS BY ATTITUDE GROUP
Political Efficacy

Trust in Government
High

Low

Total

High

20
(35.7%)*

6
(50.0%)*

26
(54.2%)**

Low

8
(72.7%)*

14
(46.7%)*

22
(45.8%)**

28
(58.3%)**

20
(41.7%)**

48
(44.0%)*

Total

X 2= 8 .0655; df=1; p<.01.
^Percentages are percent of total in cell aware. See
Table 6.
**Percents are percent of total in Table.
in group three would be greater than the proportion aware
in group four.

The data show that this is indeed the case,

that the proportion aware in group three was larger than
the proportion aware in group four.

With a Z score of

1.508, the difference approached, but did not reach, signif­
icance.^"®

The third sub-hypothesis predicted that the com­

bined proportion aware in groups one and two would be
greater than the combined proportion aware in groups three
and four.

As with the first sub-hypothesis, the opposite

was the case.

The combined proportion aware for groups

one and two was only 38.2%, while 53.6 % of groups three and
four combined were aware.
The overall awareness level was quite low.

Only 44%,

of the respondents were aware that an election had been
called (see Table 7).

This may be partially explained by

132
the fact that only 53.2 % of the sample subscribed to the
local paper, the Owosso Argus P r e s s , and 21.8% of those who
listen to the radio listen to the local radio outlet,
WOAP-AM-FM.

These two channels provided most of the readily

available information concerning the election.
respondents did not attend to the local media,
about the election may not have been received.

If the
information
The distri­

bution of awareness according to the two dimensions of
alienation is highly significant.

Those high on both

dimensions comprise 41.7% of those aware and those low on
both, 29.2%.

Those high in trust and low in efficacy make

up 16.7% of those aware while those low in trust and high
in efficacy represent 12.5% of those aware.

The distri­

bution supports the contention that those low in alienation
would exhibit a greater interest in, and therefore awareness
of, political activity. ^

The proposed withdrawal from

politics is not noted in the low trust-low efficacy g r o u p . ^
It could be that this withdrawal is from party politics as
suggested by Campbell

17

and not a significant factor in

school elections where alienation may result instead in pro.
..
18
test voting.
The study had hypothesized certain relationships
based on awareness levels but found that these did not exist
in the data.

Rather than having the highest proportionate

level of awareness, the high trust-high efficacy group had
the lowest.

The highest proportionate level of awareness

was found in the low efficacy-high trust group.

This group

133
been found to be generally supportive of the system

19

and

while inefficacious feelings generally relate to apathy, in
the present case the high levels of trust may be the determining factor.

20

It was expected that a large proportion

of those high in efficacy and low in trust would be aware
however the proportion was smaller than expected.
Awareness of the election is related to alienation
although not in the expected w a y s .

Differences exist but

these differences do not reflect the research to date.
setting, again, may be the cause.
little interest as found here.
cases.

The

School elections generate

They may indeed be deviant

The limited availability of information via tele­

vision may also have influenced the results.

Other factors

may be affecting information-seeking behaviors, factors that
may exert a greater influence upon potential voters than
alienation.
The second set of hypotheses focused on the level of
correct information held by the four groups.

Table 8 pre­

sents the mean level of correct information held by each
group.

The scores on correct information ranged from zero

to three.

The highest mean level of correct information was

1.833 for the low trust-high efficacy group followed by 1.57
for the high trust-high efficacy group.

The lowest level of

recall was 1.071 for the low trust-low efficacy group.
The main hypothesis predicted that significantly dif­
ferent levels of correct information would exist between the
four groups.

This was not the case.

Table 9 presents the

134
TABLE 8

LEVEL OF CORRECT INFORMATION

Group

Mean

Standard
Deviation

Total

High Trust-High Efficacy

1.570

1.165

21

Low Trust-High Efficacy

1.833

1.169

6

High Trust-Low Efficacy

1.125

1.246

8

Low Trust-Low Efficacy

1.071

1.141

14

results of an one-way analysis of variance performed on the
level of correct information vari a b l e .

The within group

sums of squares was substantially larger than the between
group.

More variance within the scores is left unexplained

than is explained by the trust in government-political effi­
cacy divisions.
nificant.

The F ratio reflects this and is not sig­

Difference

three sub-hypotheses.

of means tests were used to test the
No significant differences were found.
TABLE 9

ANALYSIS OF VARIANCE--LEVEL
OF CORRECT INFORMATION

Source

D.F.

Sum of
Squares

3

3.86

Within Groups

45

61.78

Total

48

65.64

Between Groups

Mean
Squares

F ratio

1.12867

.9379 (n.s.)

1.3728

135
The model proposed that correctness of information
would be related to alienation.

Again, this did not occur.

Overall levels of correct information were low.
trust-high

The low

efficacy group had the highest mean.

This was

unexpected as low trust is usually associated with use of
television as the primary source of political information
and, in Shiawassee County, no local television exists.

21

The

high trust-high efficacy group did not show the level of
information expected.

This group is generally thought of

as being supportive of school referenda and to hold correct
information as a result of this support.

In this case,

information regarding the issue may have been so sparse that
even members of this group had difficulty locating it.

The

only case where the result was truly expected from the
literature was the low level of correct information for the
low trust-low efficacy group.

Past research had shown that

those in this category avoid political information
appears to be substantiated by the data.

23

22

which

The similarlity

in means and standard deviations between the four groups is
the reason for the lack of significant differences noted
between the groups (see Table 8).
The third hypothesis focused on a comparison of
media sources and personal sources across all four groups.
Table 10 details the number of media and personal sources
for each group.

The sample, as a whole, reports hearing

about Area Vocational Technical Education from an almost
equal number of personal and media sources.

Differences do

136
TABLE 10
SOURCE OF INFORMATION BY GROUP

Source
Total
Media Personal

Group
I

High Trust-High Efficacy
(56)

II

Per
Person

12

14

26

.46

Low Trust-High Efficacy
(12)

2

6

8

.67

III

High Trust-Low Efficacy
(11)

4

0

4

.36

IV

Low Trust-Low Efficacy
(30)

12

9

21

.70

TOTAL

30

29

59

exist between the groups.

Those high on both variables

consulted the most number of sources followed closely by
those low on both.
all sources.

The two middle groups consult few over­

This pattern changes when the number of

sources per person ratio is examined.

This shows that the

group low on both variables consulted the most number of
sourcer per person (.7) followed closely by those low in
trust but high in efficacy (.67).

Those high in both

variables ranked third and those high in trust but low in
efficacy, fourth.
The model proposed that groups one and three would
rely on media and groups two and four on personal sources.
This holds for groups two (personal) and three (media) but
not for groups one and f o u r .

Group one mentions more

137
personal sources and group four, more media

sources.

The

third hypothesis had predicted that groups one and three
combined would rely on media sources to a significantly
greater extent than groups two and four combined.

The

proportion of media sources for groups one and three is
.5333 while for groups two and four the proportion is .4827.
The first is greater than the second, but the difference
is not significant.
The prediction that

those high in both trust in

government and political efficacy and those high in trust
but low in efficacy would report media sources in a signif­
icantly greater proportion than those in the remaining two
groups had been based on the findings that high trust and
efficacy are related to the use of newspapers as a source
of political i n f o r m a t i o n . ^
out.

This study did not bear this

The high trust-high efficacy group reported slightly

more personal sources and the low trust-low efficacy group
more media sources.

The remaining two groups follow the

hypothesized direction.

It would appear that in a minimal

mass mediated information election, the relationship between
efficacy and trust and type of information source is not
the same as that between the two components of alienation
and source in a more information-rich setting.

Those high

in both trust and efficacy represent those most involved
and active in the political system.

In a standard general

election, information about candidates and issues is readily
available in the mass media so no further information-seeking

138
behavior is necessary to be informed.

In the Shiawassee

County situation the mass media may not have provided
sufficient information for the highly involved group
necessitating consultation of personal sources.

The low

trust-low efficacy group poses more of a problem in
explaining.

It would appear that in Shiawassee County in

this particular situation, a high degree of alienation, as
determined by high levels of distrust and inefficacy, does
not lead to withdrawal from politics
political information,

26

25

and an avoidance of

but rather to an observer status--

one who watches but does not participate.
The derivative model did not hold up to empirical
examination as well as had been predicted.

In only one

instance was a significant relationship discovered.

There

are several possible explanations for this occurence.
The derivative model assumed that the original
Milbrath and Goel model was correct in its characteristics
of levels of participation based on differing levels of
trust in government and political efficacy.

27

Using their

descriptions as a guide, for example, that those high on
both dimensions would be "active allegiant" and have "con­
ventional participation" while those low on both would
"withdraw from politics",

the literature was searched to

determine how these characteristics related to demographics
and media use, among others.

Additionally the literature

was perused to determine how political efficacy and trust in
government related to information variables and to school

139
referenda settings.

Melding all this information, the

derivative model was developed.
The Milbrath and Goel model, however, had never
been tested in its entirety in a single study.

The assign­

ment of participation characteristics to each cell was done
on the basis of a perusal of the literature as was done for
the derivative model.

It is possible that, in a single

setting with a single sample,
may not be valid.

the Milbrath and Goel model

If that is the case, the same would po s ­

sibly hold true for its derivative.

Until further research

is undertaken on the Milbrath and Goel model, it is impos­
sible to tell if this is indeed the reason why so many ele­
ments of the derivative model were not substantiated by the
data collected in Shiawassee County.
A second possible explanation for the failure of the
model to explain the behavior of the sample concerns the
utilization of items specifically dealing with the School
Board and School politics as the individual items in the
trust in government and political efficacy scales.

The

Milbrath and Goel m o d e l , and the work of Finifter upon which
it is based, both used the standard items for both of these
scales.

It may be that the model holds for the less defined

"government" items and not for more specific items as were
used in tthe present study.

The use of the more specific

items had been suggested by Aberbach and Walker, Citrin,
Dean and McCloksy.

28

The specific political efficacy items

had also been used in the work of Carter.

29

There appeared

140
to be sufficient justification for their adoption in the
Shiawassee study.
The development of the derivative model cell charac­
teristics was based on published research.

Much of this

research had been done using more general trust and ef­
ficacy items.

It could be that the model of information

would hold if those items were used rather than the more
specific ones. Until more research is accomplished using
specific items, the final determination of whether this was
the cause for the failure of the model will not be known.
A third possible explanation for the inability of
the model to predict behavior has to do with the setting of
the study.

This included both a low interest election and

a low information • area.

The development of the Milbrath

and Goel model had been accomplished using data from general
elections rather than school referenda.

It could be that

the model holds in the more general situation but not in the
specific one.

If that is the case, the derivative model may

hold in a general election situation where it did not hold
in the school election.
Complicating this is the fact that there was so
little interest in the millage election.

The fact that so

few were aware or had correct information regarding the
issue may have biased the results.

It may not be the model

that is at fault, but rather that the Shiawassee County m i l ­
lage election was a deviant situation even for a school
election.

It may also be that the problem lies not with the

141
rationale underlying the model but rather with the infor­
mation setting.

The county receives a plethora a media

30
from the surrounding counties.

It may be impossible for

the two local media outlets to compete for the attention of
county residents.

In a more normal media setting, the model

may be a better predictor.
Given the results from the Shiawassee County study,
the derivative model should be redesigned to resemble
Figure 3.

Support for such a model, however, is not found

in the literature.
fied.

Few of the assignments could be justi­

Only those marked with an asterisk would be under­

girded by previous research results.

It is much more likely

that the setting caused the failure of the model and that
the model as proposed would be supported in a more tradi­
tional situation.
Trust in Government
High
High

Political
Efficacy

Low Awareness
*Correct Information
Information from
Personal Sources
jjigh Awareness
Incorrect Information
i n f o r m a t i o n from
Media

Figure 3:

Low
Low Awareness
Correct Infor­
mation
i n f o r m a t i o n from
Personal Sources
*Low Awareness
i n c o r r e c t Infor­
mation
Information from
Media

Shiawassee County Derivative Model

142
Campaign Awareness and Information
Hypotheses four and five dealt with increasing
levels of campaign awareness and information as the campaign
progressed.

It was predicted that as the election neared,

an increasing proportion of respondents would, first, be
aware than an election had been called and, second, hold a
high level of correct information regarding the election.
The results were mixed at best (see Table 11).

First, the

awareness level did not increase as predicted.

Week two

(.5385) was larger than week one (.1667), but not signif­
icantly so.

Week two, however, was also larger than week

three (.4286) and week four (.3), with three being larger
than four as well.

For week five the pattern shifts with

five (.3333) slightly larger than week four, but equal to
week six.

Week six was larger than week seven (.2), as

was week eight (.2857), the predicted direction but not a
significant increase.

Week nine (.75) was significantly

greater than week eight (Z=2.27; p<.05).

A major drop in

awareness took place in week ten with only 28.67> of those
contacted aware that an election had been called for the
following Monday.
The data on correctness of information were even
more disappointing.

Only twelve respondents

(117o of 109)

could correctly identify who had called the election, the
issue being voted on and the amount of the millage.
five of the weeks

For

(one, five, seven, eight and ten) not one

respondent was found to have correct information regarding

143
TABLE 11

AWARENESS AND INFORMATION BY WEEK

Week

Number
Contacted

Awareness

Information

6

1

0

2--April 9-15

13

7

2

3--April 16-22

14

6

2

4 t -April 23-29

10

3

1

3

1

0

6--May 7-13

12

4

2

7--May 14-20

10

2

0

8--May 21-27

7

2

0

25

18

5

7

2

0

44
(.4142)

12
(.2727)

1--April 2-8

5--April 30-May 6

9--May 28-June 3
10--June 4-10
Total

107a

aTwo of the questionnaires were not dated so were not
included in this analysis.
the election.

This included the final week before the

election (week t e n ) .

The hypothesis had predicted that the

proportion of those holding correct information would in­
crease significantly as the campaign progressed.
not the case.

This was

The proportion holding correct information

during week two (.2857) was larger than for week one but
this was not significant.

Week three was larger than week

two but, again, was not significantly so.
equal to week four.

Week three was

During the fifth week, no one was

144
found who had correct information.
all previous weeks, however,
a trend.

Week six was larger than

this was not the beginning of

For weeks seven and eight no one held correct

information, while for week nine only 27.87a of those aware
could correctly identify all three elements.

During the

final week before the election no one, once again, held
correct information.
The fourth hypothesis had assumed that as the date
of the election neared, an increasing proportion of regis­
tered voters would be aware that an election had been called.
This was based on the fact that the opportunities to be
exposed to campaign information would have increased as the
amount of information available in the community increased.
As the data in Table 11 show, this did not occur.

No pat­

tern of increase or decrease in awareness levels is dis­
cernible (see Figure 4).
week two and week nine.

The awareness proportions peak at
It would appear that even less than

one week before the election it was easy to have missed any
and all information regarding the upcoming election, or to
have seen but not recalled any information because of lack
of interest in the issue.
Hypothesis five had assumed that, as with awareness,
the proportion of those holding correct information would
increase as the election date neared.
not occur (see Table 11 and Figure 4).

Once again, this did
There is even less

of a discernible pattern with the correctness of

1

2

3

4

5

______ :Aware;
Figure 4:
information data.

6

7

8

9

10

Week

----- :Correct Information

Awareness and Information by Week
In fact, the level of correct information

does not follow the pattern of awareness.

The fact that

this was a low interest election is spotlighted by the data.
Few respondents were aware that the election had been called
and fewer still could correctly identify who had called it,
the issue to be decided and the amount of the millage being
requested.
It should be noted that the assumption of increasing
awareness and information may not be valid.

Similar analyses

need to be undertaken in local media-rich areas and general
election settings to determine if the assumption or the
setting in this instance caused the failure of the

146
hypothesis to predict behavior.
Hypothesis seven predicted that respondents living
in the Owosso and Corunna school districts would receive
their information regarding the election from the newspaper
to a greater extent than respondents in the outlying dis­
tricts as the major local daily was based in Owosso.

It

was first necessary to identify those respondents who had
heard anything about Area Vocational Technical Education
"since last Monday" (see Table 12).

The majority of the

respondents (687,) had heard nothing about Area Vocational
Technical Education recently.

The sample is almost evenly

split between Owosso-Corunna and the other six districts
combined, however, a larger proportion of respondents in
Owosso-Corunna (39%) had acquired information recently
(others:

25%).

This distribution is not statistically

significant.
TABLE 12
INFORMATION ACQUISITION BY SCHOOL DISTRICT

"Heard anything
since last Monday?"
Yes

No

Total

Row Percent

School District
Owosso-Corunna
Others
21
(.60)*
(.39)**

14
(.40)
(.25)

33
(.45)
(.61)

41
(.55)
(.75)

54
(.50)

55
(.50)

Column Percent

Total

35
(.32)
74
(.68)
109

147
Table 13 presents the distribution of newspaper and
non-newspaper sources by school district.

The newspaper

garnered the most number of mentions as the source of
recent information regarding the issue.

Respondents living

in Owosso-Corunna were more likely to have heard about Area
Vocational Technical Education from the newspaper (76.2%)
than respondents living in the outlying districts, however
the distribution is not significant.

The main focus of the

hypothesis was not the distribution but the proportion
mentioning newspapers.

The data show that 29.4% (16 of 54)

of all respondents from Owosso-Corunna noted newspapers
compared to 12.77o (7 of 55) of all respondents in the out­
lying districts.

The difference was found to be statis­

tically significant (Z=2.169; p<.05).
TABLE 13
SOURCE BY SCHOOL DISTRICT
School District

Source
Newspaper
Other

Total

Owosso-Corunna

16
(.76)

5
(.24)

21
(.60)

Other Districts

7
(.50)

7
(.50)

14
(.40)

Total

23
(.66)

12
(.34)

35

Owosso and Corunna are the two central school dis­
tricts in the county.

They are also the two largest commu­

nities and are situated abutting each other.

The local

newspaper and radio outlets are located in Owosso.

Given

these facts, it was assumed that respondents in these two
districts would differ in information acquisition from those
in the outlying districts.

The study found that respondents

in Owosso-Corunna were more likely to have heard of Area
Vocational Technical Education.

The centrality of their

location may have been the determining factor.

They were

geographically close to the source providing information
regarding the election and thus were more likely to hear
something about it.

The data also show, as expected, that

respondents in Owosso-Corunna were more likely to have
received information from newspapers

(76.27®) than respon­

dents in the other districts

The lack of signif­

(507®) .

icance here can be explained by a closer examination of
Table 13.

While for Owoss-Corunna, the distribution be ­

tween the two source categories is quite dramatic, for the
outlying districts the distribution is identical to what one
would have expected by chance.

Given this fact, a signif­

icant chi-square was not possible.

The trend toward greater

use of newspapers by the two central districts is present
in the data and the proportion mentioning newspapers is
significantly greater, as stated in the hypothesis.

It

would seem then that proximity to the site of publication
and information acquisition from the newspaper in an area
such as Shiawassee County are related such that as proximity
increased the likelihood of acquisition of information from
that newspaper increased.

149
Favorability
The next six hypotheses examined the effect of
several variables on favorability towards the issue of Area
Vocational Technical Education.
with the age

The first of these dealt

of the respondent.

It was predicted that a

negative relationship would exist between these two var­
iables such that as age increased favorability would de­
crease.

Such a relationship did exist in the data, however,

the relationship was not significant (r=-.17688; t=1.5964;
df=160; p < ,10).
The presence or absence of children in the home has
also been said to be a determining factor in the level of
favorability toward school monetary issues.

In the present

case, fifty-two respondents had children residing with
them and thirty did not.

The mean favorability level for

those families with children present in the home was
79.4615 (standard deviation:

29/0283) while the mean favor­

ability level for those respondents without children living
in the home was 56.667 (standard deviation:

39.8993).

The

difference was highly significant (t=2.9376; df=80; p<.001)
indicating that the presence of children in the home signif­
icantly affects favorability.
Table 14 details mean favorability level by educa­
tional attainment.

It was necessary to combine several of

the education categories due to the extremely small number
of respondents in the cel l s .

Only one respondent indicated

an eighth grade education or less.

This person was added to

150
the some high school group and a new category,
eleventh grade, was created.
category,

first to

For the graduate work

three respondents reported graduate degrees and

two, graduate work.

Therefore, these two categories were

combined.
TABLE 14
FAVORABILITY BY EDUCATION

Education Level

Favorability
Mean
Standard
Deviation

N**
n*

Firs t-Eleventh

50 .65

39 .803

20

28

High School Degree

76 .944

33 .569

18

24

Trade School

70 .000

31,.358

10

13

Some College

77 .105

30,,519

19

24

College Degree

85 .555

.683
20,

9

9

Graduate Work

99 .800

.442

5

5

*Number responding to favorability item.
**Number in sample.
Several factors need to be noted regarding these
two variables.

First, as education level increased the

proportion responding to the favorability item grew from 71%
for the least educated to 100%, for the most.

Second, the

variability of scores decreased as education level increased
as can be seen in the decreasing size of the standard devi­
ation.

Thirdly, except for one deviant case, the mean

favorability level did increase as education increased as

predicted.
Difference of means t-tests were performed between
each education level and the next lower.
nificant and two were not.

Two were sig­

The mean favorability level for

those with a high school degree was significantly greater
than the mean favorability level for those with less than a
high school degree (t=2.1492; df=36; p<.05).

The other

significant difference was found at the opposite end of the
education scale.

Those with graduate work were signif­

icantly more favorable towards the issue than those with a
college degree (t=1.94; df=12; p<.05).

The two where no

significance was found were between those with some college
and those with high school degrees, and those with college
degrees and those with some college.

No t-test was per­

formed between high school degree and trade school as the
mean favorability

score for the lower educational attain­

ment was higher and thus not in the predicted direction.
It was also predicted that favorability would in­
crease as income level increased.
(see Table 15).

This was not the case

The relationship appears to be curvilinear

with those earning less than $10,000 and those earning more
than $30,000 being less favorable towards the issue.

The

only significant difference of means occured between those
earning less than $10,000 a year and those earning between
$10,000 and $20,000 a year with the mean of the latter group
being significantly greater than the mean of the former
(t=1.7366; df=32; p<.05).

The other two comparisons were

152
not in the predicted direction, i.e., the mean favorability
level decreased as income increased.
TABLE 15
FAVORABILITY BY INCOME LEVEL

Income Level

Favorability
Mean
Standard
Deviation

n*

Less than $10,000

62.307

34.5576

13

18

Between $10,000
and $20,000

82.149

24.8783

21

26

Between $20,000
and $30,000

72.885

36.6653

26

30

More than $30,000

68.167

37.5290

18

21

*Number responding to favorability item.
**Number in sample.
It had been predicted that as occupational status
increased,

favorability would increase significantly.

was found to be partly the case (see Table 16).

This

The occu­

pational variable originally had five categories.

The cate­

gory "farmer" was dropped from the analysis as there was
only one respondent identifying herself as such.

The m a n ­

ager and professional categories were combined due to the
low number in each (four in the former and nine in the
31
latter).

The two comparisons were not significant although

the first, blue collar with white collar, approached signif­
icance (t=1.48; df=33; p < .10).
The final favorability item examined that variable

153
TABLE 16
FAVORABILITY BY OCCUPATION

Occupation

Favorability
Mean
Standard
Deviation

N**
n*

Blue Collar

64.368

39.3604

19

25

White Collar

80.000

19.4963

16

19

Manager/Professional

81.077

29.6800

13

16

*Number responding to favorability item.
**Number in sample.
in light of a previous vote for a school millage.

The hypo­

thesis predicted that those who had cast a previous vote in
favor of a school millage would be more favorable towards
the issue than someone who had cast a prior negative vote.
The data support the contention.

Thirty-nine respondents

reported a previous positive vote on a school millage.
Their mean favorability level on the current issue was
76.8947 (standard deviation:

30.9862).

reported a prior negative vote.

Seventeen respondents

Their mean favorability

level in the present case was 43.7059 (standard deviation:
40.3).

The difference was highly significant (t=2.9479;

df=54; p < .01).
The researchers into school referenda voting corre­
lates have uncovered a variety of factors relating to the
likelihood of a favorable vote.

In some instances the

results have been mixed with a characteristic being posi­
tively related to a favorable vote in some studies and in

154
others, negatively related.

A variety of these factors

were tested in Shiawassee County.
Carter had found that the most favorable voter was
under the age of fifty-one

32

, while Hamilton and Cohen re-

ported that support decreased as age increased.

33

This

trend was present in the Shiawassee County data.

Eighty-

one respondents provided answers to both ite m s .

Overall

mean age was 44.07 years while mean favorability was 71.96.
The Pearson product moment correlation between the two was
negative indicating an inverse relationship between the two
however the result was not significant.

This study did not

compare age with a favorable vote but rather with a selfreport of the likelihood of casting a favorable vote.

It is

possible that were age compared to vote in the present
study, a significant relationship would be found.

This, in

fact, did not occur as shall be seen shortly.
Carter had also found that parents with children in
school were most supportive of school referenda.
and Cohen found parents to be more supportive

35

34

Hamilton

, as did

Kowalski, who determined that those with children in the
public schools in grades ten and under would be more sup­
portive than non-parents or parents with no children in
O fi

school.

This finding was tested in Shiawassee County and

was found to be true.

The mean favorability level of those

with children present in the home was almost 23 points higher
than the mean for those without children present in the h o m e .
It would appear that parents form a special interest group

155
when it comes to schools such that they register support
for the schools across a wide variety of financial issues,
including,

in the present case, an increase in property

assessment to cover the cost of building and maintaining a
centralized center for the vocational and technical edu­
cation of high school students.
Hamilton and Cohen had found that the highly educated were more supportive of school referenda.
and Bass supported this view.

38

37

Alexander

It was hypothesized for

this study, then, that favorability would increase as level
of education increased.

Favorability towards the issue did

increase with one notable exception.

Those completing

trade school after high school were less favorable towards
the issue than those with a high school degree.
be situation-specific.

This may

The millage request was for funds

to construct and maintain a centralized vocational technical
center for high school students.

It is possible that those

who gained their technical skills in a trade school setting
may be opposed to this type of education on the high school
level, believing that a trade school is the more appropriate
place for such training.

Given the results of the study, it

can be said that increased education is related to increased
favorability.

Even though several of the relationships

were not significant, the limited size of the sample may
have influenced th i s , the trends in the data point to the
validity of such a claim.
Several studies had also examined the relationship of

156
income to favorability towards the school referenda.
Hamilton and Cohen found that respondents in the upper in­
come brackets were twice as supportive as those in lower
brackets . ^

Alexander and Bass support this finding.

Wilson and Banfeild, however,
voters

found that middle income

(homeowners) were much more likely to vote against

school proposals than the very poor or the very rich.

41

The results in Shiawassee County do not follow either of
these patterns.

In the present situation, those earning

the least and those earning the most were both less fa­
vorable towards the issue than respondents in the middle
two income brackets.

Those earning less than $10,000 a year

were, for the most part, older and retired, living on fixed
incomes.

Their opposition to the issue may have been

related to this fact.

It may be their age and life situation

as much as their income that led to the low mean favorability
rating.

Those respondents who were more affluent yet not

favorably inclined are more difficult to explain.

Once

again, however, it may not be income level but rather other
factors that are reflected in the lower mean.

Few of the

respondents were located in the managerial and professional
categories.

This means that some of those located in the

blue and white collar categories may be included in the
upper income bracket.

These job categories are, tra­

ditionally, much less inclined to vote in favor of school
referenda.

This being the case, it is possible that the

presence of these respondents in the upper income category

157
lowered the mean favorability level for that income level.
Another possible explanation for the lower mean favor­
ability level for the upper income group has to do with the
specific issue being considered.

They may believe that

money should be spent on the schools but not for vocational
education.

This would lead to an unfavorable rating.

This

explanation is not as plausible, however, as the occupationincome relationship, as will be seen.
Researchers have found that favorability towards
school monetary issues is related to occupational status.
The results, however, have not been consistent.

Carter, on

the one hand, found that the most favorable voter had an
occupation as a skilled worker or in clerical and sales
areas while the most unfavorable voter had a professional or
technical occupation.

42

Hamilton and Cohen, on the other

hand, found that professionals were more favorably inclined
than businessmen and far more supportive than those in
43
manual occupations.

Alexander and Bass found that mana­

gerial and professional occupations were associated with
greater election support

44

while Kowalski ranked occupations

from most to least favorable: professional, managerial,
white collar, manual and farmer.

45

The data from Shiawassee

County support the latter researchers.

Those employed in

professional/managerial positions were most favorable to­
wards the issue followed closely by white collar and
distantly by blue collar.

The difference in means of the

blue and white collar categories was significant, however

158
that between the white collar and professional/managerial
was not.

This second relationship may have been influenced

by the fact that the professional and managerial categories
had been combined due to the limited number of respondents
in each.

If this combining had not taken place the results

would have been different.
The mean favorability level for the manager category
was only 58.75, considerably lower than the white collar
mean of 80.0.

However, only four respondents fell into the

managerial category making the result less stable than for
the other categories.

The mean favorability level for the

professional category was 91.0, substantially higher than
the mean of the managerial category.

While the results as

presented in Table 16 appear to support the work of Kowalski,
Hamilton and Cohen and Alexander and B a s s , there is some
question as to whether this would indeed be the case given
a larger sample.

It could also be that the Shiawassee

County results as they relate

to the managerial class are

deviant.
The final favorability item examined was the rela­
tionship between a past favorable vote in a school millage
election and the likelihood of a favorable vote in the
present situation.

Kowalski had found that past voting

record was a prime determinant of a future vote.
wassee County data support this contention.

46

Shia­

Those who r e ­

ported a prior positive vote on a millage had a mean
favorability level 33 points higher than those reporting a

159
past negative vote.

It would seem that those who are

supportive of the schools are supportive regardless of
issue.

This needs future testing, however.
Profiles of the most favorable and least favorable

voters can be drawn from the Shiawassee County data.

The

most favorable voter is young, has children living in the
home, has at least a college degree, has an income of
between $10,000 and $20,000 a year, has a professional
occupation and has cast a positive vote on a school millage
issue in the past.

The least favorable voter is older, has

no children living in the h o m e , has not completed high
school, earns less than $10,000 a year, is employed in a
blue collar occupation and has voted against a school
millage in the past.
Alienation
The most surprising finding in this study was the
limited role that alienation played in the determination of
either the likelihood of a favorable vote or in the selfreport of the actual vote.

Alienation was defined as a

combination of the scores on trust in government and polit­
ical efficacy.

This dual dimension had been proposed by

48
49
50
, Thompson and Horton
, Stokes
, Aberbach
,
SI
52
Milbrath and Goel
, and Finifter.
Sufficient research

Olsen

47

support seemed to be present to permit the definition of
alienation in the present study as trust and efficacy.
Alienation had been proposed as relating to voting

160
in known ways.

Several researchers,

including Nettler,

Levin and Eden, Campbell, Kraus and Davis, Dean, and Erbe,
found that the alienated withdraw from political partici53
pation, including voting.
McDill and Ridley found that
the alienated,
favor.

54

if they vote, are less likely to vote in

Hamilton found that those low on political effi-

cacy are less likely to vote.

55

Aberbach and Walker found

that the distrustful citizen was likely to cast a protest
vote.

56

Feelings of distrust and inefficacy, according to

Horton and Thompson,

lead to negative voting"*^ while Seeman

reports similar behavior.

58

Studies specific to school referenda have reached
similar conclusions.

Thompson and Horton and Piele and Hall

both determined that the alienated are less likely to vote
but, if they vote, are much more likely to register a negative vote.

59

The work of Hamilton and Cohen, Carter, and

Giles, Gatlin and Cataldo also support this v i e w . ^
Given the extensive body of research supporting the
importance of alienation, and supporting the defining of
alienation as trust in government and political efficacy,
alienation was proposed as a significant negative factor in
both favorability towards the issue and a favorable vote.
In both cases, however, alienation was found to have little
if any influence.
The final two hypotheses,

then, examined the impact

of alienation on favorability and voting.

The first pr o ­

posed that alienation would be an independent predictor of

161
favorability towards the issue.

This was not the case

although alienation was negatively related to favorability
and the multiple regression analysis performed to test the
hypothesis was significant.
Table 17 presents the correlation matrix for the
nine variables involved.

Four of the variables were nega­

tively related to favorability.

The data show that as

alienation and age increase, favorability decreases.

The

weak negative correlations between favorability and p re­
sence of children in the home and home ownership point to
a tendency among the sample for homeowners and those with
children in the home to be less favorable towards the issue.
The strongest remaining correlations with favorability were
someone in the household having taken vocational education
and sex.

Additionally,

sex is strongly associated with

vocational education while children in the home is related
to age and working in the county (negative).

The matrix

shows that no problem with multicolinearity exists in the
data, the largest correlation being

.52010 between sex and

having taken vocational education.
Table 18 presents the beta weights for the variables
involved in the regression.

The best predictor of favora­

bility was someone in the household having taken vocational
education with a beta of .5161 (p<.003) when all other
variables were controlled.

Income is the only other posi­

tive beta in the equation, with a weak, nonsignificant
relationship.

All other variables register negative

TABLE 17
CORRELATION MATRIX

Favorability

Alienation

Children
in Home

Vocational Own
Education Home

Work in
County

Income

Age

Alienation

-.05463

Children
in Home

-.04819

.07896

.45764

-.04450

.20549

-.12105

.05335

.03078

.14827

.08008

-.05870

-.26835

-.05257

-.16523

Income

-.00751

.08804

-.15334

-.13745

-.01934 -.00244

Age

-.31363

.10648

.31184

-.23611

-.14427 -.13753

.04754

Sex

.25835

-.15370

.14862

.52010

-.21049 -.17381

.21594 -.1013

Vocational
Education
Own Home
Work in
County

163
relationships, however none are significant.

It had been

hypothesized that alienation would be a significant inde­
pendent predictor of favorability as evidenced by a signif­
icant beta.

The data show this was not the case.
TABLE 18
BETAS FOR REGRESSIONAL ANALYSIS

Variable

B

Vocational
Education

Beta

Standard Error
of B

F

Sig.*

36.6254

.5162

11.4403

10.2492

.003

-17.0199

-.2614

9.3156

3.3381

.075

Age

-1.1859

-.2373

.7317

2.6270

.112

Sex

-12.0368

-.1133

17.8721

.4536

.504

Children
at Home

-9.5189

-.0671

20.3512

.2188

.642

Work in
County

-6.0103

-.0569

14.6105

.1692

.683

Income

2.2808

.0480

6.1068

.1395

.711

Alienation

-.1637

-.0121

1.7448

.0088

.926

56.6545

9.2929

.004

Own Home

Constant

172.7077

*Significance Level
The prediction equation, however, is significant
(see Table 20).

The eight independent variables explain

31.6% of the variance in favorability (F=2.478; df=8 ;
p<.026).
R

The variable contributing the greatest change in

was vocational education (.20762) followed by age

(.05164) and home ownership (.03350)

(see Table 19).

TABLE 19
SUMMARY TABLE

Variable

F to Enter
or Remove

Sig*

Multiple R

R Square

R Square
Change

Simple R

Alienation

.0088

.926

.0546

.0029

.0029

-.0546

Vocational
Education

10.2492

.003

.4589

.2106

.2076

.4575

Work in
County

.1692

.683

.4619

.2133

.0027

.0301

Income

.1395

.711

.4678

.2188

.0055

.0075

Own Home

3.3381

.075

.5024

.2523

.0335

-.1211

Age

2.6270

.112

.5514

.3040

.0516

-.3136

Children
in Home

.2188

.642

.5553

.3083

.0043

-.0482

Sex

.4536

.504

.5617

.3155

.0072

.2584

Overall F:
Significance:
*Significance.

2.478
.026

165
TABLE 20
REGRESSION SUMMARY TABLE

Multiple R
Standard Deviation

.56174
66.38105

R Square

.31555

Analysis of Variance

Regression
Residual

DF

Sum of
Squares

Mean
Squares

8

87353.197

10919.149

43

189477.111

4406.444

F

s*

2.478

.026

*Significance.
The multiple regression analysis, performed to deter­
mine the role of alienation in the likelihood of casting a
favorable v o t e , had mixed r e s u l t s .

Alienation was nega­

tively correlated with favorability, as expected, as well
as with vocational education, working in the county and sex.
The strongest correlation was with sex followed by age and
income.

Alienation was only fifth in strength of its corre­

lation with favorability.
The regression discovered that alienation was the
weakest predictor of favorability with a beta of only
-.0121, compared to .5162 for vocational education, the best
predictor of the dependent variable.

The only portion of

the hypothesis that found support was that alienation
would be negatively related to favorability.
The fact that someone in the household having taken

166
vocational education was the best predictor of favorability
towards a millage for a centralized Area Vocational Tech­
nical Center is not surprising.

Someone who had experienced

this type of educational training would be most favorably
inclined to its continuation and expansion.

The finding

that home ownership is negatively related to favorability
61
supports the work of Wilson and Banfield.
The negative
relationship of age to favorability was also expected.
The negative beta for sex indicates that women were
more favorable towards the issue as were those who work
outside the county which is also negatively related to the
dependent variable.

The last is not surprising as many of

the technical jobs held by respondents were located outside
the county.

The positive beta for income was expected

however the negative beta for children in the home was not.
The final items seems to say that those
the home are more

with no children in

favorable towards the issue. No support

for this is found in the literature.
The eight

variables included in the regression

explain approximately 327. of the variance in favorability.
While this finding is significant, much of the variability
in the dependent variable remains unexplained by the equa­
tion.

With so much of the variance left unexplained, the

likelihood of predicting whether an individual would be
favorable towards the millage issue based on knowledge of
their scores on the independent variables is small.
A discriminant analysis was performed to test the

167
final hypothesis that alienation would be a significant
negative factor in a favorable vote for the millage.
Fifty-two respondents reported their votes in the June 11
election.

Of those, forty-two reported voting for and ten,

voting against, the millage.

Table 21 reports means and

standard deviations for the two groups and for the sample
as a whole.

Both groups score relatively high on alien­

ation, which had an upward limit of thirty-one, and are
almost equally alienated (difference of means:

.08).

Group

one, the group reporting a yes vote in the election had a
mean favorability level of one hundred
group had a ninety-two,

while the no

indicating little difference between

the two groups on a pre-election report of likelihood of
voting in favor of the issue.

Members of group one are

more likely to work in Shiawassee County, be older, have a
higher income and be male.

Members of group two are more

likely to own their own homes and to have taken, or had
someone in their family take, a vocational education class.
Table 22 presents the standardized discriminant
function coefficients.

As there were only two groups, only

one function was determined.

The variable sex represents

the largest contribution, albeit a negative one, to the
function.

It contributes most to the differentiation of the

two groups along the dimension.

Someone in the home having

taken vocational education ranks second,

followed by the

age of the respondent (negative) and likelihood of a
favorable vote (negative).

Of the nine variables,

168
TABLE 21
MEANS AND STANDARD DEVIATIONS
FOR VOTING GROUPS

Variable

Group I-"Yes"
Mean
S .D .*

Group II-"No"
Mean
S .D .

Mean

22.38

5.74

22.30

4.35

22.37

5.76

1.38

5.39

1.20

.42

1.35

.52

100.00

81.68

92.00

17.51

98.62

73.68

.48

.99

.70

1.25

.52

1.04

1.02

.99

1.50

1.58

1.12

1.13

.45

.74

.40

.52

.44

.70

44.95

15.75

38.90

8.25

43.79

14.74

Income

.86

1.70

.50

.53

.79

1.55

Sex

.45

.74

.20

.42

.40

.69

Alienation
Children
in Home
Favorable
Vote
Vocational
Education
Own Home
Work in
County
Age

*Standard Deviation.
TABLE 22
STANDARDIZED DISCRIMINANT
FUNCTION COEFFICIENTS

Variable

Coefficient

Alienation

-.0478

Children in Home

-.3927

Favorable Vote

-.4475

Vocational Education

.8001

Own Home

-.0073

Work in County

-.3685

Age

-.5483

Income

-.2828

Sex

-.8784

Total
S .D .

169
alienation ranks eighth in relative importance.
Only 13.8% of the variance in the function is
explained by the two groups (see Table 23).

The nine

var­

iables produced a small degree of discrimination between
the two groups as indicated by the large lambda and small
canonical correlation for the function.
TABLE 23
DISCRIMINANT ANALYSIS SUMMARY TABLE
Eigenvalue

Canonical
Correlation

.1597

.3711

Percent
of Total
100

Wilks
Lambda

ChiSquare

DF

Sig

.8623

6.7405

9

.664

In spite of the nonsignificant result, the function
does classify a significant percent of the fifty-two known
cases correctly (see Table 24).
TABLE 24
DISCRIMINANT ANALYSIS PREDICTION RESULTS
Actual Group

Predicted Group Membership
Yes Voters
No Voters

N

Yes Voters

28
(66.7%)

14
(33.3%)

42

No Voters

4
(40.0%)

6
(60.0%)

10

Chi-Square=4.923; df=l; p=.027

170
Sixty-five percent (65.4%) of the known cases were
classified correctly.

A slightly greater percent were

classified correctly for the group of yes voters than for
the group of no v o t e r s .
In the discriminant analysis performed to determine
if alienation was a significant factor in a favorable vote
for the millage, this was found not to be the case.
nine variables included in the analysis,

Of the

the largest contri­

bution to the function is supplied by the variable sex.
Group o n e , the yes v o t e r s , contains many more males and it
is this group that reported a favorable vote on the issue.
Sex of the respondent, then, is the best discriminator for
favorable or unfavorable votes on the issue.

Someone in the

household having taken vocational education is the second
best discriminator with those who have been exposed to
vocational education in this way the most likely to vote in
favor of the millage.

The remaining variables, in de­

scending order of their importance as discriminators, are:
age, likelihood of a favorable vote, children in the home,
work in the county,

income, alienation and home ownership.

Alienation, then, is not important as a discriminator of
persons on the favorable vote dimension.

The data also show

that the likelihood of a favorable vote prior to the elec­
tion is not a good discriminator of actual vote on the
issue.
The nine variables included in the analysis do not
do very well in explaining the function.

The results of

171
the analysis were not statistically significant.

Other

factors would have to be considered in the Shiawassee
study to determine the vote function.

The function, ho w ­

ever, does well in predicting group membership.

It is

slightly better at predicting membership in group one, the
yes vote g r o u p , than for the no vote gro u p .
Alienation, then, was not a significant factor in
either of these two analyses.
explanations for this finding.

There are several possible
First, much of the research

into alienation was accomplished using general election
data.

A low interest,

low information election may not

resemble a general election to a great enough degree for
the findings of the latter to apply in the former.

Second,

the research into school referenda and alienation relied
upon the generic trust in government and political efficacy
statements.

This study used situation-specific items.

It

is possible that alienates on the specific statements do not
behave as alienates on the generic.

The research, then,

would not be a good predictor of behavior in the present
case.

Thirdly, as has been mentioned previously, Shiawassee

County may be a deviant case.

Identical research in

another setting may provide completely different results and
may find that alienation, as defined herein, does signif­
icantly relate to both the likelihood of a favorable vote
and the actuality of such a vote.

172
"Source of Information" Types
Information about the millage election was dis­
tributed to voters through newspaper articles, radio news
programs, meetings, posters and brochures.

Additionally,

information was received by voters through conversations
with friends.

It was predicted that differences would exist

between source types--respondents who gained information
from the various sources.

Two types of analysis were pe r ­

formed to determine if this was the case.

First, source

types were compared on a variety of variables under study.
No attempt was made to identify a respondent as relying on
only one source.

The respondent was included in the pr o ­

file of each source mentioned.

Second, respondents were

assigned uniquely to one category depending on the type of
source mentioned:

mass media only (newspapers, radio);

personal only (friend, meeting); quasi-mass only (brochure,
poster); mass and quasi-mass media; mass media and personal;
and, mass, quasi-mass and personal sources.
Source Profiles
Table 25 presents the source of information by
efficacy-trust group.
information sources.

Differences do exist between the six
Newspaper readers are more likely to

fall into the high efficacy-high trust group as do those
who report getting information from friends and m e e t i n g s .
Those receiving information from brochures are more likely
to fall into the high efficacy-low trust group, while those

173
using radio, the low efficacy-low trust group.

As only

two respondents mentioned posters, it is difficult to note
a trend.
TABLE 25
SOURCE OF INFORMATION BY GROUP

Source

Group I

Newspapers

10
(43.5)

Radio

2
(28.6)

Brochure

2
(2 2 .2)

4
(44.4)

0

1
(50.0)

Poster
Friends

8
(72.7)

Meeting

4
(57.1)

Group
Group
Group
Group

I:
II:
III:
IV:

Group III

Group IV

2
(8 .2)

3
(13.0)

8
(34.8)

23

0

1
(14.3)

4
(57.1)

7

0

3
(33.3)

9

0

1
(50.0)

2

0

3
(27.3)

11

0

2
(28.2)

7

Group II

0
1
(14.3)

N

High Efficacy-High Trust
High Efficacy-Low Trust
Low Efficacy-High Trust
Low Efficacy-Low Trust

Awareness and levels of correct information also
vary between the source groups (see Table 26).

The highest

levels of awareness are in the groups getting information
from friends and meetings

62

, while the lowest level of

awareness is found among those receiving information from
radio.

The highest level of correct information is found

among those receiving information regarding Area Vocational

174
Technical Education at a meeting followed by brochures
and newspapers.
TABLE 26
AWARENESS AND INFORMATION BY SOURCE
Source

Awareness

Number of Items Correct
3
2
1
0

Newspaper

18 a
(78.3)

6
b 7
(33.3) (38.9)

Radio

5
(71.4)

1
2
(2 0 .0) (40.0)

Brochure

7
(77.8)

4
(57.1)

Poster

2
(1 0 0 .0)

2
(1 0 0 .0 )

0

Friend

10
(90.9)

3
(30.0)

4
(40.0)

Meeting

7
(1 0 0 .0)

5
(71.4)

1
(14.3)

2
(28.6)

3
(16.7)
0

2
(1 1 .1)

23

2
(2 0 .0)

7

1
(14.3)
0

0

9

0

2

2
1
(2 0 .0) (1 0 .0)
0

N

11

1
(14.3)

aPercent of N aware.
^Percent of aware holding that level of correct
information.
Differences also exist among the groups on atten­
tiveness to local media (see Table 27).

Those receiving

information from radio and newspapers are most likely to
read the Argus Press while those hearing about the issue on
radio are most likely to listen to WOAP, the local radio
outlet.

Those receiving information from personal sources

are least likely to read the newspaper while those using
a brochure for information are least likely to listen.

7

175
TABLE 27
LOCAL MEDIA USE BY SOURCE
Source

Read Argus Press

Newspaper

Listen to WOAP

N

19

4

23

Radio

6

3

7

Brochure

6

1

7

Poster

1

0

2

Friend

6

2

11

Meeting

4

2

7

Those who had heard of the issue at a meeting were
most favorably inclined with five of seven indicating a 100%
likelihood of voting :
in favor of the millage and a mean
favorability level of 84.25.

The least favorable group was

that group receiving information from newspapers with only
eight of the twenty-three reporting a 100% likelihood of a
favorable vote and a mean of 57.95.

The mean favorability

levels of the radio and friends groups were almost equal
(radio: 76.77; friend:

73.77) while the mean for the bro­

chure group was considerably less (67.14).
Table 28 compares source group and school district.
Those receiving information from newspapers are found pri­
marily in Owosso and Corunna, radio listeners in Durand and
Owosso.

Owosso contains the most brochure readers and the

largest percentage of those gaining information from
friends.

176
TABLE 28
SOURCE BY SCHOOL DISTRICT

Source

Newspaper
Radio
Brochure
Poster

District
Morrice

N

Corunna

Durand

7
(30.4)

5
(21.7)

1
(4.4)

9
(39.1)

1
(4.4)

23

0

3
(42.9)

0

3
(42.9)

1
(14.3)

7

2
(2 2 .2)

2
(2 2 .2)

1
(1 1 .1)

4
(44.4)

0

9

0

0

1
(50.0)

1
(50.0)

0

2

Friend

2
(18.2)

3
(27.3)

Meeting

2
(28.6)

1
(14.3)

0
1
(14.3)

Owosso

Perry

4
(36.4)

2
(18.2)

11

2
(28.6)

1
(14.3)

7

Tables 29 and 30 present the delineation of source
type by various demographic variables.

The majority of

respondents in each group own their own homes.

Few respon­

dents or their children have ever taken a vocational edu­
cation class in high school.

There is variance between the

groups on the presence of children in the home with those
getting information from meetings being most likely and
those hearing about the issue on radio least likely to have
children living in the home.
Differences exist as well between the groups on
economic variables.

Full-time employment ranges from 71.4%

for meetings to 47.8% for newspapers.

The majority of

workers in each group are employed in the county.

The

177
TABLE 29
SOURCE BY HOME OWNERSHIP, VOCATIONAL
EDUCATION AND CHILDREN AT HOME
Source

Home
Ownership

Newspaper
Radio

Taken Vocational
Education

Children
at Home

N

20
(87.0)

6
(26.1)

11
(47.8)

23

7
(1 0 0 .0)

1
(14.3)

3
(42.9)

7

Brochure

7
(77.8)

0

4
(44.4)

9

Poster

1
(50.0)

0

1
(50.0)

2

Friend

10
(90.9)

1
(9.1)

7
(63.6)

11

Meeting

5
(71.4)

1
(14.3)

6
(85.7)

7

highest reported income was found among those who received
information at meetings while the lowest income levels were
noted among those getting information from newspapers and
friends.
Those attending meetings are the best educated group
while those using radio, friends and newspapers have the
lowest levels of educational attainment.

Those getting

information from a brochure or a meeting are the youngest
and those using radio,

the oldest.

Finally women are the

majority in each group, however, men are more likely to be
present in the groups reporting information from newspapers
and at meetings.

178
TABLE 30
SOURCE BY DEMOGRAPHIC VARIABLES
Variable
NP
(23)

Rad.
(7)

Source
Broc,
Post.
(9)
(2)

Frie.
(11)

Meet.
(7)

Employment
Full Time

11

4

6

2

6

5

Part Time

3

0

1

0

1

1

In County

12

3

5

2

5

4

Income
<$10,000

5

1

2

0

1

0

$10-20,000

6

2

2

1

4

1

$20-30,000

5

3

1

1

3

3

>$30,000

6

1

3

0

2

3

6

2

0

0

3

0

High School/
Trade School

8

4

4

1

2

1

Some College

4

0

3

1

3

3

College Degree

2

0

2

0

0

1

Graduate Work

4

1

0

0

3

2

Education
Grades 1-11

Mean Age
Sex
Male
Female

48.0

49.0

39.7

50.5

45.7

40.0

9

2

2

1

3

3

14

5

7

1

8

4

NP=Newspaper; Rad=Radio; Broc=Brochure; Post=Poster;
Frie=Friend; Meet=Meeting

179
The differences between the source of information
types points to specific differences between the sources
themselves.

As demonstrated by Table 26, information

received from radio is fleeting.

Of all the types, this

group had the lowest level of awareness and for those aware,
the lowest level of correct information.

Information from

friends, being word of mouth, has a tendency to be at least
partially incorrect.

Those sources where the greatest

attention must be paid have the highest levels of correct
information:

meetings, brochures and, to a lesser extent,

newspapers.
The data on local media use point out the limited
readership of the Argus Press and, likewise, the limited
listenership of WOAP.

Respondents not receiving infor­

mation from one of these two sources are not likely even to
attend to them.

The two local media, then, do not seem to

hold the attention of the community at large.
Personal contact seems to relate to favorability
towards the issue.

Those most favorable received infor­

mation from meetings and to a lesser extent from friends.
This group is also most likely to be found in the least
alienated group, those high on both trust in government and
political efficacy (see Table 25).

They are also most

likely to hold correct levels of information among groups
with more than two respondents (see Table 26).

180
Profiles of Source Types
Previous research into media use had found that
those using different types of media as sources of infor63
mation differed from others in unique w a y s .
This is
substantiated in the present study.

Profiles developed for

each source type show that differences do exist.
Twenty-three respondents indicated they had heard
about Area Vocational Technical Education from the news­
paper.

This group is fairly evenly split between the high

efficacy-high trust (43.5%) and the low efficacy-low trust
(34.87,) groups

(see Table 25).

The Newspaper-type was more

likely than not to be aware than an election had been
called, however several other types were more likely to be
aware than this type (see Table 26).

Level of correct

information is also quite high among this group with 56.5%
of the group providing two or three correct responses.

As

expected, the Newspaper-type is more likely than not to
read the Owosso Argus-Press

(only one group exceeds its

readership level) but is not likely to listen to WOAP, the
local radio outlet (see Table 27).

The Newspaper-type is

not favorably inclined towards the issue with a mean favor­
ability level of only 57.95, the lowest of the six types.
Thys type is most likely to live in Owosso or Corunna, with
69.5% of the group residing in these two communities, and
to own their own h o m e s .

They are less likely to be employed

full time but more likely to work on a part time basis, and
generally to work in the county.

They are spread fairly

181
evenly across all income grou p s .

Of all source types, the

Newspaper-type is more likely to have attended trade school
and to have taken, or had someone in the family take, a
vocational education class in high school.

Finally, the

type is most likely to be in her late forties and to be
female.
The Radio-type, even though it is a mass media type,
shows some definite differences.

This type is more likely

to be found in the low trust-low efficacy group (57.17o).
While its awareness level is high (71.4%), this type is
most likely to have incorrect information with 407, having
no correct information and only 207o getting all three items
right.

The Radio-type reads the local newspaper but, while

the proportion is the greatest, less than 507. of this type
report listening to the local radio outlet, W O A P .

It seems

likely, however, that they do listen to WOAP but do not
consider it a station that they listen to regularly.

This

type is likely to favor the millage (mean: 76.67), even
though they have little-to-no personal experience with voca­
tional education either in high school or trade school.

The

Radio-type owns his own home either in Durand or Owosso and
may or may not have children living at home.

This type has

little-to-no education past high school with one each having
attended trade school and having a graduate degree.

The

Radio-Type is more likely to be employed full-time in the
county, to be middle aged (mean: 49), to make more than
$20,000 a year in total household income and to be female.

182
The third source type got information from a flyer
or brochure.

The Brochure-type is more likely to be low in

trust in government with 44.4% scoring low in trust and high
in efficacy and 33.3% rating low in trust and low in
efficacy.

The Brochure-type was aware that an election had

been called and had a high level of correct information
(three correct: 57.1%; two correct: 28.6%).

This type reads

the Argus Press (66.7%) but does not listen to WOAP (11.1%).
This type is not very favorable towards the issue (mean:
67.14) possibly because no one had had personal or family
experience with vocational education in high school although
2 2 .2%, had attended trade school upon completion of high
school.
(55.6%,).

They are also more likely to have attended college
The Brochure-type is most likely to own a home in

Owosso and to work full time in Shiawassee County, but is
less likely to have children living in the home.

The

Brochure-type, finally, is most likely to be in the late
thirties and to be female.
The fourth type is an interpersonal type, getting
information about Area Vocational Technical Education from
friends.

This Friend-type is most likely to be high in both

trust in government and political efficacy (72.7%) and to be
aware than an election had been called (90.9%).

The type

is more likely than not to have some correct information
about the issue (two or three correct:

70%).

The Friend-

type is fairly favorable towards the issue and attends to
the local paper (54.6%.) but not to the local radio outlet (18%.).

183
This type owns a home and is more likely to live in Owosso
or Durand (63.3%).

The type is fairly well-educated (54.67*

with at least some college; 18.2% with graduate degrees) and
is more likely than not to be employed full time in Shia­
wassee County.

The Friend-type is not characterized by

having taken, or had someone in the family take, a voca­
tional education class in high school.

This type is more

likely to be in the mid-forties, have children living at
home and to be female.
The fifth source type got information ragarding Area
Vocational Technical Education at a meeting.

The Meeting-

type tends to be high on both trust in government and polit­
ical efficacy although a substantial minority is low in
both.

This type is aware that an election had been called

and is most likely to hold correct information concerning
the issue.

The type is also most likely to be favorable

towards the issue (mean: 84.25).

The Meeting-type reads

the Owosso Argus Press and is more likely than all but the
Radio-Type to listen to WOAP.

This tyne is spread through­

out the county although they are slightly more likely to
live in Owosso and Corunna.

The Meeting-type is a homeowner

who works full time in Shiawassee County and has children
living at h o m e .

This type tends to be fairly highly

educated (85.7% with at least some college) but to have
little personal contact with vocational education in high
school.

The Meeting-type is approximately forty years of

age, earns a total household income of more than $20,000 a

184
year and can be either male or female.
The final type received information about the issue
from a poster.

As there were only two respondents in this

category, the determination of a profile is difficult.
ever, several characteristics can be noted.
low in trust in government,

Ho w ­

The type is

found in the low trust-high

efficacy and low trust-low efficacy groups.

As a type, they

are aware of the election and hold correct information con­
cerning the issue.

The Poster-type is employed full-time in

the county, has a medium range income and is approximately
fifty years of age.
Unique Source Types
Six unique source types were identified:

personal

sources only (n=5), quasi-mass only (n=2), mass media only
(n=12), mass and quasi-mass

(n=5), personal and mass

and a group using all three (n=2).

(n=8) ,

Differences exist b e ­

tween these six groups on a number of items.
Table 31 presents the comparison of source type by
alienation categories.

Those mentioning only personal

sources all fall into the high efficacy-high trust group.
This is the only source group where this pattern occu r s .
The other source types are spread across alienation groups.
Users of mass media are evenly distributed in three of the
groups while users of quasi-mass, mass-personal and massquasi-mass-personal are each evenly divided between two
groups.

Only the mass-quasi-mass group differs with two

185
respondents each in the high trust-high efficacy and high
trust-low efficacy groups and one in the low efficacy-low
trust group.
TABLE 31
SOURCE TYPE BY ALIENATION GROUP

Source
Group I

Alienation
Group II Group III

N
Group IV

Personal

5

0

0

0

5

Quasi-mass

0

1

0

1

2

Mass Media

4

0

4

4

12

Mass-Quasi-mass

2

2

0

1

5

Mass-Personal

4

0

0

4

8

Mass-PersonalQuasi-mass

0

1

0

1

2

Group
Group
Group
Group

I:
II:
III:
IV:

High Efficacy-High Trust
High Efficacy-Low Trust
Low Efficacy-High Trust
Low Efficacy-Low Trust

Differing levels of awareness and correctness of
information also occur among the types (see Table 32) .

Th(

highest level of awareness is found among that group using
both mass and personal sources.^

The highest level of

correct information was found among those receiving infor­
mation from personal s o u r c e s . ^

Mean favorability levels

also vary from a high of 93.75 for those reporting infor­
mation from personal sources to 57.5 for those indicating
both mass and quasi-mass sources of information.

66

186
TABLE 32
AWARENESS AND INFORMATION BY SOURCE

Source

Awareness

Number of Items Correct
2
3
1
0

N

Personal

4

3

0

1

0

5

Quasi-mass

1

1

1

0

0

2

Mass Media

7

1

3

1

2

12

Mass-Ouasi-mass

4

1

2

1

0

5

Mass-Personal

8

2

3

1

2

8

Mass-PersonalQuasi-mass

2

2

0

0

0

2

Table 33 presents the data on local media use by
source type.

The highest level of readership was found

among those groups reporting having; "heard" about the issu<
from the newspaper while the highest level of listenership
was among those reporting personal sources.
TABLE 33
SOURCE TYPE BY LOCAL MEDIA USE

Source

Read Argus Press

Listen to WOAP

N

Personal

1

2

5

Quasi-mass

1

0

2

Mass Media

10

3

12

Mass-Quasi-mass

4

1

5

Mas s-Personal

5

1

8

Mass-PersonalQuasi-mass

1

0

2

Source types are not distinguishable by school
districts (see Table 34).

Demographic differences exist,

however (see Tables 35 to 37).
spread across the groups.

Home ownership was wid e ­

Few respondents had taken or had

children who had taken vocational education in high school.
Only two source categories have direct experience with
vocational education (mass and mass-personal).

Those indi­

cating information from mass and mass-quasi-mass sources are
less likely to have children present in the home.
TABLE 34
SOURCE TYPE BY SCHOOL DISTRICT

Source
Corunna

Durand

District
Morrice

N
Owosso

Perry

Personal

1

1

0

1

2

5

Quasi-mass

0

1

0

1

0

2

Mass Media

3

3

1

4

1

12

Mass-Ouasi-mass

2

1

0

2

0

5

Mass-Personal

2

2

0

3

1

8

Mass-PersonalQuasi-mass

0

0

1

1

0

2

Those reporting mass- personal sources have the
highest levels of educational attainment while those mentioning only mass m e d i a , the lowest.

The mass media group

has the highest mean age (52.16) and quasi-mass, the lowest
(27).

Examination of the groups by sex reveals that

two groups are predominantly female (personal: 80%;

188
mass-quasi-mass: 1007o) while the remainder each contain
half males and half females.

Those reporting information

from media sources are least likely to be employed full­
time but are most likely to work in Shiawassee County.
Finally, those reporting information from mass and personal
sources report the highest levels of income while those
indicating mass media only, the lowest.
TABLE 35
SOURCE BY HOME OWNERSHIP, VOCATIONAL
EDUCATION AND CHILDREN IN HOME

Source

Home
Ownership

Taken Vocational
Education

Children
In Home

N

Personal

4

0

4

5

Quasi-mass

2

0

1

2

Mass Media

11

4

5

12

Mass-Quasimass

4

0

2

5

Mass-Personal

7

2

6

8

Mass-PersonalQuasi-mass

1

0

1

2

TABLE 36
SOURCE BY AGE AND SEX

Mean Age

Per
(5)

Ouas
'(2)

Source
Mass
MasOua
(12)
(5)

40.0

27.0

52.2

40.4

MasPer
(8)
44.3

MQP
(2)
50.5

Male

1

1

6

0

4

1

Female

4

1

6

5

4

1

TABLE 37
SOURCE BY EDUCATION, EMPLOYMENT AND INCOME
Source

Variable

p

Pera
(12)

Quas^
(2)

Mass0
(12)

1

0

4

0

2

0

High School/
Trade School

1

1

7

2

1

2

Some College

2

0

0

2

1

0

College Degree

1

1

1

1

0

0

Graduate Work

0

0

0

0

4

0

Employment
Full-Time

3

1

4

2

5

2

Part-Time

1

0

2

1

0

0

In County

2

1

5

2

4

2

0

0

3

2

0

0

$10-20,000

1

1

4

0

2

1

$20-30,000

1

0

4

0

3

1

>$30,000

2

1

1

2

3

0

Education
Grades 1-11

Income
<$10,000

MasQua^
(5)

MasPer
(8)

MQPf
(2)

aPersonal; ^Quasi-mass; cMass Media; ^ass-Quasi-mass; eMass-Personal; ^Mass-PersonalQuasi-mass

190
Profiles of Source

Category

Types

Profiles were developed

for each

unique group. The

first type reported getting information about Area Voca­
tional Technical Education from personal sources, defined
as at a meeting or from a friend.

All five respondents in

this grouping are high on political efficacy and trust in
government, a direct contradiction of the proposed model.
The Personal-type is aware that an election has been called,
holds correct information regarding the issue and is most
likely to vote in favor of the m i l l a g e .
read the local newspaper but is
to WOAP.

This type does

somewhat

not

inclined to listen

The Personal-type is a homeowner with children

living at h o m e .

He or she has had no personal experience

with vocational education and is more likely than not to
have attended college.

The type is employed either full or

part-time, but not necessarily in Shiawassee County, and
earns more than $20,000 a year in total household income.
The Personal-type is approximately forty years old and is
much more likely to be female than male.
The Mass Media-type is not characterized by a single
alienation category, being evenly divided among the high
trust-high efficacy, high trust-low efficacy and low trustlow efficacy groups.

This type is only slightly more likely

to be aware of the election than unaware and is not very
likely to hold correct information regarding the issue or
to be favorably inclined towards it.

The Mass Media-type

reads the local newspaper but tends not to listen to the

191
local radio outl e t .

They tend to own their own homes but

not to have children living in those homes.

Of all types,

they have the greatest tendency to have taken, or had some­
one in the family take, a vocational education class in high
school.

They tend to be less educated than the other types

with only 8.3% having attended college.

The Mass Media-type

is less likely to be employed either full or part-time but,
for those who are employed, is quite likely to work in the
county.

The income of the type ranges from $10,000 to

$30,000 a year in total household income.
type, finally,

The Mass Media-

is the oldest of the six types and is evenly

divided between males and females.
Only two respondents mentioned only Quasi-mass
sources defined as brochures and posters.

As was mentioned

earlier, this is too few for a true profile to be drawn
but several characteristics should be noted.

The Quasi-mass-

type is low in trust being found in both the low trust-low
efficacy and low trust-high efficacy groups.

They are young

homeowners who have never taken a vocational education cla s s ,
who may or may not have children living

in the home and who

may or may not be employed full-time.
Eight respondents mentioned both media and personal
sources.

The Mass-Personal-type may be

either high on

both trust in government and political efficacy, or
both.

low in

The type was aware than an election had been called

and was more likely than not to have some correct infor­
mation concerning the issue.

They are fairly favorably

192
inclined towards the issue.

This type tends to read the

Owosso Argus Press but not to listen to W O A P .

The Mass-

Personal-type is a homeowner with children living at home
and has had some personal experience with vocational edu­
cation.

The type is more likely than not to be employed

full-time, to work in Shiawassee County and to earn a total
household income of more than $20,000 a year.

They are

more likely to have attended graduate school.

The Mass-

Personal-type is approximately forty-four years of age and
can be either male or female.
The Mass-Quasi-mass-type tends to be high in effi­
cacy but either high or low in trust in government.

This

type is aware of the election and is as likely as not to
hold some correct information regarding the issue.
ability, however,

is low.

Press but not to WOAP.

Favor-

They attend to the Owosso Argus

The Mass-Quasi-mass-type tends to

be a homeowner in either Owosso or Corunna but not to have
children living at home.

The type has had no personal

contact with vocational education in high school but has a
tendency to have attended trade school after high school.
A tendency to have attended college is also characteristic
of this type.

The Mass-Ouasi-mass-type is more likely to

work than not and is characterized by either a low income
(less than $10,000 a year) or a high income (more than
$30,000).

The type is approximately forty years of age

and is female.

193
respondents who mentioned receiving information from per­
sonal, quasi-mass and mass media sources.

This Combination-

type included only two respondents and will be dealt with
briefly.

This type is low in trust in government but mixed

in political efficacy.

The type was aware of the election

and held correct information concerning it.

They may or

may not read the local newspaper but do not listen to the
local radio outlet.

This type has not attended college,

has not taken a vocational education class in high school,
works full-time in Shiawassee County, makes between $10,000
and $30,000 a year in total household income, is approxi­
mately fifty years of age and may be either male or female.
A Research Update
Since the completion of this research, new evidence
has come to light regarding the unidimensionality of the
political efficacy scale and the relationship of efficacy
to trust in government.

Beginning with the work of B a l c h ^ ,

many researchers have uncovered data which point to the
separation of political efficacy into two distinct but
related attitude sets:

internal or personal efficacy and

external or "political" efficacy.
Craig attempted to distinguish between these two
dimensions of political efficacy and then examined them and
their individual relationships with the concept of political
68
trust.
He defined internal efficacy as input efficacy,
the degree to which an individual perceives he has access
to the political system.

This dimension includes feelings

194
of competence and is measured by responses to "Voting is the
only way people like me can have a say in what the govern­
ment does" and "Sometimes politics seems so complicated that
a person like me can't understand".

External or output

efficacy is the degree to which the individual believes his
political activities can be successful and is measured by
responses to "People like me can have no say about what the
government does" and "Politicians don't care what people
like me think".

Craig points out that these are funda­

mentally different concepts stemming from different origins
with input being a function of personal qualities such as
socio-economic status and output reflecting variations in
evaluations of government performance.
Utilizing data collected from students in five
Chicago area colleges and universities, Craig found that
output efficacy was related to both political trust and
input efficacy but that input efficacy was not related to
political trust.

Political participation was more closely

related to input efficacy than to either output efficacy or
political trust.

Input efficacy was also rather strongly

related to community involvement regardless of level of
trust.

Finally, he reports that those who are both trustful

and efficacious are less likely to condone unorthodox
political activities.
Baloyer proposes that political cynicism (dis­
trust

of government)

is a form of criticism associated

with social conditions and a consequence of partisan and

195
ideological oppos i t i o n s . ^

Utilizing data from a national

sample in Venezuela, he reports that his findings establish
that "it is more appropriate to treat the sense of political
efficacy as dependent on criticism rather than the other way
around".^

Efficacy does not appear to have much impact on

participation while it (participation) does fluctuate with
an individual's evaluation of the performance of the govern­
ment.

He proposes that in the United States, the increased

criticism-reduced efficacy relationship may be at work to
lower levels of political participation.
Craig and Maggiotto examined the effects of internal
efficacy, the indivicual's self-perception that he is
"capable of understanding politics and competent enough to
participate in political acts," and policy dissatisfaction
upon the relationship between political behavior and polit­
ical discontent, defined as political trust, diffuse
support and external e f f i c a c y . ^

They point out that the

lack of external efficacy expresses itself in the belief
that the individual cannot influence political outcomes be ­
cause the government leaders and institutions are unre72
sponsive to his n e e d s .

This belief that the system is

unresponsive could provide a strong motivation for political
action among the internally efficacious, those citizens who
73
believe themselves to be "entitled to political p o w e r " .
They found that discontent was most strongly associated
with the respondent's degree of acceptance of political
protest and violence among the internally efficacious and

196
among those individuals who express dissatisfaction with
the government's political performance.

This was particu­

larly true for the association of external efficacy with
the degree of acceptance among the internally efficacious.
Pollock posits four distinct types of alienative
attitudes.^

In the social arena, internal efficacy is

characterized by feelings of personal control while external
efficacy finds its outlet in ideology control.

In the

political arena, internal efficacy is related to political
competence, the extent to which the individual believes
political outcomes are affected by his own attempts to
influence them, while external efficacy is related to polit­
ical responsiveness, whether or not the individual feels
that political institutions are responsive to popular
control.^

Examining involvement in purposive organizations

such as political clubs and special interest groups, he
found that this type of activity corresponded to lower
levels of political alienation, defined as external polit­
ical e f f i c a c y . ^
Pollock also examined the participatory effects of
internal and external efficacy.

77

He found that those low

on internal efficacy but high on external efficacy are
likely to participate in a traditional manner, especially
the act of voting.

This group also had the lowest propen­

sity for unconventional behavior.

Those with high inter­

nal but low external efficacy were in an optimal state for
nonconformist behavior, however, they also had a tendency

197
to participate conventionally, particularly in the areas of
78
campaigning and contacting othe r s .

He also found that

the politically competent but mistrusting individual did
not view political participation as a simple choice between
traditional behavior such as voting and extemist action.
In an examination of turnout as a measure of polit­
ical participation in Presidential elections, Miller found
that internal political efficacy could not account for
decreases in participation.

79

While these beliefs about

personal political efficacy were strongly associated with
turnout, they had not declined while turnout had declined
considerably over the last two decades.
Abramson and Aldrich also examined the decline of
political participation since the early 1950's.

80

They

suggest that the decline in turnout is the result of two
attitudinal trends:

"the weakening of party loyalties

among the American electorate and declining beliefs about
government responsiveness, that is lowered feelings of
'external' political efficacy".

81

The authors point out

that those persons who feel politically capable, the inter­
nally efficacious, may feel psychologically motivated to
participate while those who feel overwhelmed by the entire
political process may withdraw from all such action.

Uti­

lizing national election data, they found that the decline
in feelings of external political efficacy appeared to be
a partial explanation for the decline in electoral parti­
cipation and could account for as much as half of the

198
decline in turnout in Presidential elections.

82

Finally, in a study of voting in school financial
elections, Cataldo and Holm found that political efficacy,
defined as responses to three of the scale items (don't
have a say, politics seems so complicated and politicians
don't care), was positively and significantly related to
voting in favor of an increase in operating levies for the
schools.

83

Trust in government (trust to do what's right

and waste tax m o n e y ) , a child in school and willingness to
support increased taxes for mass transit were also posi­
tively and significantly related to a favorable vote.
If the political efficacy scale is no longer unidimensionsl as the research seems to suggest, then the com­
bining of the four items into a single scale could mask
any real differences present in the data.

For example, a

person high on internal efficacy but low in external effi­
cacy would be scored in an identical manner to a person
low on internal efficacy but high on external efficacy.
The differences in the relationships each has to the polit­
ical system would not be noted in the analysis.
Further, the research suggests that each of the two
concepts of political efficacy is related to trust in govern­
ment in a unique way.

As Craig points out, external effi­

cacy may be related to trust in government while internal
efficacy may not have the same relationship.

84

This means

that a person who believes the government is responsive will
also trust that government while it is not necessary to

199
trust the government to feel that the individual can have
an impact upon it.

Given these differences,

the combining of

the political efficacy scale and the trust in government
scale to create one measure of political alienation com­
pounds the problem.

Differences in the relationship of

efficacy to trust and then the relationship of these two
variables to likelihood of a favorable vote and self-report
of actual vote may have become so entangled that, as the
research reported herein seems to suggest, alienation,
defined as efficacy and trust, would appear to have no
impact on vote.

This finding is not present in any of the

literature to date, as noted earlier.

The reconceptualiza­

tion of political efficacy may provide the explanation.

200
Endnotes
^"Richard F. Carter, Voters and Their Schools: A
Technical Report from the Institute for Communication
Research (Stanford, Calif.: Stanford University Press,
30 June 1960), pp. 132-36.
2

It must be noted that many of the respondents who
were contacted in the early weeks of the campaign may have
acquired information about the issue as the campaign
progressed.
3
Thirty-one of the sixty respondents noted they had
voted in the "last one" with no definite time given.
Fif­
teen had voted in one last fall, two in June and twelve,
one year ago.
^Gerald M. Pomper, "The Presidential Election," in
Marlene M. Pomper, ed., The Election of 1976: Reports and
Interpretations (New York: David McKay Company, Inc., 1977),
{T TT.
^Turnout for the 1976 Presidential election was
62.8%.
U.S., Department of Commerce, Bureau of the Cen­
sus , 1977 City County Data Book: A Statistical Abstract
Suppliment (Washington, D.C., 1977).
f.

David I. Verway, Michigan Statistical Abstracts,
13th ed. (East Lansing, Michigan: Michigan State University Graduate School of Business Division of Research,
1978), p. 47.
^Ibid., p. 69.
Q

Warren E. Miller and Teresa E. Levetin, Leadership
and Change: The New Politics and the American Electorate
(Cambridge, Mass.: Winthrop Publishers, I n c . , 1976), p p . 27,
219-33; and, Norman H. Nie, Sidney Verba and John R.
Petrocik, The Changing American Voter (Cambridge: Harvard
University Press, 197o), pp. 85-88, 234-38.
9Ibid.
Z-score of 1.65 is needed for a difference to be
significant at an alpha level of .05 for a one-tailed test.
^ A d a W. Finifter, "Dimensions of Political Aliena­
tion," American Political Science Review 64 (June 1970):

201
89-410.
^ L e s t e r W. Milbrath and M. L. Goel, Political
Participation: How and Why Do People Get Involved in
Politics 2d ed (Chicago: Rand McNally Publishing Company.

1977).—
13

Carter, Voters and Their Schools, pp. 132-36.

■^Howard D. Hamilton and Sylvan H. Cohen, Policy
Making by Plebiscite:
School Referenda (Lexington, Mass.:
D"! C . Heath and Company, 1974) , p . 230.
■^Milbrath and Goel, Political Participation.
■^See, for example, Murray B. Levin, The Alienated
Voter: Politics in Boston (New York: Holt, Rinehart and
W i nston, I960); Gwynn Nettler, MA Measure of Alienation,"
American Sociological Review 22 (December 1957):670-77;
Sidney Kraus and Dennis D a v i s , The Effects of Mass Communi­
cation on Political Behavior (University Park: The Pennsyl­
vania State University Press, 1976); Dwight G. Dean, "Alien­
ation and Political Apathy," Social Forces 38 (March 1960):
185-95; William Erbe, "Social Involvement and Political A c ­
tivity: A Replication and Elaboration," American Sociologi­
cal Review 29 (April 1964):198-215; and, John P. Clark,
"Measuring Alienation Within a Social System," American
Sociological Review 24 (December 1959):349-52.
^ A n g u s Campbell, "The Passive Citizen," in Edward
C. Dreyer and Walter A. Rosenbaum, e d s ., Political Opinion
and Electoral Behavior (Belmont, Calif.: Wadsworth Publishing Company, 1966), pp. 175-88.
■^Philip K. Piele and John Stuart Hall, Budgets,
Bonds and Ballots: Voting Behavior in School Financial
Elections (Lexington, Mass.: D.C. Heath and Comoanv, T973),
p". TIT.—
* ‘
19

See Milbrath and Goel, Political Participation,
and Finifter, "Dimensions of Alienation".
20

See Morris Rosenberg, "The Meaning of Politics in
Mass Society," Pub1ic Opinion Quarterly 15 (Spring 1951):
5-15; Melvin Seeman, "Alienation and Engagement," in Angus
Campbell and Philip E. Converse, eds., The Human Meaning
of Social Change (New York: Russell Sage Foundation, 19/2),
pp. 467-527; Garrett J. O'Keefe and Harold Mendelsohn,
"Nonvoting: The Media's Role," in Charles Winick, ed.,

202
Deviance and Mass Media (Beverly Hills, Calif.: Sage
Publications, 1978); Jack Citrin, "Comment:
The Political
Relevance of Trust in Government," American Political
Science Review 68 (September 1974):973-88;
and, Milbrath
and Goel, Political Participation.

21

See Kurt Lang and Gladys Engel Lang, "The Mass
Media and Voting," in Eugene Burdick and Arthur J. Brodbeck,
eds., American Voting Behavior (Glencoe, 111.: The Free
Press, 1959):217-35 and Lee B. Becker, Idowu A. Sobowale
and Willian E. Casey, Jr., "Newspaper and Television
Dependencies: Their Effects on Evaluation of Govern­
ment Leaders," paper presented to the International Commun­
ications Association, Chicago, 1978.
^ S e e Elihu Katz and David Foulkes, "The the Use of
the Mass Media as 'Escape': Clarification of a Concept,"
Public Opinion Quarterly 26 (Fall 1962):377-88; Jay G.
Blumler and Denis M c Q u a i l , Television in Politics: Its
Uses and Influence (London: Paber and Faber, LTD, 1968);
Jack M. McLeod and Lee B. Becker, "Testing the Validity
of Gratification Measures Through Political Effects
Analysis," in Jay G. Blumler and Elihu Katz, eds., The Uses
of Mass Communications: Current Perspectives on Gratifixations Research (Beverly U i l l s , Cal i f .: Sage Publications,
1974), p p . 137-64; and, William Semlac and Wenmouth
Williams, Jr., "Uses, Gratifications and Avoidance:
Voting
Decisions in the 1976 Presidential Debates," paper presented
to the International Communications Association, Chicago,
1978.
23

Examination of the source of information variable
will show that this is not necessarily the case.
0/

See Bradley S. Greenberg, "Media Use and Believability: Some Multiple Correlates," Journa 1 ism O.uarterly
43 (Winter 1966):665-70, 732; Serena Wade and Wilbur Scnramm,
"The Mass Media as Sources of Public Affairs, Science and
Health Knowledge," Public Opinion Quarterly 32 (Simmer
1969):197-209; F. Gerald Kline, "Media Time Budgeting as a
Function of Demographics and Life Style," Journalism
Quarterly 48 (Summer 1971):211-21; Peter Clarke and L.
Ruggels, "Preferences Among News Media Coverage of Public
Affairs," Journalism Quarterly 47 (Autumn 1970):464-71;
Samuel J . Eldersveld,
Experimental Propoganda Techniques
and Voting Behavior," American Political Science Review 50
(March 1956):154-65; Heinz Eulau and Peter Schneider,
"Dimensions in Political Involvement," Public Opinion
Quarterly 20 (Spring 1956):128-42; Alexis S. Tan, "Political
Participation, Diffuse Support and Perceptions of Political
Efficacy as Predictors of Mass Media Use," Communication

203
Monographs 48 (June 1981):133-45; Garrett J. O'Keefe,
"Political Malaise and Reliance on Media," Journalism
Quarterly 57 (Spring 1980):122-28; and, Becker, Sobowale
and Casey, "Newspaper and Television Dependencies".
25

See Levin, The Alienated V o t e r , and Nettler,
"Measures of Alienation".
^ S e e Semlac and Williams, "Uses, Gratifications,"
and Blumler and McQuail, Television in Politics.
^ M i l b r a t h and Goel, Political Participation.
^®See Joel Aberbach and Jack Walker, "Political
Trust and Racial Ideology," American Political Science
Review 64 (December 1970):1199-1219; Herbert McCloskey,
"Consensus and Ideology in American Politics," American
Political Science Review 58 (June 1964):361-83; Dwight G.
Dean, "Alienation: It's Meaning and Measurement," American
Sociological Review 26 (October 1961):753-58; and, Citrin,
"Comment".
^ C a r t e r , Voters and Their Schools, pp. 132-36.
on

See Endnote 32, Chapter I, for a listing of all
media available in the county.
31

Mean favorability for the manager category was
58.75 and for the professional, 91.0.
qo

Carter, Voters and Their Schools, p. 67.
■^Hamilton and Cohen, Policy M a k i n g , pp. 78-84.
■^Carter, Voters and Their Schools.
qc

Hamilton and Cohen, Policy M a k i n g .
qz:

Joan P. Sullivan Kowalski, Voter Behavior and
Campaign Strategies in School Finance~Elections (Arlington,
Virginia: Educational Research Service, I n c ., T977), pp.
28-30.
37

Hamilton and Cohen, Policy M a k i n g .

■^Arthur J. Alexander and Gail V. Bass, School Taxes

204
and Voter Behavior: An Analysis of School District Property
Tax Elections (Santa Monica, C a l i f .: The Rand Corporation,
April T 9 7 7 7 T P . 65.
39
40

Hamilton and Cohen, Policy M a k i n g , pp. 78-84.
Alexander and Bass, School Taxes and Voter Be ­

havior .
4^James Q. Wilson and Edward C. Banfield, "PublicRegardingness as a Value Premise in Voting Behavior,"
American Political Science Review 58 (December 1964):876-87.
^ C a r t e r , Voters and Their Schools, p. 67.
^ H a m i l t o n and Cohen, Policy M a k i n g , pp. 78-84.
^ A l e x a n d e r and Bass, School Taxes and Voter
Behavior, p. 65.
^ Kowalski, Voter Behavior and Campaign Strategies,
pp. 28-30.
46Ibid., p. 32.
4^Marvin E. Olsen, "Two Categories of Political
Alienation," Social Forces 42 (March 1969):288-99.
4®Wayne Thompson and John E. Horton, "Political
Alienation as a Force in Political Action," Social Forces
38 (March 1960):190-95.
49

Donald E. Stokes, "Popular Evaluations of Govern­
ment: An Empirical Assessment," in Harlan Cleveland and
Harold D. Lasswell, eds., Ethics and Business: Scientific,
Academic, Religious, Political and Military (New York:
H arp e r , 1962), p p . 61-73.
■*®Joel Aberbach, "Alienation and Political Behavior,"
American Political Science Review 63 (March 1969):86-99.
■^Milbrath and Goel, Political Participation, p. 69.
"^Finifter, "Dimensions of Alienation".
^ S e e Murray B. Levin and Murray Eden, "Political

205
Strategy for the Alienated Voter," Public Opinion Quarterly
26 (Spring 1962):47-63; Nettler, "Measure of Alienation";
Campbell, "Passive Citizen"; Kraus and Davis, Effects of
Mass Communication; and Erbe, "Social Involvement".
54

Edward L. McDill and Jeanne Clare Ridley, "Status,
Anomia, Political Alienation and Political Participation,"
American Journal of Sociology 68 (September 1962):205-13.
■’"’Howard D. Hamilton, "The Municipal Voter: Voting
and Nonvoting in City Elections," American Political
Science Review 65 (December 1971):1135-40.
■^Aberbach and Walker, "Political Trust".
■^John E. Horton and Wayne E. Thompson, "Powerless­
ness and Political Negativism: A Study of Defeated Local
Referendums," American Journal of Sociology 67 (March
1962):485-93.
58

Seeman, "Alienation and Engagement".

59

See Thompson and Horton, "Alienation as a Force,"
and Piele and Hall, Budg e t s , Bonds and Ballots .
^ S e e Michael W. Giles, Douglas S. Gatlin and Everett
F. Cataldo, "Parental Support for School Referenda," Journal
of Politics 38 (May 1976):442-51; Hamilton and Cohen, Policy
Maki n g ; a n d , Carter, Voters and Their Schools.
^ W i l s o n and Banfield, "Public-Regardingness".
62

The two respondents noting posters will be noted
in the tables but not in the text as any conclusions drawn
from only two respondents could be highly misleading.
63

See the Review of Literature, Chapter I I , for a
detailed discussion.
^ T h o s e noting all three categories are also aware,
however the low number in the category makes the result less
stable.
65

Once again, the small overall number in the quasi­
mass and mass-personal-quasi-mass groups casts doubt on the
reliability of the finding.

206

66

The other mean favorability levels were: MassPersonal, 71.125; Mass Media, 58.64; Quasi-mass, 70 (based
on two respondents); and, Mass-Personal-Quasi-mass, 100
(based on one respondent).
^ G e o r g e I. Balch, "Multiple Indicators in Survey
Research: The Concept 'Sense of Political Efficacy',"
Political Methodology 1 (1974):l-43.
fLQ

Stephen C. Craig, "Efficacy, Trust and Political
Behavior: An Attempt to Resolve a Lingering Conceptual
Dilemma," American Politics Quarterly 7 (April 1979):
225-239.
69

Enrique A. Baloyar, "Criticism, Cynicism and Polit­
ical Evaluation: A Venezuelan Example," American Political
Science Review 73 (December 1979):987-1002.
70Ibid., p. 1 0 0 0 .
7^Stephen C. Craig and Michael A. Maggiotto, "Polit­
ical Discontent and Political Action," Journal of Politics
43 (May 1981):515.
72I b i d .
73I b i d ., p. 516.
7^Philip H. Pollock III, "Organizations and Alien­
ation: The Mediation Hypothesis Revisited," The Sociological
Quarterly 23 (Spring 1982):143-55.
75Ibid., p. 147.
76

For further data on the impact of political effi­
cacy on other areas of life, see David 0. Sears, Richard R.
Lau, Tom R. Tyler and Harris M. Allen, J r . , "Self-Interest
v s . Symbolic Politics in Policy Attitudes and Presidential
Voting," American Political Science Review 74 (September
1980):670-84; and, J. Maxwell Elden, "Political Efficacy at
Work: The Connection Between More Autonomous Forms of
Workplace Organization and a More Participatory Politics,"
American Political Science Review 75 (March 1981):43-58.
77Philip H. Pollock III, "The Participatory Conse­
quences of Internal and External Political Efficacy,"

207
The Western Political. Quarterly 36 (September 1983):
400-409.
~
78Ib i d ., p. 405.
79

Warren E. Miller, "Disinterest, Disaffection,
and Participation in Presidential Politics," Political
Behavior 2 (1980):7-32.
---------80

Paul R. Abramson and John H. Aldrich, "The Decline
of Electoral Participation in America," American Political
Science Review 76 (September 1982) :502-2T"!
81Ibid., p. 502.
82Ibid., p. 515.
83

Everett F. Cataldo and John D. Holm, "Voting on
School Finances: A Test of Competing Theories," The Western
Political Quarterly 36 (December 1983):619-31.
QA

Craig, "Efficacy, Trust and Political Behavior".
See also, Paul R. Abramson and Ada W. Finifter, "On the
Meaning of Political Trust: New Evidence from Items Intro­
duced in 1978," American Journal of Political Science 25
(May 1981):297-307; and, Burt Useem, "Trust in Government
and the Boston Anti-Busing Movement," The Western Political
Quarterly 35 (March 1982):81-91.

CHAPTER V
SUMMARY AND CONCLUSIONS
Summary
Over the past several decades, researchers from a
variety of fields have investigated the role played by
alienation in political participation, media use, attitudes
towards schools, interpersonal relationships, societal
structures and communication patterns.

The research had

shown that alienation was integral to many of these pro­
cesses.

This attitude set had a major impact on how

people interacted with each other and the societal system
around them.
The various components of alienation had also been
investigated by a variety of authors from the fields of
political science, communication and sociology.

Two dimen­

sions that had appeared in many of the studies as being
interrelated were trust in government or political cynicism
and political efficacy or powerlessness.

These two com­

ponents of alienation were said to explain the varieties of
political participation observed in voting studies during
the last forty y e a r s .
Taking this body of literature as a guide, this
study proposed to examine how political alienation defined
208

209
as trust in government and political efficacy related to
information holding in a local election campaign.

Since

the research into school referenda was slight, and since
the opportunity presented itself,

the study was designed

to test this relationship in a school millage election.

The

study proposed and tested a model of awareness, information
and information source based on trust in government and
political efficacy

derived from a model of political par-

ticipation developed by Finifter

1

and Milbrath and Goel.

2

The derivative model predicted that varying levels
of awareness, correctness of information and source of
information would be found in the sample based on their
scores on trust in government and political efficacy scales.
Following the suggestions of Aberbach and others,
scales were made situation specific.

3

the

Rather than referring

to "government" and "government officials" the items
referred to the "school board," "school board members" and
"school politics".
The study took place on Shiawassee County, Michigan
in the Spring of 1979.

The Shiawassee County Intermediate

School District was attempting to pass a millage for the
construction and maintenance of a centralized Area Voca­
tional Technical Education C e n t e r .

The millage effort had

failed twice in the previous ten y e a r s .

A research team

from Michigan State University, funded by the Michigan
State Department of Education, undertook a study of the
County and the issue in an effort to bring about a favorable

210
vote.

The study was part of that larger effort.
Data were collected by telephone from 109 respon­

dents between the day the election was called (April 12,
1979)

and the day before the actual election (June 10, 1979).

Additionally, an attempt was made to recontact all respon­
dents during the week following the election.

The original

questionnaire collected information in six general areas.
The first area dealt with political attitudes and included
the six items dealing with a sense of political efficacy and
the five items measuring trust in government.

The second

section included items measuring awareness of the election
and knowledge of who had called the election, the issue to
be decided, and the amount of millage requested.

Thirdly,

the instrument collected data dealing with media use, spe­
cifically newspaper readership and subscription and radio
listenership.

Television was not included as there are

no local television outlets in the county.
The fourth area of interest on the questionnaire
was a series of questions measuring information source from
those who had heard something of Area Vocational Technical
Education "since last Monday".

Fifthly, data were collected

on voting in general elections, voting in millage elections,
and the likelihood of casting a favorable vote in the elec­
tion under study.

The final area of the questionnaire

measured demographic characteristics including home owner­
ship, education, age, income, children present in the home
and sex of respondent.

The questionnaire also included a

211
series of items measuring attitudes toward centralized
vocational education using a metric multidimensional
scaling design.
Fourteen hypotheses and several sub-hypotheses were
tested.

In addition, profiles were developed of the most

favorable voter, least favorable voter, source of informa­
tion types and source category types.

The first three

hypotheses examined the validity of the derivative model.
The first hypothesis predicted that levels of awareness
would vary significantly between the four alienation groups:
those high on efficacy and trust, those low on efficacy and
trust, those high on efficacy but low on trust, and those
low on efficacy but high on trust.

Support was found for

this hypothesis which was highly significant.

Sub­

hypotheses had proposed that levels of awareness would vary
in specific ways and these were not supported.

It had been

predicted that the proportion aware in the group high on
both variables would be greater than the proportion aware
in the group high in efficacy and low in trust.
was true.

The reverse

It had also been predicted that the proportion

aware in the group low on efficacy and high on trust would
be greater than the proportion aware in the group low on
both.

This was found to be so but the difference was not

significant.
The second hypothesis examined level of correct
information according to group membership.

It predicted

that significantly different levels of correct information

212
would exist between the four groups.
case.

This was not the

The analysis of variance showed no significant

difference.

The sub-hypotheses were also not significant.

The third hypothesis predicted that groups would
vary as to primary source of information, either depending
primarily on media or personal sources.

The high efficacy-

high trust and high trust-low efficacy groups were to rely
on media sources.

The other two were to have used personal

sources to a greater extent.
for this hypothesis.

Partial support was found

The high-high and low-low groups did

not follow the expected pattern.

Those high on both var­

iables mentioned more personal sources while those low on
both, more media sources.
as expected.

The two middle groups did fall

No significant difference was found.

Hypotheses four and five predicted that as the
election neared, the proportion holding correct information
about the election and the proportion aware that the elec­
tion had been called would increase.
Rather than increasing,

This did not occur.

in many instances the proportion

aware or with correct information dropped as the election
neared.

Only one significant difference was found.

The

proportion aware during Week nine was significantly greater
than the proportion aware during W eek eight.
Hypothesis seven predicted that those respondents
living in the Owosso and Corunna school districts would
report getting information from the newspaper to a greater
extent than those respondents living in the outlying

213
districts.

Support was found for this contention.

The

difference in the proportion reporting newspapers for
the Owosso-Corunna districts was significantly greater than
the proportion from the other districts.
The next six hypotheses examined favorability
towards the issue of Area Vocational Technical Education.
A variety of demographic variables were tested to determine
if they were associated with the likelihood of casting a
positive vote on the issue.

Age was found to be negatively

related but the correlation was not significant.

The

presence or absence of children in the home was signif­
icantly related.

The mean favorability level of those with

children was significantly greater than the mean favora­
bility level of those with no children in the home.
Favorability did increase as education increased,
with one notable exception.

Those who had attended trade

school after high school were less favorably inclined toward
the issue than high school graduates.
ferences were found.

Two significant dif­

Those who had completed high school

had a significantly greater mean favorability level than
those with less than a high school education and those with
graduate work,

including those who had earned a graduate

degree, had a significantly greater mean favorability level
than those with a college deagree but no further study.
The relationship of income to favorability was
mixed.

Favorability did not increase as income level

increased.

Instead a curvilinear relationship was noted

214 .
with favorability increasing as income increased from less
than $10,000 a year to between $10,000 and $20,000 a year.
For the next two categories of income, however, favor­
ability decreased.

The increase just noted was signif­

icant .
Occupational status was not related to favorability
as had been expected.

While favorability did increase as

occupational status increased when the managerial and
professional categories were combined, separation of these
two categories showed managers to be much less favorable
towards the issue than white collar workers.
The final favorability item was significant.

Those

who had cast a positive vote in a school millage election in
the past were substantially more likely to report a greater
likelihood of voting in favor of the present issue.
The final two hypotheses examined the relationship
of alienation to the likelihood of casting a favorable
vote and the actual vote on the millage.

In both instances,

a multiple regression analysis and a discriminant analysis,
alienation was found to have little or no impact on either
of the two dependent variables.

Of seven predictors of the

likelihood of a favorable vote, alienation was the least
important.

Of nine variables used in the discriminant

analysis, alienation was eighth in relative importance.
The study also created profiles of source types and
source category types.

The Newspaper-type was generally

aware that an election had been called, had correct

215
information regarding the issue, read the local newspaper
but did not listen to the local radio station, was not
very favorable towards the issue.

He owns his own home,

works in the county, has been exposed to vocational edu­
cation, is in his late forties and "he" is more likely to
be a "she".
The Radio-type is aware of the election but has
incorrect information about the issue, reads the local paper
and listens to the local radio station.

They favor the

millage but have little experience with vocational edu­
cation.

They are homeowners, have not attended school past

high school, are employed full time in the county, are
middle aged and female.
The Brochure-type is aware of the election and has
correct information regarding it.

They read the local paper

but do not listen to the local radio station.

They are not

favorable towards the issue and have had little experience
with vocational education.

They are home owners, are

employed full time in the county, do not have children
living at h o m e , are in their late thirties and are generally
women.
The Friend-type is aware of the election but has
incorrect information regarding it.

They are fairly

favorable towards the issue, read the local paper but do
not listen to the local radio station.

They are well-

educated home owners who are employed full time in the
county.

They are in their mid-forties, have children and

216
are female.
The Meeting-type is aware and has correct infor­
mation.

They are most favorable towards the issue, read

the paper and listen to the radio station.

They are

home owners who work full time in the county, are fairly
highly educated and are in their early forties.
There were four unique source category types con­
taining more than two respondents.

The Personal-type is

high on both efficacy and trust, is aware of the election
and holds correct information regarding it.

This type does

not attend to the local media, is a home owner with children
present in the home,

is employed, earns more than $20,000

a year, is approximately forty years old and is female.
The Mass Media-type is not aware of the election and
does not hold correct information concerning it.

The type

reads the local paper but does not necessarily listen to the
local radio station.

They are home owners, with little

education past high school and are in their mid-forties.
The Mass-Personal-type is aware of the election
and has correct information concerning i t .

They read the

paper but so not listen to the radio station.

They are

home owners with children, are employed full time in the
county, are highly educated and are in their mid-forties.
The Mass-and-Quasi-mass-type is aware of the elec­
tion but is likely to hold incorrect information about it.
They attend to the local paper but not the local radio
station.

They are home owners with no children at home,

217
have attended college, have either a high or low income,
are in their forties and are female.
Conclusions
Several conclusions can be drawn from the study.
First, a number of hypotheses were confirmed. Levels of
awareness do differ significantly among the four trust in
government-political efficacy groups.

Respondents in the

Owosso and Corunna school districts do report getting infor­
mation from the newspaper to a greater extent than those
respondents living in the outlying districts, perhaps indi­
cating that respondents in those two districts generally
rely on the local newspaper more than those in the other
districts.

The mean favorability level of respondents with

children in the home was significantly greater than the
level for respondents without children in the home pointing
to the possibility that parents form a unique interest group
regarding school issues.

There were several significant

relationships between favorability and various income and
education levels.

Past positive voting did relate signif­

icantly to the self-report of the likelihood of casting a
favorable vote in the present case.

Finally, the multiple

regression equation was significant with the nine variables
included as independent explaining 32% of the variance in
favorability.
Second, the derivative model did not predict
behavior for several possible reasons.

The original model

218
had not been tested in a research setting as a whole.

If

the original model is invalid then the derivative model
may be invalid as well.

It is also possible that the use

of the situation-specific items was problemmatic.

It is po s ­

sible that the setting in a low information election under­
mined the validity of the model.

It may also be that the

new research on the bi-dimensionality of the political
efficacy scale used herein as a unidimensional scale ex­
plains the lack of predictive power of the m o d e l .

It is too

early to conclude that the model is entirely invalid.

At

this point in time, all that can be said is that the model,
in this situation, did not operate as well as expected.
The third conclusion that can be drawn from the
study is that the school millage election is a low interest
election situation.

The low levels of both awareness and

correct information and the eventual low turnout all point
to an election setting where few people were informed, and
fewer still cared enough to vote.

This finding is consis­

tent with the literature on elections.

The school millage

does not attract the interest that a general election d o e s .
This is substantiated here.
Fourthly, attention to local media in a situation
where that media must compete against a large number of
voices from outside the community is low.

Shiawassee

County is a media-rich area with many radio stations, tele­
vision stations and newspapers available in the area.

This

plethora of voices, however, overpowers the local outlets.

219
With few people attending to the local media, information
regarding local issues and events will be limited.
seems to have occurred here.

This

Those who read the local paper

were much more likely to be aware of the election than those
who had no contact with the local media.
informed electorate,

The concept of an

then, may be in danger in areas similar

to Shiawassee County where much information is available
but little-to-none of it is community specific.
Fifth, a variety of demographic variables influence
favorability towards a school referendum issue.

This study

as well as others in the past agree that certain factors
will lead to a favorable or unfavorable vote.

School offi­

cials planning on using referenda need to be aware of the
presence or absence of these factors to determine the
approach to take in assuring passage of the proposal.
Sixth, alienation, defined as trust in government
and political efficacy, does not seem to play a large part
in the likelihood of a favorable vote or in an actual "yes"
vote when combined with a variety of demographic factors.
It is possible that the role assigned to alienation has
been overblown.

It is also possible, however, that alien­

ation works differently in a school millage election, gen­
erally, or that its impact, or lack thereof,
stance was situation specific.

in this in­

A better explanation, though,

based on recent research, is that the effect of alienation
was dissipated by the conceptualization of alienation used
in this study.

If alienation had been treated as three

220
separate variables, its true effect may have been noted.
Seventh, it appears advisable to localize polit­
ical attitude scales.

While no comparison data of general­

ized scales have been offered herein, recent research
points to the necessity of specific referents for political

4

attitudes.

Eighth, the source profiles point to the need for
research into a community when planning a campaign for a
school monetary proposal.
tween source types.

Favorability levels varied b e ­

School officials should be aware of

the sources to which their most favorable supporters attend
and to which their opponents attend and should plan their
information campaigns accordingly.
Finally, media planners for all types of elections
need to be aware of the media situations in which they
will be working.

In general elections, where the party

has a primary role to play, the media environment must be
understood so that party cues can be transmitted to the
voter.

In an area like Shiawassee County, this may mean

a reliance on personal and quasi-mass sources rather than
on mass media sources, as it appears that newspapers and
radio are not used by voters as a source of information.
In nonpartisan elections, the same advice can be offered,
however, in this situation exploration of media alternatives
may be more crucial as the party cue is absent.

221
Limitations
Several limitations need to be noted.
sample for the study was small.

First, the

This was the result of a

low response rate brought about by a number of factors.
Because this study was part of a larger study, it was
necessary to complete the primary portion of the question­
naire before asking

the respondent for further information.

The first portion of the questionnaire was difficult and
lengthy.

Respondents may not have been willing to com­

plete the instrument because of this and therefore were not
included.

Problems were also found with the voter regis­

tration lists.

The lack of telephone numbers and the high

incidence of incorrect addresses led to the inability of the
interviewers to locate sample m e m b e r s .

The combination of

these two factors resulted in the samll sample size.
The small sample size leads to a second limitation
of the study.

In some of the statistical tests, the sub­

groups being compared are extremely small.
then, are highly unstable.

The statistics,

Significance, or a lack thereof,

may be the result of the smallness of the individual cells
rather than a true relationship, or a lack thereof, in the
population as a whole.
extremely difficult,

Generalization to the population is

then, because of the limited number of

respondents.
Thirdly, no attempt was made to equalize the number
of respondents in the four cells of the m o d e l .

Respondents

were assigned on the basis of an arbitrary division of the

222
scale scores at the mid-point with no consideration given
to the number of respondents in each.

This uneven cell

size may have influenced the results in unknown ways.
A fourth limitation is sample-related as well.
As noted in Chapter IV, the sample is not representative
of the county as a whole.

Two school districts are not

represented at all and the others are not represented
proportionate to their size in the population.

The sample

also underrepresents the rural community and men and over­
represents those over the age of sixty-five.

The biases

that may have resulted from these factors are unknown and
limit the generalizability of the results to the county as
a whole or to other locations.
The fifth limitation of the study is the setting
of the study.

Attempting to test a derivative model for the

first time in a low-information,
setting may be problemmatic.

low-interest election

As the situation is so dif­

ferent from the more common general election, the results,
or lack thereof, may not be generalizable.

The fact that

so few people were aware of the election and fewer still
had correct information, made the testing of the model
difficult at best.

The very real possibility that the Shia­

wassee County school millage election was a deviant situation
imits the usability of the results of the study in differ­
ent situations.

223
Suggestions for Future Research
There are several suggestions for future research
that can be made.

First,

it seems most crucial that the

Milbrath and Goel model of political participation and
political alienation, defined as trust in government and
political efficacy, be tested in a research setting.

The

model needs to be subjected to empirical testing as a whole
to see if the proposed relationships can find support in
data drawn from a single study.
Second, the derivative model needs to be tested in
a variety of situations.

It should be tested in a more

isolated area, one where little outside media spill in.
It should also be tested in a media-rich area where the
level of local media is quite high, a metropolitan setting,
for example.

It also needs to be tested in a general

election setting, either Congressional or Presidential or
both, as well as in local-only elections and school elec­
tions.

Only after such extensive testing has been accom­

plished can the validity of the model be judged.
Along with the second suggestion, the model needs
to take the bi-dimensionality of political efficacy into
consideration.

This can be accomplished by eliminating the

items found to correlate with internal efficacy and including
only those items found to relate to external efficacy.

The

model could also be reconceptualized as a four-by-two table
with two levels of trust in government and two levels each
of internal and external political efficacy.

In this way,

224
the varying relationships among the three variables could
be determined as well as their individual impacts on commun­
ication and political participation variables.
Fourth, more research needs to be accompliches using
situation-specific attitude scale items.

The trust in

government and political efficacy items need to be localized
and specialized to meet the needs of particular research
questions.

The scales need to be investigated to see it

their unidimensionality no longer holds in specific as well
in generic situations.
Fifth, research into media use in school elections
needs to be explored.
in this area.

Little research has been undertaken

Focus must be placed on the role of mass,

quasi-mass and personal sources as means of information
acquisition.

This may be the most crucial need.

As media

become more specialized and with the advent and widespread
adoption of cable television,
media may no longer be valid.

the theories based on mass
New ways of looking at

communication media need to be developed in order to under­
stand the impact of these new media on the electorate.
Finally, more emphasis needs to be placed on the
role alienation plays in voting, media use and communication
patterns.

It would appear that alienation may not be as

important as previous research has indicated.
may exert a greater influence on the voter.
change as situations change.

Other factors
Concepts must

This may be the case with

alienation, that the theories and suppositions that held

225
several decades ago are no longer valid.

This needs to be

determined if the body of literature on alienation is to be
kept current and generalizable.

226
^Ada W. Finifter, "Dimensions of Political Alienation," American Political Science Review 65 (June 1970):
389-410.
?
Lester W. Milbrath and M. L. Goel, Political
Participation: How and Why People Get Involved in Politics
2nd ed. , (Chicago: Rand McNally College Publishing Company,
1977).
3

See Joel Aberbach and Jack Walker, "Political Trust
and Racial Ideology," American Political Science Review
64 (December 1970):1199-1219; Jack Citrin, "Comment: The
Political Relevance of Trust in Government," American
Political Science Review 68 (September 1974):973-88;
Dwight G. Dean, "Alienation and Political Apathy," Social
Forces 38 (March 1960):185-95; and, Herbert McClosky,
"Consensus and Ideologies in American Politics," American
Political Science Review 58 (June 1964):361-83.
^See Paul R. Abramson and Ada W. Finifter, "On the
Meaning of Political Trust: New Evidence from Items Intro­
duced in 1978," American Journal of Political Science 25
(May 1981):297-307; and, Bert Useem, "Trust in Government
and the Boston Anti-Busing Movement," The Western Political
Quarterly 35 (March 1982):81-91.

APPENDIX 1

FIRST WAVE QUESTIONNAIRE

SHIAWASSEE COUNTY CENTRALIZED
VOCATIONAL EDUCATION QUESTIONNAIRE
Respondent Name _____________________________
Telephone Number ____________________________
Callbacks
__________________ ______ _____
Date
Time
Interviewer
Hello, I'm
____________
^__ calling from Michigan State
University.
We are conducting a survey of attitudes toward
vocational education in Shiawassee County and would like to
ask you a few questions. You have been randomly selected
from the county voter registration lis t s . Of course your
participation is completely voluntary and your answers will
be confidential and used only in combination with the
answers of othe r s . Your cooperation is extremely important
to insure adequate representation of countywide attitudes
and to make the survey a success.
I can call back at a more
convenient time (MAKE APPOINTMENT IF NECESSARY),
In the questions that follow I'll give you a pair of words
or ideas that persons such as yourself from Shiawassee
County have used to describe vocational education. Then,
you'll ask yourself how different each of the pair is from
the other according to your opinion. If you think that the
two items are very similar, tnen you could say that they are
very close together.
If you think that they are very dif­
ferent, then you could say they are very far apart.
In­
stead of using words like "close togther" or "far apart" you
could easily use a number.
Take cold rain and war m sunshine for example. If you think
that these two are similar, seem to go together, or are
often associated with each other, then you would say a
small number.
On the other hand, if you think that cold
rain and warm sunshine are different, do not seem to go
together, or are seldom associated with each other, then you
would say a large number.
Thus a large number would mean that two items are very dif­
ferent; a small number would mean that the two are very
similar.
Zero would mean that they are the same, identical,
that there is no difference between them.
In order to help make your estimates more accurate, I ’ll
give you a mental r u l e r . Think of what "practical train­
ing" means to you. Now think of "work skills." Think of the
difference between "practical training” and "work skills."
Call t h i s difference t e n ! In other w o r d s , we'll say that
pratical training and work skills are ten units a p a r t ;
that the distance between them is now ten on your mental
ruler.
Keep this difference ten in mind and compare the
227

228
other pairs to i t . You will compare each pair of items to
the difference between practical training and work skills
which is now ten.
Also, use any number that accurately reflects your opinion.
You may use decimals, fractions, or whole numb e r s . Zero
is the lower limit and there is no upper limit.
Of any pair
cannot be estimated, just tell me to go on the the next.
Here are a few practice pairs.
Remember, the difference
between pratical training and work skills is t e n .
How far apart are:

(**WRITE RESPONSES ON

RESPONSE SHEET**)

Cold rain and w arm sunshine? ....
Good and bad?.....................
Good and Practical Training? ....
Work Skills and Necessary?

....

Concerned and Your Friends? ....
Happy and Yourself?

....

Would you like to review any part of the instructions?
Also, any time that you would like the instructions reviewed,
just ask.
Remember,
opinion.

there are no right or wrong answers only your

Are you ready to begin? OK.
In other w o r d s , if you think that the items are farther apart
than Practical Training and Work Skills then you would say
a number larger than ten.
If, in your opinion, the two
items are closer together than Practical Training and
Work Skills, then you would use a number smaller than ten.

PRACTICE QUESTIONS
Cold rain and warm sunshine____________________
Good and bad
Good and practical training
Work skills and necessary
Concerned and your friends
Happy and yourself
Remember:

________
________
________
________
________
________

If practical training and work skills are ten units
apart, in your opinion how far apart are the
following pairs?

1.
2.
3.
4.
5.

Job opportunities and practical training? ________
Job opportunities and work skills__________________
Job opportunities and college
'
Job opportunities and expensive
________
Job opportunities and centralized
vocational education________________________ ________
6 . Job opportunities and tax increase_________ ________
7. Job opportunities and duplication of
programs_____________________________________ ________
8 . Job opportunities and needed_______________ ________
Remember:
9.
10.
11.
12.
13.
14.
15.
16.

Job opportunities and discipline problems _________
Job opportunities and intermediate
school district_____________________________ ________
Job opportunities and your vote for_______ ________
Job opportunities and your vote against
________
Practical training and work skills________ ________
Practical training and college_____________ ________
Practical training and expensive___________________
Practical training and centralized
vocational education

Remember:
17.
18.
19.
20.
21.
22.
23.
24.

If practical training and work skills are ten
units apart, in your opinion how far apart are:

If practical training and work skills are ten
units a p a r t , in your opinion how far apart a r e :

Practical training and tax increase_______ ________
Practical training and duplication of
programs_____________________________________ ________
Practical training and needed______________________
Practical training and discipline
problems_____________________________________ ________
Practical training and intermediate
school district_____________________________ ________
Practical training and your vote for______ ________
Practical training and your vote against
_______
Work skills and college____________________ ________

230
Remember:
25.
26.
27.
28.
29.
30.
31.
32.

Work skills and expensive
' _____
Work skills and centralized vocational
education____________________________________ ________
Work skills and tax increase
'
______
Work skills and duplication of programs
Work skills and needed_____________________ ________
Work skills and discipline problems
'
Work skills and intermediate school
district_____________________________________ ________
Work skills and your vote for______________ ________

Remember:
33.
34.
35.
36.
37.
38.
39.
40.

44.
45.
46.
47.
48.

52.
53.
54.

If practical training and work skills are ten
units apart, in your opinion how far apart are:

College and your vote for
'
College and your vote against______________ ________
Expensive and centralized vocational
education
'
Expensive and tax increase_________________ ________
Expensive and duplication of programs
.....
Expensive and needed
'
Expensive and discipline problems__________________
Expensive and intermediate school
district
'

Remember:
49.
50.
51.

If practical training and work skills are ten
units apart, in your opinion how far apart are:

Work skills and your vote against_________ ________
College
and expensive
'
College
and centralized vocational
education____________________________________ ______ __
College and tax increase
'
College and duplication of programs_______ ________
College and needed__________________________ ________
College and discipline problems____________ ________
College and intermediate school district
'

Remember:
41.
42.
43.

If practical training and work skills are ten
units apart, in your opinion how far apart are:

If practical training and work skills are ten
units a p a r t , in your opinion how far apart a r e :

Expensive and your vote for________________ ________
Expensive and your vote against____________________
Centralized vocational education and
tax increase_________________________________________
Centralized vocational education and
duplication of programs_____________________________
Centralized vocational education and
needed_______________________________________ ________
Centralized vocational education and
discipline problems_________________________ ________

231
55.
56.

Centralized vocational education and
intermediate school district
'
Centralized vocational education and
your vote for_______________________________ ________

Remember:
57.
58.
59.
60.
61.
62.
63.
64.

If practical training and work skills are ten
units apart, in your opinion how far apart are:

Centralized vocational education and
your vote against
Tax increase and duplication of programs
Tax increase and needed
Tax increase and discipline problems______
Tax increase and intermediate school
district
Tax increase and your vote for
Tax increase and your vote against________
Duplication of programs and needed

Remember:

' _____
______
'
________
'
'
________
'

If practical training and work skills are ten
units apart, in your opinion how far apart are:

65.

Duplication of programs and discipline
problems_____________________________________ ________
6 6 . Duplication of programs and intermediate
school district_____________________________ ________
67. Duplication of programs and your vote for '
6 8 . Duplication of programs and your vote
against______________________________________ _____ '
69. Needed and discipline problems______________ _ _ ____
70. Needed and intermediate school district
________
71.
Needed and your vote for___________________ _____ _
72. Needed and your vote against_______________ ________
Remember:
73.
74.
75.
76.
77.
78.

If practical training and work skills are ten
units apart, in your opinion how far apart are:

Discipline problems and intermediate
school district_____________________________ ________
Discipline problems
and your vote for
________
Discipline problems
and your vote against _______
Intermediate school
district and your
vote for_____________________________________ ________
Intermediate school
district and your
vote against______________________________________ _
Your vote for and your vote against
’
_____

Now I have just a few more questions:
79.

Do any other ideas or feelings about vocational
education come into your mind, besides those

232
which we have already mentioned?
80.

Who do you usually talk to to get
information about school-related topics?
PROBE IF NECESSARY.

81.

What do they do for a living?

82.

Do you regularly read a newspaper?______________ _
(1)

83.

YES

(2)

Which One?

NO

(RECORD 1 FOR ALL MENTIONED)

Detroit Free
Detroit News
Owosso Argus
84.

2

3

4

5

6

7

8

'

9+

2

3

4

5

6

7

8

9+

Which radio station do you usually________ ______
listen to?
(1)

87.

Press

Lansing State Journal
___ Flint Journal
Other:

About how many hours of radio did_________ ______
you listen to yesterday?
1

86.

Press

About how many hours of television did
you watch yesterday?
1

85.

(GO TO Q.84)

WOAP

OTHER: __________________

There are different ways that high school ______
level Vocational Education can be offered
--by the local high school, or by a co­
operative effort between school districts.
Which of these do you prefer -- the local
high school or a cooperative effort between
scnool districts?
(1)
(2)
(3)
(4)
(5)

LOCAL HIGH SCHOOL
COOPERATIVE EFFORT AMONG SCHOOL DISTRICTS
BOTH
C A N ’T SAY
OTHER: ______________________

233
88.

If the local high school did not have
_
enough money or enough students to
offer training in several kinds of Voc a ­
tional skills, would you be in favor of a
centralized Vocational Program to serve all
of the high schools in the County.
(1)
(2)
(3)

89.

Would you approve of the Shiawassee
Intermediate School District planning,
organizing and operating a centralized
Vocational Program for all high schools
in the county?
(1)
(2)
(3)

90.

YES
NO
CAN'T SAY

YES
NO
CAN'T SAY

If additional millage, up to 2 mill(s)
were necessary to finance a centralized
Vocational Education Program, based on
how you feel now, would you vote for or
against such a millage request?
(1)
(2)
(3)

_

_

FOR
AGAINST
UNDECIDED/CAN'T SAY

91.

If there was a millage election for a
_
centralized Vocational Education Program,
what kinds of information would you like to
know about the millage request before the
election?

92.

What school district are you in?
(1)
(2)
(3)
(4)

93.

Byron
Corunna
Durand
Laingsburg

(5)
(6)
(7)
(8)

New Lothrop
Morrice
Owosso
Perry

How much education have you completed?
(1)
(2)
(3)
(4)

8th Grade or less
(5)
(GO TO Q .95)
(6)
9th-12th Grade
(7)
H . S . Degree
(8)
Trade School After H.S.

Some College
College Degree
Graduate Work
Graduate Degree

234
94.

Did you ever take a Vocational Education
class in high school?
(1)

95.

YES

(2)

NO

Do you have any children?
(1)

YES

(2)

NO

(GO TO Q.98)

96.

What are their ages? (RECORD TO THE
NEAREST YEAR)

97.

IF CHILDREN 13 Y R S , OR OLDER:
Have your ______
children ever taken a vocational education
class in high school?
(1)

98.

YES

(2)

NO

Would you tell me your age?

'

(RECORD TO NEAREST YEAR)
99.

Do you currently have either a full or a
part-time job?
(1)

Full time

(2)

Part Time

100.

What is your occupation?

101.

Do you work in Shiawassee County
or somewhere else?
(1) Shiawassee
(GO TO Q.103)

102.

103.

(3) NO
(GO TO Q. 103)

(2) Somewhere Else

What county do you work in?
(1)
(2)
(3)

Ingham (Lansing)
Genesee (Flint)
Other:
'

Is your total household income more or
less than $10,000?
(1)
(-)

Less than $10,000 (GO TO Q.104)
More than $10,000
103a.

Is it more than $20,000?
(2)
(-)

No (GO TO Q.104)
Yes

______

_____ _

235
103b.

And finally, is it
more than $30,000?
(3)
(4)

(5)
(6)
(7)
(8)
104.

No (GO TO Q.104)
Yes (GO TO Q.104)

Student
Unemployed
E.etired
Refused

RECORD SEX:

(1)

MALE

(2)

(Thank you very much for your time.
your help.

FEMALE
We appreciate

APPENDIX 2

MAIN WAVE QUESTIONNAIRE

236
INTRODUCTION AND OUESTIONS 1 THROUGH 78 IDENTICAL TO
FIRST WAVE QUESTIONNAIRE.

Now I'm going to read you some things that people have
told me when I have interviewed them and ask you to tell
me if you agree or disagree with them.
After I read each
statement, just tell me whether you agree or disagree with
it.
79.

People like me don't have any say about
what the School Board d o e s .
AGREE
DISAGREE
DEPENDS
DON'T KNOW
NO RESPONSE

80.

Voting is the only way people like me can
have any say about how the School Board runs
things.
AGREE
DISAGREE
DEPENDS
DON'T KNOW
NO RESPONSE

81.

Sometimes school politics seems so compli­
cated that a person like me can't really
understand what's going on.
AGREE
DISAGREE
DEPENDS
DON'T KNOW
NO RESPONSE

82.

I don't think school officials care much
what people like me think.
AGREE
DISAGREE
DEPENDS
DON'T KNOW
NO RESPONSE

____

237
83.

Generally speaking, those people we elect
to the School Board loose touch with the
people pretty quickly.
AGREE
DISAGREE
DEPENDS
DON'T KNOW
NO RESPONSE

84.

School Board members are only interested in
people's votes and not in their opinions.
AGREE
DISAGREE
DEPENDS
DON'T KNOW
NO RESPONSE

Now I have a few more questions along this same line that
I'd like to ask you.
85.

First of all, do you think that people on the ___
School Board waste a lot of money we pay in
taxes, waste some of it, or don't waste very
much of it at all?
WASTE A LOT
SOME OF IT
NOT VERY MUCH AT ALL
DON'T KNOW
NO RESPONSE

86.

How much of the time do you think you can
trust the School Board to do what is right -just about always, most of the time, or only
some of it?

“
87.

JUST ABOUT ALWAYS
MOST OF THE TIME
SOME OF IT
DON'T KNOW
NO RESPONSE

Would you say the School Board is pretty much
run by a few big interests looking out for
themselves or that it is run for the benefit
of all the people?
A FEW BIG INTERESTS
BENEFIT OF ALL
DON'T KNOW
NO RESPONSE

238
88.

Do you feel almost all of the people
running the School Board are smart people or
do you think that quite a few of them don't
seem to know what they're doing?
SMART PEOPLE
DON'T KNOW WHAT THEY'RE DOING
DON'T KNOW
NO RESPONSE

89.

Do you think quite a few of the people
running the School Board are crooked, not
very many are crooked, or do you think
hardly any of them are crooked?
QUITE A FEW
NOT VERY MANY
HARDLY ANY
DON'T KNOW
NO RESPONSE

90.

Are you aware of any upcoming special
county-wide elections in Shiawassee County?
YES (GO TO 90a)
NO
(GO TO 90b)
90a.

IF YES: Who has called that countywide election:
is it the County
Attorney, the Intermediate School
District, the State Department of
Education or someone else?
COUNTY ATTORNEY
ISD
STATE ED DEPT.
SOMEONE ELSE
OTHER
NO RESPONSE

90b.

IF NO:
Has the Shiawassee County
Intermediate School District called
any special elections for the near
future?
YES (GO TO 91)
NO
(GO TO 93)

239
91.

What is that county election about -- a
millage for a new grade school, a millage
for area vocational technical education or
a millage for the expansion of the special
education programs?
NEW GRADE SCHOOL
AREA VOCATIONAL TECHNICAL EDUCATION
SPECIAL EDUCATION
OTHER
NO RESPONSE

92.

How much millage is
county-wide millage
less than one mill,
mills, or more than

being asked in that
election -- is it
between one and two
two mills?

LESS THAN ONE
BETWEEN ONE AND TWO
MORE THAN TWO
OTHER
REFUSED
MEDIA
93.

Do you regularly read a newspaper?
YES
93a.

NO

(GO TO 94)

Which newspaper or papers would that be?
DETROIT FREE PRESS
DETROIT NEWS
OWOSSO ARGUS PRESS
LANSING STATE JOURNAL
FLINT JOURNAL
"OTHERS
(SPECIFY: _________________ )

94.

Which newspapers do you subscribe to or
receive at your home?
DETROIT FREE PRESS
DETROIT NEWS
OWOSSO ARGUS PRESS
STATE JOURNAL
FLINT JOURNAL
OTHERS (SPECIFY: _________________ )

95.

How many hours a day to you listen to the
radio?
1

2

3

4

5

6

7

OR MORE

240
96.

Which radio station do you usually listen to?
WOAP
96a.

97.

LIST OTHER STATIONS MENTIONED
(GO TO 97)

Is that WOAP AM, FM or both?
AM
FM
__ BOTH

Since last Monday, have you heard anything
about Area Vocational Technical Education?
YES (GO TO 97a)
97a.

IF YES:

NO

Was it a poster?
YES

97f.

NO

Was it in a flyer or brochure?
YES

97e.

NO

From a Friend?
YES

97d.

NO

On the radio?
YES

97c.

(GO TO 98)

Was that in the newspaper?

YES
97b.

NO

NO

Was it at a meeting?
YES

NO

VOTING
98.

How many times have you voted in a general
election in the past two years?

99.

How many times have you voted in a school
election in the past two years?

100. When was the last time you voted in a school
millage election?
(IF NONE, GO TO 102)

241
101.

Would you tell me how you voted in that
school millage election?
FOR
REFUSED

AGAINST

DON'T KNOW

During the school election on June 11, voters in Shiawassee
County will be asked to vote for or against a 1.95 m i l l ­
age to finance area vocational technical education.
102.

Do you think you will vote in that election?
YES
NO
(GO TO 104)
DON'T KNOW
REFUSED

103.

If 0 means no chance and 100 means definitely
in favor, what's the chance that you will vote
in favor of the millage?
"DON'T KNOW
"REFUSED

104,

What school district are you in?
BYRON
"CORUNNA
"DURAND
"LAINGSBURG

105

NEW LOTHROP
MORRICE
"OWOSSO
"PERRY
"OTHER

Do you own or rent the place you are
living in?
OWN
"RENT
"OTHER (SPECIFY:

106

How much education have you completed?
SOME COLLEGE
8TH GRADE OR LESS
"COLLEGE DEGREE
" (GO TO Q. 107)
9TH-12TH GRADE
"GRADUATE WORK
"GRADUATE DEGREE
"H.S. DEGREE
"TRADE SCHOOL AFTER HTS"

107,

Did you ever take a vocational education
class in high school?
YES

NO

____

242
108.

Do you have any children that currently_______ __
live at home with you?
YES

109.

NO

REFUSED

Have your children ever taken a vocational
education class during high school?
YES

110.

NOT MARRIED
~7&5 TO 110)

NO

__

CHILDREN NOT IN HIGH
SCHOOL YET
__

Would you tell me your age?
RECORD TO NEAREST YEAR

111.

Do you currently have either a full or a
part-time job?
FULL TIME

PART TIME

NO
CGO TO 114)

112.

What is your occupation?

113.

Do you work in Shiawassee County or some­
where else?

__

SHIAWASSEE
SOMEWHERE ELSE
114.

Is your total household income more or less
than $10,000?
LESS THAN $10,000 (GO TO Q.115)
MORE THAN $10,000
114a.

Is it more than $20,000?
NO (GO TO Q.115)
YES
114b.

And finally, is it more
than $30,000?
NO
(GO TO 115)
YES (GO TO 115)

STUDENT
'UNEMPLOYED
'RETIRED
REFUSED

___

243
115.

RECORD SEX:

(1) MALE

(2) FEMALE

(Thank you very much for your time.
h e l p .)

_

We appreciate your

APPENDIX 3

CALL-BACK QUESTIONNAIRE

244
SHIAWASSEE PROJECT

CALL-BACK QUESTIONNAIRE

INTRODUCTION:
Hello, my name is
, calling from Michigan State
University.
We're doing a study of voting behavior in
Shiawassee County and I have a few questions I'd like to
ask you.
This will just take a few minutes and I'd appre­
ciate your help.
1.

First, an election was held in Shiawassee County on
June 11. Would you tell me what was on that ballot?
(IF DOESN'T MENTION AREA VOCATIONAL TECHNICAL EDUCA­
TION, OR SOMETHING SIMILAR TO THAT, PROBE.)
IF MENTIONS A V T E , CHECK YES.
IF DOESN'T, CHECK NO.

2.

How much millage was being asked for the Area Voca­
tional Education Center?
IF RESPONDS 1.95 OR BETWEEN ONE AND TWO MILL, CHECK
YES.
ANY OTHER RESPONSE, CHECK NO.
IF DON'T KNOW, CHECK DK.

3.

During the past few months did you hear anything
about Area Vocational Technical Education?
IF NO, GO TO QUESTION 4.
IF YES, Where did you hear about it?
CHECK APPROPRIATE COLUMN:

4.

NEWSPAPER, RADIO,
TELEVISION, CITIZEN'S
COMMITTEE, FRIENDS/
NEIGHBORS, BROCHURES

Were you contacted by the Intermediate School Dis­
trict or the Citizen's Committee regarding voting in
the June 11 election for the Area Vocational Tech­
nical Education Center?
IF YES, CHECK YES.
IF NO, CHECK NO.

5.

Did you vote in the election on June 11?
IF YES, CHECK YES.
IF NO, CHECK NO.

(GO TO 6)

245
6.

Would you please tell me whether you voted for or
against the millage for the Area Vocational
Technical Education CEnter?
IF FOR, CHECK FOR.
IF AGAINST, CHECK AGAINST.

Thank you very much for your time.
appreciated.

Your help is greatly

APPENDIX 4

CODEBOOK

246
SHIAWASSEE COUNTY CENTRALIZED
VOCATIONAL EDUCATION CODEBOOK
CARD 1
COLS.
01-08

QUESTION
NUMBER
Identification number
Col.
01

Sex
(1)
(2)

Male
Female

02

School District
(1) Byron
(2)
Corunna
(3) Durand
(4) Laingsburg
(5) New Lothrup
(6) Morrice
(7)
Owosso
(8) Perry
(9) Other

03

Vote
(1)
(2)
(3)
(4)

04-06

For
Against
Undecided/Can't Say
No Response

Questionnaire ID Number
Card Number
(00) Card 1
(01) Card 2
(02) Card 3
Card 4
(03)
Card 5
(04)
Card 6
(05)
Card 7
(06)
Card 8
(07)
Card 9
(08)
Card 10
(09)
(10) Card 11
(11) Card 12

247
PRACTICE QUESTIONS
09-12
13-17
18-21
22-26
27-30
31-35
36-39
40-44
45-48
49-53
54-57
58-62

2021
Cold rain and warm sunshine
2223
Good and bad
2425
Good and practical training
2627
Work skills and necessary
2829
Concerned and your friends
3031
Happy and yourself

CARD 2
01-08
09-12
13-17
18-21
22-26
27-30
31-35
36-39
40-44
45-48
49-53

1
2
3
4
5

54-57
58-62
63-66
67-71

6

72-75
76-80

8

7

Identification Number
0102
Job opportunities and
training
0103
Job opportunities and
0104
Job opportunities and
0105
Job opportunities and
0106
Job opportunities and
vocational education
0107
Job opportunities and
0108
Job opportunities and
programs
0109
Job opportunities and

(SEE ABOVE)
practical
work skills
college
expensive
centralized
tax increase
duplication of
needed

CARD 3
01-08
09-12
13-17

9

18-21
22-26

10

27-30
31-35
36-39
40-44
45-48
49-53

11
12
13

Identification Number (SEE ABOVE)
0110
Job opportunities and discipline
problems
0111
Job opportunities and intermediate
school district
0112
Job opportunities and your vote for
0113
Job opportunities and your vote against
0203
Practical training and work skills

248
54-57
58-62
63-66
67-71
72-75
76-80

14
15
16

0204
Practical training and college
0205
Practical training and expensive
0206
Practical training and centralized
vocational education

CARD 4
01-08
09-12
13-17
18-21
22-26

17
18

27-30
31-35
36-39
40-44

19

45-48
49-53

21

54-57
58-62
63-66
67-71
72-75
76-80

22

20

Identification Number (SEE ABOVE)
0207
Practical training and tax increase
0208
Practical training and duplication of
programs
0209
Practical training and needed

0210
Practical training and discipline
problems

0211
Practical training and intermediate
school district

23
24

0212
Practical training and your vote for
0213
Practical training and your vote against
0304
Work skills and college

CARD 5
01-08
09-12
13-17
18-21
22-26
27-30
31-35
36-39
40-44
45-48
49-53
54-57
58-62
63-66
67-71
72-75
76-80

25
26
27
28
29
30
31
32

Identification Number (SEE ABOVE)
0305
Work skills and expensive
0306
Work skills and centralized vocational
education
0307
Work skills and tax increase
0308
Work skills and duplication of programs
0309
Work skills and needed
0310
Work skills and discipline problems
0311
Work skills and intermediate school
district
0312
Work skills and your vote for

249
CARD 6
01-08
09-12
13-17
18-21
22-26
27-30
31-35
36-39
40-44
45-48
49-53
54-57
58-62
63-66
67-71
72-75
76-80

33
34
35
36
37
38
39
40

Identification Number (SEE ABOVE)
0313
Work skills and your vote against
0405
College and expensive
0406
College and centralized vocational
education
0407
College and tax increase
0408
College and duplication of programs
0409
College and needed
0410
College and discipline problems
0411
College and intermediate school
district

CARD 7
01-08
09-12
13-17
18-21
22-26
27-30
31-35
36-39
40-44
45-48
49-53
54-57
58-62
63-66
67-71
72-75
76-80

41
42
43
44
45
46
47
48

Identification number (SEE ABOVE)
0412
College and your vote for
0413
College and your vote against
0506
Expensive and centralized vocational
education
0507
Expensive and tax increase
0508
Expensive and duplication of programs
0509
Expensive and needed
0510
Expensive and discipline problems
0511
Expensive and intermediate school
district

CARD 8
01-08
09-12
13-17
18-21
22-26
27-30
31-35

49
50
51

Identification Number (SEE ABOVE)
0512
Expensive and your vote for
05i3
Expensive and your vote against
0607
Centralized vocational education and
tax increase

250
36-39
40-44

52

45-48
49-53

53

54-57
58-62

54

63-66
67-71

55

72-75
76-80

56

0608
Centralized vocational education
duplication of programs
0609
Centralized vocational education
needed
0610
Centralized vocational education
discipline problems
0611
Centralized vocational education
intermediate school district
0612
Centralized vocational education
your vote for

and
and
and
and
and

CARD 9
01-08
09-12
13-17

57

18-21
22-26

58

27-30
31-35
36-39
40-44
45-48
49-53

59

54-57
58-62
63-66
67-71
72-75
76-80

60
61
62
63
64

Identification Number (SEE ABOVE)
0613
Centralized vocational education and
your vote against
0708
Tax increase and duplication of
programs
0709
Tax increase and needed
0710
Tax increase and discipline problems
0711
Tax increase and intermediate school
district
0712
Tax increase and your vote for
0713
Tax increase and your vote against
0809
Duplication of programs and needed

CARD 10
01-08
09-12
13-17

65

18-21
22-26

66

27-30
31-35

67

36-39
40-44

68

Identification Number (SEE ABOVE)
0810
Duplication of programs and discipline
problems
0811
Duplication of programs and intermediate
school district
0812
Duplication of programs and your
vote for
0813
Duplication of programs and your
vote against

251
45-48
49-53
54-57
58-62

69

63-66
67-71
72-75
76-80

71

70

72

0910
Needed and
0911
Needed and
district
0912
Needed and
0913
Needed and

discipline nroblems
intermediate school
your vote for
your vote against

CARD 11
01-08
09-12
13-17

73

18-21
22-26
27-30
31-35

74

36-39
40-44

76

45-48
49-53

77

54-57
58-62

78

75

Identification Number (SEE ABOVE)
1011
Discipline problems and intermediate
school district
1012
Discipline problems and your vote for
1013
Discipline problems and your vote
against
1112
Intermediate school district and your
vote for
1113
Intermediate school district and your
vote against
1213
Your vote for and your vote against

CARD 12
Identification Number (SEE ABOVE)
People like me have no say about
what the School Board does
(1) Agree
(2) Disagree
(3) Depends
(4) Don't Know
(5) No Response

01-08
09

79

10

80

Voting only way tohave
(1) Agree
(2) Disagree
(3) Depends
(4) Don't Know
(5) No Response

11

81

School
politics toocomplicated
(1) Agree
(2) Disagree
(3) Depends
(4) Don't Know
(5) No Response

say

School officials don't care what people
think
(1) Agree
(2) Disagree
(3) Depends
(4) Don't Know
(5) No Response
People on School Board loose touch
(1) Agree
(2) Disagree
(3) Depends
(4) Don't Know
(5) No Response
School Board only interested in votes
(1) Agree
(2) Disagree
(3) Depends
(4) Don't Know
(5) No Response
School Board wastes tax monies
(1) Waste a lot
(2) Some of it
(3) Not very much at all
(4) Don't Know
(5) No Response
Trust School Board to do right
(1) Just about always
(2) Most of the time
(3) Some of it
(4) Don't Know
(5) No Response
School Board run by a few big interests
(1) A few big interests
(2) Benefit of all
(4) Don't Know
(5) No Response
People running School Board are smart
people
(1) Smart people
(2) Don't know what they're doing
(4) Don't know
(5) No Response

253
19

89

People running School Board are crooked
(1) Quite a few
(2) Not very many
(3) Hardly any
(4) Don't know
(5) No Response

20

90

Any
(1)
(2)
(9)

upcoming county-wide elections
Yes
No
No response

21

90a

Who
(1)
(2)
(3)
(4)
(5)
(8)
(9)

called election
County Attorney
ISD
State Ed. Dept.
Someone else
Other
NA
No response

22

90b

Has
(1)
(2)
(8)
(9)

SCISD called election
Yes
No
NA
No response

23

91

What's election about
(1) New grade school
(2) Area vocational technical education
(3) Special education
(4) Other-nursing home
(5) Other
(8) NA
(9) No response

24

92

How
(1)
(2)
(3)
(4)
(8)
(9)

much millage is being asked
Less than one
Between one and two
More than two
Other
NA
No response/refused

93

Read newspaper regularly
(1) Yes
(2) No
(9) No response

MEDIA
25

254
93a
26
27
28
29
30
31

Which one(s)
(below coded 1 if y e s , 0 if no)
Detroit Free Press
Detroit News
Owosso Argus Press
Lansing State Journal
Flint Journal
Others

(1)

(2)
(3)
(4)

(9) No response
94
32
33
34
35
36
37

Which newspapers subscribed to
(coded 1 if yes, 0 if no)
Detroit Free Press
Detroit News
Owosso Argus Press
State Journal
Flint Journal
Others

(1)

(2)
(3)
(4)
(9) No response/refused
38

95

Hours listen to radio
(0) None
(1) One
(2) Two
(3) Three
(4) Four
(5) Five
(6) Six
(7) Seven or more
(9) No response

39

96

Radio station listened to
(1) WOAP

(2)
(3)
(4)
(8) NA
(9) No response

255
40

96a

WOAP
(1)
(2)
(3)
(8)
(9)

41

97

Heard anything about area vocational
technical education
(1) Yes
(2) No
(8) Refused
(9) No response

42
43
44
45
46
47

97a
97b
97c
97d
97e
97f

48

98

AM, FM or Both
AM
FM
Both
NA
No response

Source
(coded 1 if yes, 2 if no, 8 if NA and
9 if no response)
In Newspaper
On radio
From a friend
Flyer or brochure
A poster
At a meeting
Times voted in a general election in
past two years
(0) None
(1) One
(2) Two
(3) Three
(4)
(8) Refused
(9) No response

49

99

50

100

Times voted in school election
(0) None
(1) One
(2) Two
(3) Three
(8) Refused
(9) No response
Last time voted in school election
(0) None

(1)
(2)
(3)
(8) Refused
(9) No response

How
(1)
(2)
(3)
(4)
(8)
(9)

voted
For
Against
Don't know
Refused
NA
No response

Likelihood of voting June 11
(1) Yes
(2) No
(3) Don't know
(4) Refused
(9) No response
Likelihood of voting in favor
Code number as given except:
(777) Refused
(888) NA
(999) No response
School district
(1) Byron
(2) Corunna
(3) Durand
(4) Laingsburg
(5) New Lothrup
(6) Morrice
(7) Owosso
(8) Perry
(9) Other
Own
(1)
(2)
(3)
(8)
(9)

or rent
Own
Rent
Other
Refused
No response

Respondent's education
(1) 8th grade of less
(2) 9th-12th grade
(3) H.S. degree
(4) Trade school after high school
(5) Some college
(6) College degree
(7) Graduate work
(8) Graduate degree
(9) No response/refused

257
59

107

Ever taken voc ed
(1) Yes
(2) No
(7) NA
(8) Refused
(9) No response

60

108

Children at home
(1) Yes
(2) No
(3) Not married
(8) Refused
(9) No response

61

109

Children taken voc ed
(1) Yes
(2) No
(3) Children not in high school yet
(7) Refused
(8) NA
(9) No response

62-63

110

Respondent's age
(code as given)

64

111

Employment status of respondent
(1) Full
(2) Part
(3) No
(8) Refused
(9) No response

65-67

112

Occupation of respondent
(000) Unknown job listed
(100) Housewife
(300) Retired
(400) Farmer
(500) Blue collar
(600) White collar
(700) Managerial
(800) Professional
(888) NA
(999) Refused

68

113

Work in Shiawassee County
(1) Shiawassee
(2) Somewhere else
(7) NA
(8) Refused
(9) No response

258
69

114

Total household income
(1) Less than $10,000
(2) More than $10,000 but less than

$20,000
(3) More than $20,000 but less than
$30,000
(4) More than $30,000
(5) Student
(6) Unemployed
(7) Retired
(8) Refused
(9) No response
70

115

Sex of respondent
(1) Male
(2) Female

SELECTED BIBLIOGRAPHY

SELECTED BIBLIOGRAPHY

Articles
Aberbach, Joel. "Alienation and Political Behavior."
American Political Science Review 69 (March 1969) :
:—
Aberbach, Joel, and Walker, Jack. "Political Trust and
Racial Ideology." American Political Science
Review 64 (December 1970):Il99-1219.
Abramson, Paul R . , and Aldrich, John H. "The Decline of
Electoral Participation in America." American
Political Science Review 76 (September 1982) :

fot-tt ;---------------

Abramson, Paul R . , and Finifter, Ada W. "On the Meaning of
Political Trust: New Evidence from Items Introduced
in 1978." American Journal of Political Science
25 (May 1981) :297-31)7':---------------------------Agger, Robert E.; Goldstein, Marshall N . ; and Pearl,
Stanley A. "Political Cynicism: Measurement and
Meaning." Journal of Politics 23 (August 1961):
477-508.
Alexander, Herbert E. "Communication and Politics: The Media
and the Message." Law and Contemporary Problems
34 (Spring 1969):255-77.
Anderson, James A., and Meyer, Timothy P. "Functionalism and
the Mass Media." Journal of Broadcasting 19
(Winter 1975):11-22:
Atkin, Charles K . ; Galloway, John; and Nayman, Oguz B.
"News Media Exposure, Political Knowledge and
Campaign Interest." Journalism Quarterly 53
(Summer 1970):231-37.
Balch, George I. "Multiple Indicators in Survey Research:
The Concept ’Sense of Political Efficacy'."
Political Methodology 1 (1974):1-43.
Baloyar, Enrique A. "Criticism, Cynicism and Political
Evaluation: A Venezuelan Example." American
259

260
Political Science Review 73 (December 1979):
987-1002.----------------Becker, Jerome D . , and Preston, Ivan L. "Media Use and
Political Activity." Journalism Quarterly 46
(Spring 1969):129-34.
Bennett, Stephen E., and Klecka, William R. "Social Status
and Political Participation: A Multivariate
Analysis of Predictive Power." Midwest Journal of
Political Science 14 (August 1970):355-82.
Blumler, Herbert. "Suggestions for the Study of Mass Media
Effects." In American Voting Behavior, pp. 197-208.
Edited by Eugene Burdick ana Arthur J. Brodbeck.
Glencoe, 111.: The Free Press, 1959.
Bogart, Leo, and Orenstein, Frank E. "Mass Media and Commun­
ity Identity in an Interurban Setting." Journalism
Quarterly 42 (Spring 1965):179-88.
Buchanan, William. "An Inquiry into Purposive Voting."
Journal of Politics 18 (May 1956):281-96.
Burstein, Paul. "Social Status and Individual Political
Participation in Five Countries." American Journal
of Sociology 77 (May 1972):1 0 8 7 - l K T
Campbell, Angus. "The Passive Citizen." In Political Opinion
and Electoral Behavior, pp. 175-88.
Edited by
Edward C. Dreyer and Walter A. Rosenbaum. Belmont,
Calif.: Wadsworth Publishing Company, 1966.
'Surge and Decline: A Study of Electoral Change."
Public Opinion Quarterly 24 (Fall 1960):397-418.
Cataldo, Everett F., and Holm, John D. "Voting on School
Finances: A Test of Competing Theories." The Western
Political Quarterly 36 (December 1983):619-31.
Chaffee, Steven H . , and McLeod, Jack M. "Individual vs.
Social Predictors of Information Seeking." Journalism
Quarterly 50 (Summer 1973):237-45.
Citrin, Jack. "Comment: The Political Relevance of Trust
in Government." American Political Science Review
68 (Septmeber 1974):973-88.
Clark, John P. "Measuring Alienation Within a Social System."
American Sociological Review 24 (December 1959):
-----------S4T-52.------Clarke, Peter, and Ruggels, L. "Preferences Among News

261
Media Coverage of Public Affairs." Journalism
Quarterly 47 (Autumn 1970):464-71.
Connelly, Gordon M . , and Field, Harry H. "The Non-Voter
Who He I s , What He T h i n k s ." Public Opinion
Quarterly 8 (Spring 1944):175-87.
Conway, M. Margaret.
"Voter Information in a Nonpartisan
Local Election." Western Political Quarterly
21 (March 1968): 69^77":
Cowart, Andrew T. "Electoral Choice in the American States:
Incumbency Effe c t s , Partisan Forces, and Divergent
Partisan Majorities." American Political Science
Review 67 (September 1973):835-52.
Craig, Stephen C. "Efficacy, Trust and Political Behavior:
An Attempt to Resolve a Lingering Dilemma."
American Politics Quarterly 7 (April 1979):225-39.
Craig,

Stephen C . , and Maggiotto, Michael A. "Political
Discontent and Political Action." Journal of
Politics 43 (May 1981):514-22.

Danielson, Wayne A. "Eisenhower's February Decision: A
Study of News Impact." Journalism Quarterly 33
(Fall 1956):431-41.
Dean, Dwight D., "Alienation and Political Apathy." Social
Forces 38 (March 1960):185-95.
________ . "Alienation: It's Meaning and Measurement."
American Sociological Review 26 (October 1961):
753-55.-----------------Deutschmann, Paul J . , and Danielson, Wayne A. "Diffusion
of Knowledge of the Major News Story." Journalism
Quarterly 37 (Summer 1960):345-55.
Dreyer, Edward C. "Media Use and Electoral Choices: Some
Political Consequences of Information Exposure."
Public Opinion Quarterly 35 (Winter 1971-72):
544-T3.
------Eckhardt, Kenneth, and Hendershot, Gerry.
"Transformation
of Alienation into Public Opinion." Sociological
Quarterly 8 (Autumn 1971):459-67.
Elden, J. Maxwell.
"Political Efficacy at Work: The Con­
nection Between More Autonomous Forms of Workplace
Organization and a More Participatory Politics."
American Political Science Review 75 (March 1981):
45-58.------------------------------

262
Eldersveld, Samuel J. "Experimental Propoganda Techniques
and Voting Behavior." American Political Science
Review 50 (March 1956):154-65.
Erbe, William. "Social Involvement and Political Activity:
A Replication and Elaboration." American Socio­
logical Review 29 (April 1964):198-215.
Eulau, Heinz, and Schneider, Peter. "Dimensions in Political
Involvement." Public Opinion Quarterly 20 (Spring
1956):128-42.
Finifter, Ada W. "Dimensions of Political Alienation."
American Political Science Review 64 (June 1970) :
39"5-"410.---------------------------Giles, Michael W . ; Gatlin, Douglas S.; and Cataldo,
Everett F. "Parental Support for School Referenda."
Journal of Politics 38 (May 1976):442-51.
Glaser, William A. "Television and Voting Turnout." Public
Opinion Quarterly 29 (SprjLng 1965) :71-86.
Greenberg, Bradley S. "Voting Intentions, Election Expecta­
tions and Exposure to Campaign Information."
Journal of Communication 15 (September 1965):
149-60.------------------________ . "Media Use and Believability: Some Multiple
Correlates." Journalism Quarterly 43 (Winter 1966):
665-70, 732.
Hamilton, Howard D. "The Municipal Voter: Voting and No n ­
voting in City Elections." American Political
Science Review 65 (December 1971):1135-40.
Hinckley. Barbara; Hofstetter, C. Richard; and Kessel,
John H. "Information and the Vote: A Comparative
Election Study." In Controversies in American
Voting Behavior, pp. 274-96. Edited by Richard
G. Niemi and Herbert F. Weisberg. San Francisco:
W. H. Freeman Company, 1976.
Horton, John E., and Thompson, Wayne E. "Powerlessness and
Political Negativism: A Study of Defeated Local
Referendums." American Journal of Sociology 67
(March 1962):485-93.
Iyengay, Shanto. "Subjective Political Efficacy as a Measure
of Diffuse Support." Public Opinion Quarterly 44
(Summer 1980):249-56.
Katz, Elihu, and Foulk e s , David. "On the Use of the Mass
Media as 'Escape': Clarification of a Concept."

263
Public Opinion Quarterly 26 (Fall 1962):377-88.
Key, V. 0., Jr. "Varieties of Participation and the Political
System." In Political Opinion and Electoral Behavior,
p. 51-56. Edited by Edward C. Dreyer and Walter A.
Rosenbaum. Belmont, Calif.: Wadsworth Publishing
Company, 1966.
Kline, F. Gerald. "Media Time Budgeting as a Function of
Demographics and Life Style." Journalism Quarterly
8 (Summer 1971):211-21.
Kramer, Gerald H. "Short-Term Fluctuations in U.S. Voting
Behavior." American Political Science Review 65
(March 1971JTT3TT4T:-------------------------Lang, Kurt, and Lang, Gladys Engel. "The Mass Media and
Voting." In American Voting Behavior, pp. 217-35.
Edited by Eugene Burdick and Arthur J. Brodbesk.
Glencoe, 111.: The Free Press, 1959.
Levin, Murray B., and Eden, Murray. "Political Strategy for
the Alienated Voter." Public Opinion Quarterly
26 (Spring 1962):47-63.
Litt, Edgar. "Political Cynicism and Political Futility."
Journal of Politics 25 (May 1963):312-23.
Meiklejohn, Alexander. "The First Amendment is an Absolute."
in Free Speech and Association: The Supreme Court
and The First Amendment, p p . 1-22. Edited by
Philip B. Kurland. Chicago: The University of
Chicago Press, 1975.
Mendelsohn, Harold. "Broadcast vs. Personal Sources of
Information in an Emergent Public Cris i s : The
Presidential Assassination." Journal of
Broadcasting 8 (Spring 1964):147-56.
Middleton, Russell. "Alienation, Race and Education."
American Sociological Review 28 (December 1963):
-----------973-7 T.------Miller, Arthur H. "Political Issues and Trust in Govern­
ment 1964-70." American Political Science Review
68 (September 1974):951-72.
_________. "Rejoinder to 'Comment' by Jack Citrin: Political
Discontent or Ritualism." American Political
Science Review 68 (September 1974):989-1001.
Miller, Warren E. "Disinterest, Disaffection, and Parti­
cipation in Presidential Politics." Political

264
Behavior 2 (1980):7-32.
Minar, David W. "The Community Basis of Conflict in
School System Politics." American Sociological
Review 31 (December 1966):822-35.
McClosky, Herbert. "Consensus and Ideology in American
Politics." American Political Science Review 58
(June 1964)73^1^3':
McCombs, M. E., and Mullins, L. E. "Consequences of Educa­
tion, Media Exposure, Political Interest and Infor­
mation-Seeking Orientations." Mass Communications
Review 1 (August 1973):27-31.
McDill, Edward L., and Ridley, Jeanne Clare. "Status, Anomia,
Political Aleination and Political Participation."
American Journal of Sociology 68 (September 1962):
205-13.
McLeod, Jack M . , and Becker, Lee B. "Testing the Validity
of Gratification Measures Through Political Effects
Analysis." In The Uses of Mass Communications:
Currect Perspectives on Gratifications Research,
p p . 137-64. Edited by Jay 0. Blumler and Elihu K a t z .
Beverly Hills, Calif.: Sage Publications, 1974.
McLeod, Jack; Ward, Scott; and Tancil, Karen. "Alienation
and Use of the Mass Media." Public Opinion Quarterly
29 (Winter 1965):584-94.
Nafziger, Ralph 0.; Engstron, Warren C.; and MacLean,
Malcolm S., Jr. "The Mass Media and An Informed
Public." Public Opinion Quarterly 15 (Spring 1951):
--------105-14.
Nettler, Gwynn. "A Measure of Alienation." American Socio­
logical Review 22 (December 1957):670-77.
Nie, N. H . ; Powell, G . B . ; and Prewitt, K. "Social Struc­
ture and Political Participation: Developmental
Relationships." American Political Science Review
63 (June and September 1969):361-78, 802-32.
O'Keefe, Garrett J. "Political Malaise and Reliance on
Media." Journalism Quarterly 57 (Spring 1980):
122-28.
O'Keefe, Garrett J . , and Mendelsohn, Harold. "Nonvoting:
The Media's Role." In Deviance and Mass M e d i a .
Edited by Charles Winick. Beverly Hills, Calif.:
Sage Publications, 1978.
O'Keefe, Garrett J.; Mendelsohn, Harold; and Lui, Jenny.

265
"Voter Decision Making 1972 and 1974." Public
Opinion Quarterly 40 (Fall 1976):320-30.
Olsen, Marvin E. "Alienation and Political Opinions."
Public Opinion Quarterly 29 (Summer 1965):200-12.
'Two Categories of Political Alienation." Social
Forces 47 (March 1969):288-99.
Paige, Jeffrey. "Political Orientation and Riot Partici­
pation." American Sociological Review 36 (October
1971) :810^27T
Pietila, Viekko. "Alienation and the Use of Mass Media."
ACTA Sociologicia 13, 14 (1970):237-52.
Poindexter, Paula M. "Non-News Viewers." Journal of
Communication 30 (Autumn 1980):58-65.
Pollock, Philip H. III. "Organizations and Alienation: The
Mediation Hypothesis Revisited." The Sociological
Quarterly 23 (Spring 1982):143-55.
'The Participatory Consequences of Internal and
External Political Efficacy." Western Political
Quarterly 36 (September 1983):400-409.
Pomper, Gerald M. "The Presidential Election." In The
Election of 1976: Reports and Interpretations,
pp. 54-82. fedited by Marlene M. Pomper.
New Y o r k :
David McKay Company, Inc., 1977.
Prisuta, Robert H. "Mass Media Exposure and Political
Behavior." Educational Broadcast Review 7
(June 1973):67-73.
Quarles, Rebecca Colwell. "Mass Media Use and Voting
Behavior: The Accurary of Political Perceptions
Among First Time and Experienced Voters." Commun­
ication Research 6 (October 1979):407-36.
Rarick, Galen R. "Political Persuasion: The Newspaper and
the Sexes." Journalism Quarterly 47 (Summer 1970):
360-64.
Robinson, Michael J. "American Political Legitimacy in an
Era of Electronic Journalism: Reflections on the
Evening News." In Television as a Social Force:
New Approaches to Television Criticism, pp. 97-139.
Edited by Douglass Cater and Richard A d l e r . New
York: Praeger Publishers, 1975.
Rosenberg, Morris. "The Meaning of Politics in Mass

266
Society." Public Opinion Quarterly 15 (Spring
1951) :5-151-------- ---------------_________ . "Some Determinants of Political Apathy." Public
Opinion Quarterly 18 (Winter 1954-55) :349-65"!
Ross, John E . , and Bostain, Lloyd R. "Communications
Activities of Wisconsin Farm Families in Winter­
time." Journal of Broadcasting 2 (Fall 1958):
319-28.-----------------------Samuelson, Merrill; Carter, Richard F.; and Rugg e l s , Lee.
"Education, Available Time and Use of Mass Media."
Journalism Quarterly 40 (Autumn 1963):491-96, 617.
Sargent, Leslie W . , and Stempel, Guido III. "Poverty,
Alienation and Mass Media Use." Journalism
Quarterly 45 (Summer 1968) :324-2TT!
Sears, David 0.; Lau, Richard C.; Tyler, Tom R . ; and Allen,
Harris M. Jr. "Self-Interest vs Symbolic Politics
in Policy Attitudes and Presidential Voting."
American Political Science Review 74 (September
T 9Wr67 0 - 8 ' 4 .---------------------Seeman, Melvin. "On the Meaning of Alienation." American
Sociological Review 24 (December 1959):783-91.
________ . "Alienation and Social Learning in a Reformatory."
American Journal of Sociology 69 (November 1963):
270-8T.-----------------________ . "Alienation, Membership and Political Knowledge:
A Comparative Study." Public Opinion Ouarterly
30 (Fall 196 6 ) :353-67.
________ . "Alienation and Engagement." In The Human Meaning
of Social C h a n g e , pp. 467-527. Edited by Angus
Campbell and Philip E. Converse.
New York: Russell
Sage Foundation, 1972.
Seeman, Melvin, and Evans, John. "Alienation and Learning
in a Hospital Setting." American Sociological
Review 27 (December 1962):772-78.
Star, Shirley A., and Hughes, Helen MacGill. "Report on an
Educational Campaign: The Cincinnati Plan for the
United Nations." American Journal of Sociology
55 (January 1950):389-401.
Stokes, Donald E. "Popular Evaluations of Government: An
Empirical Assessment." In Ethics and Business:
Scientific, Academic, Religious, Political and

267
Military, pp. 61-73. Edited by Harlan Cleveland
and Harold D. Lasswell. New York: Harper, 1962.
Stokes, Donald E., and Miller, Warren E. "Party Government
and the Saliency of Congress." Public Opinion
Quarterly 26 (Winter 1962):531-46.
Stone, Clarence E. "Local Referendums: An Alternative to the
Alienated Voter Model." Public Opinion Quarterly
29 (Summer 1965):213-22.
Tan, Alexis S. "Political Participation, Diffuse Support and
Perceptions of Political Efficacy as Predictors of
Mass Media Use." Communication Monographs 48
(June 1981) :133-45:----------- ----Templeton, Frederic. "Alienation and Political Participation:
Some Research Findings." Public Opinion Quarterly
30 (Summer 1966):249-61.
Thompson, Wayne, and Horton, John E. "Political Alienation
as a Force in Political Action." Social Forces
38 (March 1960):190-95.
Tichenor, Philip J . , and Wackman, Daniel R. "Mass Media
and Community Public Opinion." American Behavioral
Science 16 (March-April 1973):593-607.
Troldahl, Verling C. "Studies of Consumption of Mass Media
Content." Journalism Quarterly 42 (Autumn 1965):
596-606.
Tufte, Edward R. "Determinants of the Outcomes of Midterm
Congressional Elections." American Political
Science Review 69 (September 1975):812-26.
Useem, Bert. "Trust in Government and the Boston AntiBusing Movement." The Western Political Quarterly
35 (March 1982):81^5T:
VandenHaag, E. "Of Happiness and of Despair We Have No
Measure." In Mass Culture, pp. 504-36.
Edited
B. Rosenberg and D M . White. Glencoe, 111.:
The Free Press, 1957.

by

Vinyard, Dale, and Sigel, Roberta S. "Newspapers and Urban
Voters." Journalism Quarterly 48 (Autumn 1971):
486-93.
Wade, Serena, and Schramm, Wilbur. "The Mass Media as
Sources of Public A f f a i r s , Science and Health
Knowledge." Public Opinion Quarterly (Summer 1969):
197-209.

268
Warren, Donald I. "Mass Media and Racial Crisis: A Study of
the New Bethel Church Incident in Detroit." Journal
of Social Issues 28 (1972):111-32.
Westley, Bruce H . , and M o b i u s , Joseph P. "A Closer Look
at the Non-Television Household." Journal of
Broadcasting 4 (Spring 1960):164-77!
Westley, Bruce H . , and Severin, Werner J. "A Profile of
the Daily Newspaper Non-Reader." Journalism
Quarterly 41 (Spring 1964):45-50, 156.
Wilson, James Q . , and Banfield, Edward C. "PublicRegardingness as a Value Premise in Voting
Behavior." American Political Science Review 58
(December 1964):876-87.
Woelfel, Joseph. "Foundations of Cognitive Theory: A Multi­
dimensional Model of the Message-Attitude-Behavior
Relationship." In Message-Attitude-Behavior Rela­
tionship: Theory, Methodology and Application, ~pp.
89-ll6. Edited by Donald f*. Cushman and Robert D.
McPhee. New York: Academic Press, 1980.
Woelfel, Joseph, and Danes, Jeffery E. "Multidimensional
Scaling Models for Communication Research." In
Multivariate Techniques in Human Communication
Research, pp. 333-64. Edited by Peter R. Monge and
Joseph N. Cappella. New York: Academic Press, 1980.
Wolfinger, Raymond E.; Wolfinger, Barbara K . ; Prewitt,
Kenneth; and Rosenhack, Sheila. "America's Radical
Right: Politics and Ideology." In Political Opinion
and Electoral Behavior, pp. 189-226. Edited by
Edward C. Dreyer and Walter A. Rosenbaum. Belmont,
Calif.: Wadsworth Publishing Company, 1966.
Books
Alexander, Arthur J . , and Bass, Gail V. School Taxes and
Voter Behavior: An Analysis of School District
Property Tax Elections. Santa Monica, Cal i f . : The
Rand Corporation, April 1974.
Berelson, Bernard R . ; Lazarsfeld, Paul E.; and McPhee,
William N. Voting. Chicago: The University of
Chicago P r e s s , T754.
Blumler, Jay G . , and Katz, Elihu, e d s . The Uses of Mass
Communication: Currect Perspectives on Gratifica­
tion Research. Beverly Hills, Calif.: Sage
Publications, 1974.

269
Blumler, Jay G . , and McQuail, Denis. Television in Politics
Its Uses and Influences. Lon d o n : Faber and Faber,
l/TD.',' T953-;------------Burdick, Eugene, and Brodbeck, Arthur J . , e d s . American
Voting Behavior. Glencoe, 111.: The Free t r e s s , 1959.
Campbell, Angus, and Converse, Philip E . , eds. The Human
Meaning of Social Change. New York: Russell Sage
Foundation, 1972.
Campbell, Angus,; Converse, Philip E.; Miller, Warren E.;
and Stokes, Donald E. The American V o t e r . Chicago:
The University of Chicago Press, i960.
Campbell, Angus; Gurin, Gerald; and Miller, Warren E.
The Voter D e c i d e s . Evanston: Row, Peterson and
Company, 1954.
Carter, Richard F. Voters and Their Schools; A Technical
Report from the Institute for Communication R e st rch.
Stanford, C a l i f .: Stanford University P r e s s , 19Z ,.
Cater, Douglass, and Adler, Richard, eds. Television as a
Social Force: New Approaches to Television
Criticism. New York: Praeger Publishers, T9 7 5 .
Cleveland, Harlan, and Lasswell, Harold D . , eds. Ethics and
Business: Scientific, Academic, Religious, Political
and M ilitary. New York: Harper, 1962?
Cushman, Donald P., and McPhee, Robert D . , eds. MessageAttitude-Behavior Relationship: Theory, Methodology
and Application. New York: Academic Press, 1980.
Dreyer, Edward C., and Rosenbaum, Walter A., eds. Political
Opinion and Electoral Behavior. Belmont, Calif.:
Wadsworth Publishing Company, 1966.
Emerson, Thomas I. Toward a General Theory of the First
Amendment. foew York: Vintage Books, 1966.
Gamson, William E . Power and Discontent. Homewood, 111.:
Dorsey Press’^ 1968.
Hamilton, Howard D . , and Cohen, Sylvan H. Policy Making by
Plebiscite: School Referenda. Lexington, Has s .7
D. C. Heath and Company, 1974.
Kowalski, Joan P. Sullivan. Voter Behavior and Campaign
Strategies in School Finance Elections. Arlington,
Va.: Educational Research Service Inc., 1977.

270
Kraus, Sidney, and Davis, Dennis. The Effects of Mass
Communication on Political Behavior University
Park: The Pennsylvania State University Press,
1976.
Kurland, Philip B. Free Speech and Association: The Supreme
Court and the First Amendment. Chicago: The
University of Chicago Press, 1975.
Lang, Kurt, and Lang, Gladys Engel. Voting and Nonvoting:
Implications of Broadcasting Returns Before Polls
are Closed. Waltham, Mass.: Blaisdell Publishing
Company, 1968.
Lazarsfeld, Paul F.; Berelson, Bernard; and Gaudet, Hazel.
The People's Choice. New York: Columbia University
P r ess, 1944.
Levin, Murray B. The Alienated Voter: Politics in B o s t o n .
New York: Holt, Rinehart and Winston, 1960.
Mendelsohn, Harold, and O'Keefe, Garret J. The People
Choose a President: Influences on Voter Decision
M a k i n g . New York: Praeger Publishers, t n c ., 1976.
Milbrath, Lester W . , and Goel, M. L. Political Participation:
How and Why Do People Get Involved in Politics, 2nd
ed. Chicago: Rand McNally College Publishing
Company, 1977.
Miller, Warren E., and Levetin, Teresa E. Leadership and
Change: The New Politics and the American Elector­
ate . Cambridge, Mass.: Winthrop Publishers, I n c .,
T976.
Nie, Norman H . ; Hull, C. Hadlai; Jenkins, Jean G . ; Steinbrenner, Karin; and Bent, Dale H. Statistical
Package for the Social Sciences, 2nd ed. New York:
McGraw-Hill Book Company, 1975.
Niemi, Richard G . , and Weisberg, Herbert F., eds. Contro­
versies in American Voting Behavior. San Francisco:
W. H . Freeman Company, 19/6.
Monge, Peter R . , and Cappella, Joseph N., eds. Multivariate
Techniques in Human Communication Research. New
York: Academic Press, 1980.
Piele, Philip K . , and Hall, John Stuart. Budgets, Bonds and
Ballots: Voting Behavior in School Financial
Elections. Lexington, Mass.: D. C. Heath and
Company, 1973.

271
Pomper, Marlene, ed. The Election of 1976: Reports and
Interpretations. New York: David McKay Company,
In c . , 197 1 .
Reeder, Ward G. Campaigns for School T a x e s : A Manual for
Conducting Sucn Campaigns. New York: The MacMillan
Company, 1946.
Robinson, John P.; Rusk, Jerald G . ; and Head, Kendra B.
Measurement of Political Attitudes. Ann Arbor,
M i chigan: Institute for Social Research, The
University of Michigan, 1966.
Rosenberg, B . ; and White, D. M . , eds. Mass Culture.
Glencoe, 111.: The Free P r e s s , 1957.
Selltiz, Claire; Wrightsman, Lawrence S . ; and Cook,
Stuart W. Research Methods in Social Relations,
3rd ed. New York: Holt, Rinehart and Winston, 1976.
Sudman, S. Reducing the Cost of Surveys. Chicago: AldineAtherton, 1967.
Verway, David I. Michigan Statistical Abstracts. 13th ed.
East Lansing, Michigan: Michigan State University
Graduate School of Business, Division of Research,
1978.
Wimmer, Roger D . ; and Dominick, Joseph R. Mass Media
Research: An Introduction. Belmont*! Cal i f . :
Wadsworth Publishing Company, 1983.
Woelfel, Joseph, and Fink, Edward L. The Measurement of
Communication Processes: Galileo Theory and Method.
New York: Academic Press, 1980.
Government Documents
Federal Communications Commission. Amendment of Multiple
Ownership Rules (Newspapers^ Second Report and
Ord e r . 50 F .'C.C . 2d 1045'; "32 T T T r ^ d V ' T O T T U " "
Fed. Reg. 6449 (1975), on Reconsideration 53 F.C.C
2d 589, 33 R.R. 2d. 1603 (1975).
FCC vs. National Citizens Committee for Broadcasting, 436
U.S. 775, 98 S. Ct. 2096 (1978).
Michigan Department of Commerce, Office of Economic Expan­
sion. Economic Profile of Shiawassee Cou n t y .
Lansing, Michigan (September 1966).
National Citizens Committee for Broadcasting v. F.C.C.

272
55 F 2 d . 938 (D.C. Cit. 1977), 39 R.R. 2d. 1463
(1977) .
Temporary Special Levy Commission. Summary Report and
Research Repor t s , Volume 1. Olympia: Washingtc
State Legislature, 1971.
U.S. Constitution. Amend.

I.

U.S. Department of Commerce, Bureau of the Census. 1977
City County Data Book: A Statistical Abstract
Supplement (1977).
Unpublished Papers
Barnett, George A., and Hughes, Carol E. "The Effects of
Media Exposure on Uncertainty Reduction and Voting:
A Preliminary Analysis." Paper presented to the
International Communications Association,
Chicago, Illinois, 1978.
Becker, L.B.; McLeod, J. M . ; and Ziemke, D. "Correlates
of Media Gratifications." Cited in Kim A. Smith
and Warren Bechtolt, Jr.
"Adolescents' Use of
the Content of the 1976 Presidential Debates."
Paper presented to the International Communications
Association, Chicago, Illinois, April
1978.
Becker, Lee B . ; Sobowale, Idowu A . ; and Casey, William E.
Jr. "Newspaper and Television Dependencies:
Their Effects on Evaluation of Government Leaders."
Paper presented to the International Communications
Association, Chicago, Illinois, 1978.
Block, Martin, and Woelfel, Joseph. "Area Vocational Educa­
tion Campaign Strategies and Techniques Project,
Phase I: Communication Strategies." East Lansing,
Michigan: College of Communication Arts and
Sciences Michigan State University, September 1978.
Block, Martin, and Woelfel, Joseph. "Area Vocational Educa­
tion Campaign Strategies and Techniques Project,
Phase II: Campaign Development, Phase III: Campaign
Evaluation." East Lansing, Michigan: College of
Communication Arts and Sciences Michigan State
University, June 1979.
Cohen, Sylvan, H. "Voting Behavior in School Referenda: An
Investigation of Attitudes and Other Determinants
by Q Technique and Survey Research." Ph.D. disser­
tation, Kent State University, 1971.

273
Fleitas, Daniel William. "The Underdog Effect: An Experi­
mental Study of Voting Behavior in a Minimal
Information Election." Ph.D. dissertation,
Florida State University, 1970.
Krischak, Donald Eugene. "An Analysis of an Unsuccessful
Rural Community College Millage Election to Deter­
mine How Selected Communication Techniques and
Demography Interacted to Influence Voter Decision."
Ph.D. dissertation, Michigan State University, 1971.
McLeod, Jack M . ; Bybee, Carl R . ; Luetscher, William D.; and
Garramore, Gina. "Mass Communication and Voter
Volatility." Paper presented to the Association of
Educators in Journalism, Seattle, Washington, 1978.
Palmgreen, Philip. "Mass Communication and Political
Knowledge: The Effects of Political Level and
Mass Media Coverage on Political Learning." Ph.D.
dissertation, University of Michigan, 1975.
Rowe, Ronald Roger. "Selective Factors Influencing Voter
Response to a School Bond Issue." Ph.D. dissertation,
U.S. International University, 1971.
Semlac, William, and Williams, Wenmouth Jr. "Uses, Grat­
ifications and Avoidances: Voting Decisions in the
1976 Presidential Debates." Paper presented to the
International Communications Association, Chicago,
Illinois, 1978.
Sheinkopf, Kenneth G . ; and O'Keefe, M. Timothy. "The Eagleton Affair: A Study of Mass Diffusion." Cited in
Sidney Kraus and Dennis Davis, eds. The Effects of
Mass Communication on Political Behavior. University
Park: The Pennsylvania State University Press, 1976.
Sigel, Roberta S. "Election With an Issue: Voting Behavior
of a Metropolitan Community in a School Fiscal
Election." Cited in Howard D. Hamilton and Sylvan H.
Cohen. Policy Making by Plebiscite: School Referenda.
Lexington, Mass.: d T C. Heath and Company, 1974.
Smith, Kim A., and Bechtolt, Warren Jr. "Adolescents' Use
of the Content of the 1976 Presidential Debates."
Paper presented to the International Communications
Association, Chicago, Illinois, 1978.