INFORMATION TO USERS

This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted.

The following explanation of techniques is provided to help you understand markings or patterns 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) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity.

2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame.

3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete.

4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced.

5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received.

Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 76-24,617 HOWELL, Susan Robinson, 1948- AMERICAN POLITICAL PARTIES IN A PRESIDENTIAL CAMPAIGN: A STUDY OF THE 1972 LOCAL CAMPAIGN ACTIVISTS. The Ohio State University, Ph.D., 1976 Political Science, general

Xerox University Microfilms,Ann Arbor, Michigan 48106

Copyright by Susan Robinson Howell 1976

I AMERICAN POLITICAL PARTIES IN A PRESIDENTIAL CAMPAIGN:

A STUDY OF THE 1972 LOCAL CAMPAIGN ACTIVISTS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

Susan Robinson Howell, B.A., M.A.

*****

The Ohio State University

1976

Reading Committee: Approved By

Professor John H. Kessel Professor Herbert B. Asher Professor C. Richard Hofstetter ______/ Adviser Department of Political Science To Bill and my parents

ii ACKNOWLEDGEMENTS

X am indebted to my advisor, John H. Kessel, for his time spent in editing and reviewing the manuscript and for his continual guidance and encouragement throughout this project. I am also grateful to C. Richard Hofstetter for the use of his data from the Television and Electronic News Project and to the Polimetrics Laboratory at The Ohio State University for its services.

iii VITA

November 6, 1948..... Born - Cincinnati, Ohio

1969...... B.A., Anderson College Anderson, Indiana

1971-1974...... Research Associate, Polimetrics Laboratory, The Ohio State University Columbus, Ohio

1974-197 5 ...... Teaching Associate, Department of Political Science The Ohio State University Columbus, Ohio

1975-197 6 ...... Visiting Assistant Professor, Department of Political Science University of Cincinnati Cincinnati, Ohio

PUBLICATIONS

"Political Information: The Effects of System and Individual Character­ istics," Comparative Political Studies, Vol. 8, No. 4 (January, 1976), pp. 413-435.

FIELDS OF STUDY

American Government: Political Participation and Voting Behavior Public Opinion Congress Public Policy

Research Methods

Comparative Government: Western Europe Comparative Political Behavior TABLE OF CONTENTS

Page ACKNOWLEDGMENTS...... iii

VITA ...... iv

LIST OF TABLES ...... vii

LIST OF FIGURES...... x

Chapter One: An Introduction to Parties in a Presidential Campaign and the 1972 Campaign Activists . 1

Introduction...... 1 The Decentralized American Party System ...... 2 Candidate Organizations and the Regular Party .... 4 The 1972 County Leaders ...... 9 The Parties as Organizations...... 20 Study Overview...... 26 The Data...... 29

Chapter Two: Motivations, Consensus, and Obligations. . . 33

Amateurs and Professionals ...... 34 Intra-Party Issue Consensus ...... 44 - • Obligations to the Party...... 73 Conclusions ...... 82

Chapter Three: Communications and Candidate Images. .. . 87

Introduction...... 87 Communications Channels ...... 88 Communications Channels and Other Variables ...... 98 Candidate Images...... 108 Candidate Images and Other Variables...... 118 Conclusions ...... 128

v Page Chapter Four: The Issue Agendas: A Summary Measure of Campaign Unity...... 131

Introduction ...... 131 The National Issue Agenda...... 132 The Local Issue Agendas...... 136 Constituency Influence on the Local Agenda ...... 139 Measuring Deviation from the National Issue Agenda . . 142 Organizational Influences...... 149 Conclusion ...... 153

Chapter Five: Discussion and Conclusions ...... 158

Introduction ...... 158 Dimensions of Decentralization ...... 158 The Psychological Unity of American Parties...... 162 Comments on Future Research...... 169

LIST OF REFERENCES...... 173

vi LIST OF TABLES

Page Table 1.1: Educational Attainment of County Party Leaders: 1961 and 1972 (In Percentages)...... 12

Table 1.2: Variation in the Income of County Party Leaders, 1961 and 1972 (In Percentages)...... 13

Table 1.3: Variation in the Occupational Backgrounds of County Party Chairmen, 1958, 1961 and 1972 (In Percentages)...... 14

Table 1.4: Political Experience of County Leaders, 1961 and 1972...... 16

Table 1.5: Previous Party and Public Offices Held (In N u m b e r s ) ...... 17

Table 1.6: Political Offices Held by 1972 County Leaders During the Campaign (In Numbers)...... 18

Table 1.7: Political Aspirations of County Leaders...... 19

Table 2.1: Motivations for Involvement in the Campaign. . . 37

Table 2.2: Amateurs and Professionals Among the 1972 County Campaign Leaders ...... 40

Table 2.3: Amateurs' and Professionals' Interest in I s s u e s ...... 42

Table 2.4: Amateurs', Semi-Professionals', and Professionals' Interest in Issues ...... 43

Table 2.5: 1972 County Campaign Leaders' Perceptions of the Candidates' Issue Positions...... 46-50

Table 2.6: Perceptions of the Candidates: A Measure of Intra-Party Consensus...... 53

vii Page

Table 2.7: The Issue Positions of the 1972 County Campaign Leaders...... 56-62

Table 2.8: Intra-Party Issue Consensus...... 66-67

Table 2.9: Median Distances of Activists from Candidate on Issues ...... 72

Table 2.10: The Individual Index of Agreement ...... 74

Table 2.11: Obligation to the Party Felt by the 1972 County Campaign Leaders ...... 77-78

Table 2.12: Correlation Matrix for Obligation Items (Yule's Q ) ...... 80

Table 2.13: Guttman Scale of Obligations to the Party Felt by the 1972 County Campaign Leaders...... 81

Table 3.1: Local Party Leaders' Sources of Information About the National Campaign ...... 91

Table 3.2: Bivariate Relationships Between all Sources of Information...... 92

Table 3.3: Most Important Source of Information About the National Campaign for 1972 County Campaign Leaders ...... 94

Table 3.4: Background Characteristics of the County . Campaign Leaders and Communications Channels...... 96-97

Table 3.5: Obligation to Party and Communications Channels...... 100-101

Table 3.6: Agreement with Candidate and Use of Communications Channels ...... 102

Table 3.7: Amateurism and the Use of Communications Channels...... 103

Table 3.8: Being Within the Issue Consensus and the Use of Communications Channels...... 104

Table 3.9: Communications Channels and Perceptions of the Candidates' Issue Stands ...... 106-107

viii Page Table 3.10: Republican Assessment of Candidate Images: Average Number of Positive and Negative Comments About Candidates by Topic...... 110

Table 3.11: Democratic Assessment of Candidate Images: Average Number of Positive and Negative Comments About Candidates by Topic...... Ill

Table 3.12: The General Public's Image of the C a n d i d a t e s ...... 112

Table 3.13: Candidate Images ...... 119

Table 3.14: Issue Agreement and Candidate Images ...... 120-121

Table 3.15: Being Within the Party's Issue Consensus and Having a Positive Image of the Candidate ...... 122

Table 3.16: Amateurism and Candidate Images...... 124

Table 3.17: Obligation to the Party and Candidate Images . 126

Table 3.18: Communications Channels and Candidate Image. . 127

Table 4.1: National Issue Agendas: 1972 ...... 134

Table 4.2: 1972 Local Issue Agendas: Issues in Order of Importance by P a r t y ...... 138

Table 4.3: Regional Influences on the 1972 Local . Issue Agenda ...... 140

Table 4.4: Urban-Rural Influences on the 1972 Local Issue Agenda ...... 141

Table 4.5: Relationships Between Organizational Influences and Distance from the National Agenda in the 1972 Campaign Organizations ...... 151

ix LIST OF FIGURES

Page Figure 2.1: Number of Issues on Which 1972 Republican County Campaign Leaders are Outside Their Own Party Issue Consensus...... 69

Figure 2.2: Number of Issues on Which 1972 Democratic County Campaign Leaders are Outside Their Own Party Issue Consensus...... 70

Figure 4.1: Summary Measure of Distance from the National Issue Agenda: Republicans...... 147

Figure 4.2: Summary Measure of Distance from the National Issue Agenda: Democrats...... 148

x CHAPTER ONE AN INTRODUCTION TO PARTIES IN A PRESIDENTIAL CAMPAIGN AND THE 1972 CAMPAIGN ACTIVISTS

Introduction

American political parties are usually described as loose federa­

tions of state and local party organizations. The neat hierarchical

structure that appears on paper from the National Committees to the pre­

cinct leaders masks the considerable autonomy that state and local party

leaders have over recruitment, nominations, and campaigns in their juris­

dictions. In addition to local autonomy there is a variety across states

in the degree of organization. Yet every four years Republicans and Demo­

crats are engaged in a national effort to elect a presidential candidate,

requiring at least some degree of cohesion and coordination.

During the campaign committees supporting particular candidates and

the formal parties must work together. These committees have a variety

of relationships with the regular machinery from state to state. Some­

times the personal organization replaces the regular party; sometimes they

perform parallel functions. Usually the organizations are indistin­ guishable because regular party workers fill so many key positions in the 1 new organizations. We are well aware of the authority problems in the inter-election period. We might expect the same sorts of organizational problems in the parties during campaigns since they only exist on a nationwide basis for a few months. 2

This dissertation is a study of the campaign organizations in the

1972 campaign. It deals with the sources of decentralization and

problems of organization and authority well known to students of parties.

The focus will be on the county leaders of both the Committee to Re-elect

the President and the Citizens for McGovern-Shriver. It examines their

characteristics, motivations, and attitudes toward the campaign and the

national party elites and attempts to relate these variables to their

behavior in relation to the national party. The dissertation asks what

factors incline local leaders to follow the national party and what

inclines them to diverge.

The Decentralized American Party System

It is no news that American political parties represent a radical departure from the responsible party model. There are numerous reasons why these parties will probably never resemble the cohesive, programma­

tic parties of some European countries. Frank Sorauf suggests geographic, cultural, and institutional factors all mitigate against strong national parties. Rural areas are not easily mobilized by door to door campaigns and voter contact. The existence of a long-term minority party in some areas reduces the ability of that party to attract able candidates and workers, diminishing its effectiveness in offering an alternative program.

The direct primary seriously inhibits the local organization's control over nominations. The more offices for which a primary is required, the less the party organization will be regarded as the route to public office. Finally, in some state or localities, patronage is simply not an acceptable way of rewarding party service, reducing the opportunity 2 for parties to offer direct tangible rewards for participation.

Other parties scholars have pointed to a variety of factors that

also contribute to decentralization of authority. Control from the top

is mostly formal, and lower levels of party hierarchy frequently control

the level above by selection of convention delegates. An example of

this viewpoint is David Truman's essay on federalism and the American party system. Because the leadership of the presidential party does not control the structure which nominates legislative candidates, we can only expect localized parties.^ Austin Ranney, too, attributed much of the lack of cohesive parties to the decentralized nominating procedure."*

Finally, Arthur Holcombe's study of American parties supports the charac­ terization of parties as the instruments for the coordination of factions.

One of their functions is to make possible mutual concessions and adjust conflicts among factions. He concluded that the most important fact concerning the alignment of the major parties is the relations of the fL factions within the party.

Perhaps the most obvious indicator of the lack of authority at the national level is the position of the national committees in the party power structure. These Committees meet only two or three times a year and delegate their work to the national chairman and his staff. ^ Much of the role of the chairman and his staff depends upon his personal

Q relationships with party leaders. Since this study is concerned with the nature of parties during a national campaign, the activisties of the national chairman at this time deserve note. The national chairman's role in the campaign depends largely on his relationship with the 4

candidate. The candidate may choose to leave the old National Committee

staff intact and rely on his own separate and personal organization, the

"divided" system. Or he may appoint his own team to the National Commit­

tee staff including his pre-convention manager, the "unified" system. ^

If the pre-convention manager is not made National Chairman he will still normally run the campaign and decide on a role for the Chairman

to play. Clearly, the National Committee and its staff speak with

little authority during the inter-election period, and at campaign time may well be relegated to a subordinate role.^

Candidate Organizations and the Regular Party

Organizations supporting particular presidential aspirants arise during the contest for the party's nomination. By tradition, the na­ tional committees are expected to remain neutral. They must make arrange- arrangements for the national conventions where the candidates for pre­ sident and vice president will be chosen. In the meanwhile, someone must seek out potential delegates at district and state conventions, raise money, hand out press releases, arrange transportation to get the aspirant to the states where primaries are being held, and so on. Con­ sequently, the successful candidate has an organization in being by the time he gives his acceptance speech. There are leaders who are accustomed to working directly with the candidate, and scores of emplo­ yees who attend to the necessary logistics. Neither leaders or workers are ready to step aside at the moment of victory. The existence of two separate organizations poses a real organiza­

tional problem. The members of the candidate's personal organization

feel they are responsible for his success, and therefore qualified to

conduct the fall campaign. The members of the formal party organization

feel they are the professionals who have been getting ready for years,

and are therefore qualified to conduct the fall campaign. ^ A history

of campaign organizations illustrates the variety of combinations to

which this dilemma has led.

In 1960 Kennedy's pre-convention campaign manager was his brother,

Robert. The basic idea of JFK's team was to run a unified campaign

effort, so the national committee was reorganized to serve the campaign.

Although Henry Jackson was chairman, Robert Kennedy was clearly the boss

of this unified system. It was inevitable that Robert Kennedy assumed

leadership because the^Kennedys brought their campaign organization with

them, and Jackson had only a small staff. Kennedy supervised the work

of the national committee through "campaign coordinators" from the

Kennedy staff who directed the divisions of the committee. Some units

of the national committee were abolished and their workers were assigned to other campaign tasks. To mobilize state and local support

for the national ticket, state coordinators were selected from those who supported Kennedy before the convention. These coordinators dealt with state central committees and county central committees. The

Citizens for Kennedy-Johnson, acting as an auxiliary organization, set up a parallel organization in all states except the South, which reached down to the county level 6

A different pattern was used by Nixon in 1960. He decided to lean

on his own campaign organization and to use the national committee in a

subordinate role. In some areas the Nixon staff completely controlled

the campaign effort. In other areas the Nixon people held the leader­

ship positions while the national committee division remained intact.

Organizations of the National Volunteers for Nixon-Lodge, as the commit­

tee was called, were set up in most states and were directed and par­

tially supplied by Nixon headquarters in Washington. These organizations

were directed not to compete with the regular party. Their relations 14 with the party varied from state to state.

In 1964 the Republicans achieved a more complete fusion of the

regular party and the citizen committees at the national level than

they had in 1960. Goldwater appointed his men to key positions at the

national committee and in the less important Citizens for Goldwater-

Miller Committee. Although both committees established regional offices,

close contact with each other prevented some of the usual waste and

duplication.^ At the center of decision-making in the campaign was a group composed of Goldwater intimates and conservative intellectuals.

Cooperation at the local level between the citizens' committee and the regular party varied. In some states they were the same organization and in others they had little contact and functioned separately.

In 1964 Johnson had at his disposal an almost ready-made unified campaign organization. The national committee was already staffed with

Kennedy men who reported through key Johnson appointees in the White

House. A system of state and regional coordinators was established, 7

with regular state party chairmen often being the state coordinators.

So LBJ did not have to encounter an established party unwilling to submit

to his direction of the campaign.

Hubert Humphrey in 1968 faced a party plagued by factions in the

states, factions due to Vietnam, urban unrest, and the Chicago violence.

Partly because there was no pre-convention plan for a campaign, Humphrey's

organization was slow in getting started. Lawrence O'Brien, Joseph

Napolitan and Ira Kapenstein tried to pull together the parts of the

organization existing in Washington, and eventually state and regional

chairmen were set up. The campaign began with Humphrey's personal

loyalists and then the regular party fell in as the campaign was a

success. In the end Humphrey was relying quite heavily on the regular

party people at both local and national levels.^

In contrast, the top of Nixon's campaign was quite divorced from

the party. In fact, his headquarters were in New York, and the regulars

from Washington were invited only once a week to the daily staff meetings. The New York headquarters was run exclusively by Nixon people, but the organization at local levels consisted of regulars.

This study deals with the 1972 campaign organizations. That year neither campaign was characterized by close cooperation with the existing party machinery. In Nixon's case it was by his own choosing. There was no grand plan to the Nixon campaign, organizationally. Theodore White in The Making of the President 1972 describes the Nixon organization as made up of "teams," each headed by a man who was left on his own to accomplish the job. The most important team was the Committee to 8

Re-elect the President whose function it was to mobilize the voters.

The Committee in Washington was to be entirely divorced from the

17 Republican National Committee.

The fissure between McGovern's personal organization and the party

regulars began at the top of the organization. O'Brien, a party regu­

lar, felt the campaign needed one manager and a strict hierarchical

organization. He became the liaison with the leadership of the party.

Gary Hart, who has been with McGovern through the primaries, disagreed

with O'Brien and favored leaving initiative to the volunteers who had

fought the primary battles. Hart's people were the new people who felt

the campaign would be won with their own activists plus volunteers

across the country. Old-timers referred to this as the politics of

exclusion; some state Democratic chairmen and pillars of the Democratic

18 Party were virtually ignored. Gary Hart's version of this relationship was understandably different. He ordered cooperation at state and local

levels with the Democratic Party organization, and perceived the atti­

tude of the McGovern campaign as one of intolerance for the disregard of fellow Democrats it is probable that locales had varying degrees of friction between these two groups. Like most candidate organizations,

the Citizens for McGovern-Shriver was organized down to the local level.

Similar to the Nixon committee, there were regional coordinators, and officers below them responsible for five or six states each. In every state a state coordinator established a campaign committee, set up head­ quarters, distributed literature, and raised money. 9

It is clear from this brief history that the campaign organization

can be various combinations of regular party people and the candidate's

personal associates; and how strained relations are between these two

groups will depend on variables in each campaign such as the attractive­ ness of the nominee, his attitude toward the established party, the personnel of the national committees, the anticipated margin of victory or defeat, and the competetiveness of the primaries.

The 1972 County Leaders

Regardless of who dominates the campaign at the elite level, at the local level the regular party and the candidate organization tend to converge. A campaign organization structured down to the local level will fill the positions with regular party people who are willing simply because it is easier to work within an already existing party structure than to recruit new people. At times, of course, it is necessary to go outside the regular party to adequately man the campaign?®

In order to get a clearer picture of who county leaders were in

1972 we will examine their career experiences and status characteristics and compare these results to results of some previous studies of local activists. Several factors should be kept in mind while making these comparisons. First, the previous studies were in specific states or localities while the 1972 data are from a national sample. Second, the earlier studies examined regular party county chairman, while the pre­ sent research sampled county leaders in a presidential campaign. And, there is an eleven-year time lapse between the older studies and the present one which may well compound increases in status variables such 10

as education and income.

One characteristic always attributed to those active in politics

is high status. Tables 1.1, 1.2, and 1.3 show this to be true in 1972 as well as in previous years. The 1972 county leaders are a highly educated group with two-thirds having a college degree. They are also more educated than the 1961 chairmen, although this difference parallels

the general increase in everyone's education. In other words, the differ­ ence between the education of the population and that of the activists is the same in 1972 as it was in 1961.

Comparing the parties, the 1972 Democrats had attained a higher educational level than Republicans, a fact that will be explained more fully by occupations. Comparing the earlier parties is misleading because, as Patterson points out, one of the flaws of state-wide data is that the status characteristics of the dominant party workers are higher than they would be in a competitive situation. Both Kansas and

Oklahoma had unbalanced party situations encouraging the talented and ambitious activists to become identified with the dominant party.

Turning to income, again all three groups of chairmen are privi­ leged. The 1972 leaders earned a larger income than the 1961 chairmen, but this, too, could be due to inflation and the increase in earnings of the whole population. In contrast to education, the Republicans earned more than the Democrats in 1972, another finding which becomes clearer when we examine occupations. 11

The 1972 county leaders were asked what they considered their main

occupation to be. When these results are compared to the 1961 occupa­

tions several substantial differences appear. First is the decrease

in the percentage of lawyers, a career traditionally associated with politics. In the earlier studies lawyers comprised about one-fourth of the dominant parties, but eleven years later lawyers totaled only one-tenth. As the number of lawyers has declined, housewives and pro­ fessionals other than lawyers have made substantial gains. At most, housewives comprised 6.8% in the early 1960's, while in 1972 18.8% of Republican respondents were housewives, and the increase in other professionals was from a 1961 high of 18.3% to 44.5% in 1972.

The third difference is the decline in businessmen in the Demo­ cratic ranks. What the Democrats lack in businessmen they make up for in a large number of other professionals and persons not in the labor force. Teachers and social workers expand the proportion of Democrats categorized as other professional, and those not in the labor force are primarily students. This trend is important to the new amateur politics based on policy and purposive incentives. It is the housewives, social workers, teachers, or students who are likely to be motivated by other than material or social gratifications. So although the parties were both made up of those with high status occupations, the type of occupa­ tion differed. Public officials and'insurance and real estate salesmen were more common among the Republicans, and students and teachers among

Democrats. This illuminates the previous findings that the Democrats had lower incomes and higher education. The Republican higher paying TABLE 1.1: EDUCATIONAL ATTAINMENT OF COUNTY PARTY LEADERS: 1961 AND 1972 (In Percentages)

Oklahoma 1961 Kansas 1961 National Sample 1972 Democratic Republican Democratic Republican Democratic Republican

Education

Grade School 1.7 1.7 12.3 7.2 0.0 1.0 Some High School 6.8 10.0 5.5 2.4 0.5 0.0 High School Diploma 10.2 25.0 24.7 7.2 8.1 15.1 Some College 28.8 20.0 26.0 28.9 22.7 21.4 College Degree 15.3 23.3 13.7 18.1 37.8 34.4 Graduate Degree 35.6 20.0 13.7 32.5 30.8 28.1 No Response 1.7 0.0 4.1 3.6

Total 100.1 100.0 100.0 99.9 100.0 100.0

N's 59 60 73 83 185 192

Source of Oklahoma and Kansas Data: Samuel C. Patterson, "Characteristics of Party Leaders," Western Political Quarterly, Vol. 16, No. 2 (June 1963), pp. 332-352. TABLE 1.2: VARIATION IN THE INCOME OF COUNTY PARTY LEADERS, 1961 AND 1972 (In Percentages)

Kansas 1961 Oklahoma 1961 National Sample 1972 Democratic Republican Democratic Republican Democratic Republican Income Range

Less than $3,000 19.2 6.0 ------Less than $4,000 -- -- 1.7 13.3 8.0 2.6 $3,000 to $5,000 17.8 9.6 ------$4,000 to $6,000 ------11.9 13.3 6.5 1.0 $5,000 to $8,000 20.5 19.3 ------—— $6,000 to $8,000 ------8.5 16.7 4.9 1.6 $8,000 to $10,000 17.8 13.3 16.9 16.7 8.6 2.6 More than $10,000 19.2 42.2 57.6 38.3 68.6 85.9 No response 5.5 9.6 3.4 1.7 3.2 6.3

Total 100.0 100.0 100.0 100.0 100.0 100.0

N's 73 83 59 60 185 192

Source of Oklahoma and Kansas data: Samuel C. Patterson, "Characteristics of Party Leaders," Western Political Quarterly, Vol. 16, No. 2 (June 1963), pp. 332-352. TABLE 1.3: VARIATION IN THE OCCUPATIONAL BACKGROUNDS OF COUNTY PARTY CHAIRMEN, 1958, 1961 AND 1972 (In Percentages)

Oklahoma 1961 Kansas 1961 Wisconsin 1958 National Sample 1972 Democratic Republican Democratic Republican Democratic Republican Democratic Republican Occupation ------

Farmer-Rancher 18.6 25.0 28.8 14.5 15.6 3.1 1.1 (2) 2.6 (5) Businessman 22.0 20.0 19 2 32.5 17.2 45.3 7.6 (14) 27.6 (53) Sales-Clerical 10.2 13.3 15.1 6.0 7.8 4.7 2.1 (4) 3.6 (7) Attorney 23.7 11.7 13.7 27.7 18.8 29.7 8.6 (16) 12.5 (24) Other Professional 15.3 18.3 8.2 9.6 6.3 12.5 44.5 (82) 24.0 (46) Laborer 1.7 5.0 1.4 3.6 23.4 0.0 2.1 (4) 1.5 (3) Public Official 1.7 0.0 1.4 1.2 1.6 0.0 0.5 (1) 3.6 (7) Housewife 1.7 0.0 6.8 1.2 4.7 0.0 15.7 (29) 18.8 (36) Retired 3.4 3.3 5.5 0.0 3.1 3.1 Other non-■labor force Other 1.7 1.7 0.0 0.0 0.0 0.0 15.7 (29) 5.7 (11) No response 0.0 1.7 0.0 3.6 1.6 1.6 Missing Data 2.1 (4) 0.0 (0)

Total 99.9 100.0 100.1 99.9 100.0 100.0 100.0 100.0

N ’s 59 60 73 83 64 64 185 192

Source of Oklahoma, Kansas and Wisconsin Data: Samuel C. Patterson, "Characteristics of Party Leaders," Western Political Quarterly, Vol. 16, No. 2 (June 1963), pp. 332-352. ’ £ 15 business and public jobs do not necessarily require a high education, while teachers, social workers and students have at least college

training with middle incomes.

The career patterns and aspirations of the Oklahoma and national samples are summarized in Tables 1.4 - 1.7. The Oklahoma data suggest that previous position in the party or public office reduces the possi­ bility of becoming a county chairman. More than half of the county leaders in Oklahoma have never served in a public or private office.

The 1972 county leaders are considerably more experienced. Approximately

70% had held a public or party office previously and 19% had experience in both. The party ranks were by far the greatest source of leaders in 1972, illustrating the convergence of the regular party and the committees at the local levels. As we would expect, the Republicans were recruited from the regular party more frequently than the Demo­ crats, and more of the latter party's chairmen were novices. Even so, the overlap is significant in view of the dissension in the McGovern campaign. Only 40% (74) of the Democrats and 2470 (47) of the Republi­ cans had never held a party office. One-half (93) of the Democrats and

3770 of the Republicans said that they were not currently holding a party office.

A tabulation of previous party and public offices held by Oklahoma county leaders, presented in Table 1.5 shows that those who have held party office held it at the precinct level. In 1972, however, both precinct and county officers were readily available for chairmanships. TABLE 1.4: POLITICAL EXPERIENCE OF COUNTY LEADERS, 1961 AND 1972

I

Oklahoma 1961 National Sample 1972 Democratic______Republican Democratic______Republican N %N % N %N % Type of Political Experience

Held other party office 8 13.6 13 21.7 89 48.1 95 49.5 Held public office 23 38.9 14 23.3 3 1.6 6 3.1 Held both other party office and public office 3 5.1 3 5.0 22 11.9 50 26.0 Never held other party office or public office 25 42.4 30 50.0 71 38.4 41 21.4

Total 59 100.0 60 100.0 185 100.0 192 100.0

Source of Oklahoma Data: Samuel C. Patterson, "Characteristics of Party Leaders," Western Political Quarterly, Vol. 16, No. 2 (June 1963), pp. 332-352. TABLE 1.5: PREVIOUS PARTY AND PUBLIC OFFICES HELD (In Numbers)

______Oklahoma 1961______National Sample of Countv Leaders: 1972 Office Democratic Leaders* Republican Leaders* Dem. * Rep.: (59) (60) (185) (192) Party Office Party Office Precinct 24 16 Precinct 82 102 County 3 9 County 53 99 District 5 5 City 31 59 State 1 9 State 29 56 YDems/YGOP/ Club Member 111 122 Women's clubs 7 6 Total 306 438 Total 40 45

Never held Elective Public Office Party Office 74 47 Municipal 16 10 Elective Office County 15 6 City/County 17 43 State 3 1 Other Public National 0 0 Office 12 23

Total 34 17 Total 29 66

Never held Public Office 160 136 Never held Public or Party Office 42 28

*Each party leader was counted for each party or public office he held which means that the total N's for this table are higher than the total sample size.

Source of Oklahoma Data: Samuel C. Patterson, "Characteristics of Party Leaders," Western Political Quarterly, Vol. 16, No. 2 (June 1963), pp. 332-352. TABLE 1.6: POLITICAL OFFICES HELD BY 1972 COUNTY LEADERS DURING THE CAMPAIGN (In Numbers)

Office Democrats * Republicans (185) (192) Party Precinct/Ward 43 46 City or Town 15 27 County 40 84 State 18 39 Political Club 65 84 No Party Office Held Now 93 71

Public City/County 5 23 Other Elected 13 18 None 168 153

No Office 43 1

Each party leader was counted for each party or public office he held which means that the total N's for this table are higher than the total sample size. TABLE 1.7: POLITICAL ASPIRATIONS OF COUNTY LEADERS

(1961) (1958)

Political Aspiration Democrats Republicans Democrats Republicans

Desire to run for elective public office 23.7% 40.0% 48.4% 23.4%

National Sample (1972) Democrats Republicans

Desire to run for public office 36.8% 36.5% Desire Party Office 38.4% 30.2%

Source of Oklahoma and Wisconsin Data: Samuel C. Patterson, "Characteristics of Party Leaders," Western Political Quarterly, Vol. 16, No. 2 (June 1963), pp. 332-: 20

Second, among the politically experienced, party office was held much

more frequently in 1972 than in 1961. And third, the 1961 county

leaders who had held public office did so at the municipal and county

levels, while in 1972 other public officials were also recruited. To

sum up the differences in careers, it seems that the 1972 local leaders were not only more politically experienced, but that experience was

partisan in nature.

The data in Tables 1.5 and 1.6 illustrate the considerable over­

lap between the formal party and the candidate committees. 69% of the

Democrats and 76% of the Republicans said they had held party office

prior to the 1972 campaign. And during the '72 campaign one-half of

the Democrats and 63% of the Republicans said they were holding party

offices. Although this shows more overlap in the Republican Party,

the data probably underestimate the amount of prior party experience since one can be involved and not hold a formal party office.

Political aspirations in 1972 were quite similar to 1961; in neither groups is political aspiration especially high. Another point to note about Table 1.7 is the lack of aspiration for other party offices.

Apparently the purely partisan career holds no more appeal for those already involved in party work than a public career.

The Parties as Organizations

To suggest approaches .to the study of these activists as a unit we might look to organization theory. But there has been very little previous work applying organization theory to political parties. 21

Lee F. Anderson undertook a search of organization theory for models

and criteria which would be of use in studying state and local parties.

He seems to be particularly attracted to the "natural systems" model

of organizational analysis in which the realization of goals is only

one of several needs to which the organization is oriented. In addition,

an organization strives to survive and maintain its equilibrium, even

after goals have been obtained. Anderson proposes to conceive of state

and local parties as natural systems and to study them on dimensions of

(1) autonomy -- the degree of dependence on external actors, (2) con­

trol — party organizations differ in control over members and (3)consen­

sus -- what portion of an organization is in mutual agreement, what is

the consensus between higher and lower participants. 22 a .11 three of

these dimensions will be explored in the coming chapters.

Joseph A. Schlesinger in an extensive application of organization

theory to parties discusses four aspects of party organization: (1) the prime place of office-seeking; (2) the effects of the political system;

(3) the role of what he calls a "nucleus," the party workers and organi­ zation gathered around one elective office; and (4) how various nucleii

interact and o v e r l a p ? ^ This study is not concerned with office-seekers so we will focus on the other incentives considered by Schlesinger.

Along the lines of the work on amateurs and professionals, he discusses Wilson and Clarke’s material, solidary, and purposive incen­ tives saying that the dominance of one or the other will have conse­ quences for the organization. 24 Material incentives produce flexibility 22

in defining goals, the solidary (social) incentives reduce internal

flexibility and purposive (policy) incentives lead to rigidity in 25 defining goals. One of the first questions in this study is to

determine the particular motives of the 1972 county leaders. In doing

so we should note that there is evidence that the amateur and profes­

sional incentives are less important than certain social traits in

determining amateur or professional behavior. C. Richard Hofstetter,

in a study of the determinants of joining amateur versus professional

organizations found that incentives were not very significant when

controls were applied for social integration, age, and economic status.

The socially integrated, older, and those of higher status had more

professional role perceptions and joined the professional organiza- 26 tions.

In assessing the effects of the political systems on the parties

Schlesinger says three conditions; the many independently elected offices at different levels, primaries, and no formal prerequisites for office, have led to an open system of political opportunity. And that system combined with two parties creates the interesting notion of the "illusion of organization." This means that with hardly any organized effort a candidate can be nominated and campaign.^ Although this could not occur in the Presidential race it does point to the question of how cohesive and well managed the organizations are. Tests of local party organization based on structural criteria show weak results. What, then does hold national party organizations together in spite of these structural weaknesses? 23

Also applicable to the Presidential contest is Schlesinger1s hypo­

thesis about the interdependence of association leaders (our county

leaders) and the office seekers (the presidential candidates). In

highly competitive races, such as the Presidency, the organization finds

its leadership crucial, and the leadership is more dependent on the or­

ganization than in non-competitive races. Interdependence is also pre­

sent in multi-nuclear relations. The smaller nucleus (county leaders)

can offer close contact with the voters, while the larger nucleus can

provide the attractive leaders, the issues, and possibly coattails?®

In this study we have only one nucleus, using Schlesinger's defini­

tion, and in the competitive presidential contest there should be evi­

dence of a dependence on attractive candidates. The multi-nuclear

hypothesis also applies. Some 32 % of the county leaders of the candi­

date committees were also regular county chairmen so they would be con­

cerned with how the larger nucleus affected local offices. If

Schlesinger's hypotheses are true it is important that local leaders

perceive positive attractive leadership at the top in order to want to make their own contribution to the campaign. To quote Schlesinger,

"local organizations care little about continuity and substance in the national party, but much about its ability to present appealing tickets.^

To what extent did the local leaders in 1972 perceive an attractive

ticket? Their perceptions, according to Schlesinger would affect their performance. 24

Another and newer body of theory applicable to the problems of

organization is implementation theory. Policy scientists writing on

implementation have listed numerous problems in coordinating a organ­

ization to achieve its stated objective. Since this work was built on

studies of bureaus and agencies, some of these problems are more rele­

vant for parties than others. One implementation problem relevant to

both agencies and parties is the problem of non-compliance by subor­

dinates. In his study of administrative feedback Kaufman suggests three

reasons for non-compliance. The subordinate has doubts about what he is

supposed to do, he is incapable of doing what he is supposed to do, 30 or he refuses to do what he is directed to do. Later in his study

Kaufman found considerable communication upward by subordinates instead

of the non-compliance he had expected. Therefore, this research is using part of Kaufman's theoretical framework, and not necessarily his

findings.

All of Kaufman's reasons for non-compliance are problems in party organizations. In large organizations there are conflicting and confusing directives from above, and it is left to the subordinates to interpret their situation and decide what is required. Considering the number of headquarters above the local political leaders, ambigui­ ties in directives seem inevitable. The local leader receives mail, and phone calls from state headquarters, regional headquarters and national headquarters. In addition he is watching his candidate on television and reading about him in the newspapers. The effects of 25

these sources of information will be explored more fully in Chapter

Three, but suffice it to say that the number and the autonomy of the

layers of authority above him leaves the local leader somewhat free to

fashion his own policies. On the second point, that subordinates are

incapable of doing what they are supposed to do, constituency pressure may make it impossible to go along with the national party leadership.

Lack of resources such as manpower and money, may make it difficult to comply with the leadership. Consider Kaufman's third point, the re­ fusal to comply. Local party leaders have frequently clashed with na­ tional party leaders, particularly in a presidential campaign. The pro­ blem of refusal is probably exacerbated by the nomination of a candi­ date representing a minority of his party such as McGovern.

The ability of an organization to elicit obedience depends on rewards and punishments, the subordinate's feeling that he ought to obey, his identification with the organization, and upon his confidence that the superior knows morep-*- All of these have been problems of party organizations. The research already reviewed documents some of these problems in the regular party, and it seems likely they will exist in a candidate organization as well. A last problem of implementation cited by policy scholars and applicable to political parties is the cooptation by a local constituency. Studies of federally imposed civil rights regulations and of public work projects have demonstrated how local interests can determine if and how a federal regulation is admin- istered. Party regulars are known to be quite sensitive to their con­ stituents, and this has been a hurdle to coordinating the national party. 26

Study Overview

In describing the 1972 party organizations, the focus will be on what factors contribute to local leaders' following the national party.

Drawing upon the theories and hypotheses above, certain characteristics and attitudes of local leaders were chosen because they would presumably affect behavior in relation to the national party.

The first factor chosen is the local leader's motivation to parti­ cipate in this campaign. Most have a variety of motives but we are in­ terested in the principal incentive. Although the literature on acti­ vists has perhaps overemphasized the importance of amateur and pro­ fessional incentives, still we repeatedly find that consequences are hypothesized for groups dominated by one or the other. Issue agree­ ment with the presidential nominee is the second variable expected to affect behavior in relation to the national party. The local leaders' perceived proximity to their candidate on a number of issues is mea­ sured, the expectation being the closer they perceive themselves to be, the more they will cooperate with the national leaders. This variable taps both Schlesinger's hypothesis about the need for attractive leader­ ship and Kaufman's idea about the importance of identification with the organization and its leadership. A third variable is more directly derived from Schlesinger, and that is the perceived attractiveness of the presidential candidate. An undesireable image of the elites in an organization makes eliciting obediance and cooperation more difficult.

Goldwater and McGovern both represented minorities in their parties and both had very disrupting influences on their party organizations. 27

Another factor presumed to affect relations with the national party

is the extent to which a local leader feels he is in the mainstream of

his party. Consensus was one of the dimensions for the study of parties

suggested by Anderson in his search of organization theory. If a local

leader is in agreement with his co-workers on most issues he will feel

a stronger identification with the organization than if he is "outside

the ballpark." So issue agreement both with the candidate and with the

rest of the party will be considered in studying consensus within each

organization.

Another dimension for the study of organizations is control. Most studies of control in American parties reveal considerable autonomy on

the part of local leaders; we know the many barriers to a hierarchical authority system. Control in this study refers to the obligations activists feel toward the party, the degree of support for the party as a hierarchical structure able to hold its members accountable. Perhaps this is not true control as intended by Anderson and organization theorists. But the more support there is for the party as an authority, the easier it will be for leaders to coordinate the party.

The final variable in this study expected to affect relations with the national party is the source of county leader's information about the national campaign. Communication has always been considered essential for efficient organization?^ Eldersveld's study cited the lack of intra-party communication as evidence for the freedom of precinct leaders ?^ Communication channels could well affect the picture local leaders have of what is going on in the campaign and therefore their 28 behavior. The question is whether they rely primarily on sources out­

side the party or sources within the party, the prediction being that

reliance on external sources leads to deviation from the national

party.

In the coming chapters considerable attention will be paid to univariate analysis of each of these variables and differences between

the parties. The state of research on this subject is such that we cannot even guess what simple distributions on these variables would be.

Following the univariate analysis hypotheses suggesting relationships among these variables are introduced and tested.

The final step will be to introduce a measure of the extent to which local leaders follow the national party. It is this measure of cohesion that we predict will be influenced by the variables derived from the organization and implementation literature and theories. More specifically this measure of cohesion is the resemblance between the local campaigns and the national campaigns. How closely do local leaders fashion their campaigns to the national effort? What is the impact of their motivations, the intra-party consensus, their obliga­ tions, and their image of their nominee on this cohesion?

In succeeding chapters we shall look at these questions in more detail and set forth the operational measures for each of the variables. 29

The Data

The. data are from the Television Election News Coverage Project funded by the American Enterprise Institute and directed by C. Richard

Hofstetter of the Ohio State University. The national cross-section sample contained 106 counties, the primary sampling units. Once the counties were selected the state chairpersons for the Committee to

Re-elect the President and Citizens for McGovern-Shriver were called and asked whom they relied upon most in the chosen county. They could name as many as three persons. The nominees from that county were interviewed by National Analysts of Philadelphia, Pennsylvania who had a completion rate of 807o. The final sample consisted of 401 county campaign leaders, 197 from Citizens for McGovern-Shriver and 204 from the Committee to Re-elect the President. For a more detailed discussion of the project see Chapter Two of Television and Civic Education, Vol. 2,

1976, by C. Richard Hofstetter, a Polimetrics Laboratory Report (mimeo),

The Ohio State University. FOOTNOTES

1. Robert Agranoff (ed.), The New Style in Election Campaigns (Boston: Holbrook Press, Inc., 1972), pp. 3-27. Frank Sorauf, Party Politics in America, 2nd edition (Boston: Little, Brown & Co., 1972), pp. 255-261.

2. Frank Sorauf, Party Politics in America, 2nd edition (Boston: Little, Brown & Co., 1972), pp. 104-106.

3. Samuel C. Patterson, "Characteristics of Party Leaders," Western Politics Quarterly, Vol. 16, No. 2 (June, 1973), pp. 332-352, reprinted in David W. Abbott and Edward T. Rogowsky, eds., Political Parties: Leadership, Organization, Linkage (Chicago; Rand McNally, 1971), p. 40.

4. Daiid B. Truman, "Federalism and the Party System," Arthur W. MacMahon, Federalism, Mature and Emergent (New York: Doubleday & Co., 1955), p. 118.

5. Austin Ranney and Willmoore Kendall Democracy and the American Party System (New York: Harcourt, Brace & World, Inc., 1956), Chapters 10-12. Austin Ranney The Governing of Men (New York: Henry Holt & Co., 1958), pp. 269, 312-327 and Austin Ranney, "Changing the Rules of the Nominating Game," in James D. Barber (ed.) Choosing the President (Englewood Cliffs, N.J.: Prentice-Hall, 1974), pp. 71-93.

6. Arthur Holcombe, Our More Perfect Union (Cambridge: Harvard University Press, 1950).

7. Frank Sorauf, Party Politics in America, 2nd edition, p. 117.

8. Ibid., pp. 118-119.

9. David Ogden, Jr. and Arthur Peterson, Electing the President, revised edition (San Francisco: Chandler, 1968), pp. 95-121.

10. Cornelius P. Cotter and Bernard C. Hennessy, Politics With­ out Power (New York: Atherton, 1964).

11. Agranoff, pp. 8-12.

30 31

12. Ogden and Peterson, pp. 95-98, and Theodore White, The Making of the President 1972 (New York: Atheneum Publishers, 1973), p. 316.

13. Ogden and Peterson, pp. 102-107.

14. Ibid., pp. 133 and 152.

15. Ibid., p. 109.

16. Theodore White, The Making of the President, 1968 (New York: Athemeun, 1969), pp. 414-426.

17. Theodore White, pp. 274-279.

18. Theodore White, The Making of the President, 1972 (New York: Bantam, 1973), p. 316.

19. Gary W. Hart, Right From the Start (New York: Quadrangle, 1973), pp. 277-278.

20. Ogden and Peterson, pp. 122-155.

21. Samuel C. Patterson, "Characteristics of Party Leaders," p. 35.

22. Lee F. Anderson, "Organization Theory and State and Local Parties," in William J. Crotty, Approaches to the Study of Party Organ­ ization, (Boston: Allyn and Bacon, Inc., 1968), pp. 389-397.

23. Joseph A. Schlesinger, "Political Party Organization," in Handbook of Organizations, ed. by James G. March (Chicago: Rand McNally, 1965), p. 768.

24. Peter B. Clarke and James Q. Wilson, "Incentive Systems: A Theory of Organization," Administrative Science Quarterly, VI (Sept­ ember, 1961), pp. 129-166.

25. Schlesinger, "Political Party Organization," p. 770.

26. C. Richard Hofstetter, "Organizational Activists: The Bases of Participation in Amateur and Professional Groups," American Politics Quarterly, Vol. 1, No. 2 (April, 1973), pp. 244-276.

27. Schlesinger, p. 774.

28. Ibid., pp. 790-792.

29. Ibid., p. 792. 32

30. Herbert Kaufman, Administrative Feedback (Washington: Brook­ ings, 1973), pp. 2-3.

31. Ibid., p. 5.

32. F. A. Lazen, "The Failure of Federal Enforcement of Civil Rights Regulations in Public Housing 1963-1971: The Co-optation of a Federal Agency by its Local Constituency," Policy Sciences (1973), pp. 263-273. Philip Selznick, TVA and the Grass Roots. (New York: Harper and Row, 1966).

33. Harold Guetzkow, "Communication in Organizations," in Micropolitics ed. by John Kessel, George Cole, and Robert Seddig (Chicago: Holt, Rinehart, Winston, 1970), pp. 475-494.

34. Eldersveld, Political Parties: A Behavioral Analysis (Chicago: Rand McNally, 1964), p. 408. CHAPTER TWO MOTIVATIONS, CONSENSUS, AND OBLIGATIONS

What motivates people to be active in party politics? Incentives may be material, social, ideological or simply for the fun and excite­ ment of it. Some political activists are so accustomed to working in party politics that their participation has become a habit. One body of literature divides activists into two types, amateurs and professionals, based on their incentives. Consequences are suggested for organizations dominated by one type or the other, and it is these consequences that make the examination of motivations relevant to a study of party unity.

A second question important to party unity is to what extent do members of a party organization agree with each other and with their presidential candidate? The greater the consensus, the easier it will be to coordinate a campaign.

What obligations do party activists feel toward their organiza­ tion? Do they support the idea of a party which holds its members accountalbe? Or do they prefer a looser organization? A strong feeling of obligation to the party indicates an acceptance of the party as a legitimate authority.

This chapter explores these questions giving particular attention to party differences, consequences for the organizations, and the possi­ ble influence of campaign events on the findings.

33 34

Amateurs and Professionals

What led the county leaders to become active in the '72 presiden­ tial campaign? The most widely used categorization of motives for party activism has been that of amateurs and professionals. Scholars seem to agree on what motivates the amateurs, a commitment to a set of princi­ ples or to an issue.^ The amateur's initial participation is due to issues or ideology and those motives continue to tie him to the party organization. The interest in policy-oriented activism has been sparked by the seeming increase in these types in recent campaigns. Many party organizations, such as the 1972 Democratic National Convention, have changed their structure so as to encourage persons who wish to parti­ cipate out of a sense of ideology. Wilson, writing on political organ­ izations attributes the interest in issues to the decline of patronage and material rewards, factors which have contributed to the decline of parties in general. Since tangible rewards have become less accessible, this leaves purpose, principle and ideology as a major source of incen- 2 tives for party organizations. Wilson cites the rise of amateur poli­ tical clubs whose members are concerned about the goals of public policy and the views of the candidates. Although these amateurs are not insensitive to partisan or personal advantage, they are more ser-

O ious about the content of public policy than other activists.

In contrast to the single motive distinguishing the amateurs, there number of motives gathered under the label of professional. Most of these motives have to do with social satisfactions gained from party activity. C. Richard Hofstetter lists party loyalty, an intrinsic enjoyment of politics, and the development of organizationally-linked friendships.^ Wilson's motives for professionals include material

rewards, personal ambition, social and business contact, party loyalty,

links to the community, and fun and excitement."’ Because of the social

orientation of his motives, the professional is more concerned with the

means of conducting party activity than with the goals of that activity.

There is some evidence that motives change with the length of one's

involvement in the party, becoming more professional over time. Elders-

vled found in his study of Detroit activists that 74% of the Democrats

and 85% of the Republicans said that their original motive was some

ideological or moral satisfaction, but only 7% and 20% reported those

C. as their current motives. Ippolito, too, found differences between

initial and current incentives, the latter being less policy oriented.^

Finally, Conway and Feigert in their study of party leaders in an

Illinois and a Maryland county reported the two-thirds or more of the

precinct chairmen who entered party work for ideological or other

impersonal reasons said that what they would miss most were more per- g sonal and social satisfactions.

The dominance of amateur motives has implications for the organiza­

tion. Joseph Schlesinger has pointed out that organizations with pur-

positive incentives lose flexibility due to the adherence of the members g to their issue stands. Concern for policy also implies a concern for mechanisms such as intra-party democracy to insure that the policies are

followed."^ The concern over procedure inclines amateur politicians

to form clubs. However, Wilson suspects that most people who are active in local

political parties are in it for neither material rewards nor policy ends,

but find politics basically enjoyable.^ Eldersveld, too, found that

three quarters of the Democrats and over half of the Republicans men­

tioned social contact and the fun of being on the inside as the chief 12 attractions of party work. Three incentives, habit, party loyalty, and policy, were clearly more important to the 1972 county leaders than any other motivations (Table 2.1). However, the distributions of these incentives was different for the two parties. The single most popular response was that politics was a way of life, confirming Wilson's idea about the intrinsic enjoyment of politics as well as the fact that activists come from politically active families. While this was by far the most common incentive among Republicans (54.9%), the Democrats were motivated by both way of life (40.1%) and policy incentives (40.1%) equally. We will probably find more amateurs in the Democratic party.

Only 23% of the Republicans mentioned policy incentives.

Party loyalty was another motive receiving a substantial number of responses from both parties (34% of the Democrats and 45.5% of the

Republicans). Notice that the motives indicating long service such as party loyalty and way of life were cited more frequently by Republicans who we have found to more experienced in party politics.

Two other motives were mentioned by over 10% of the activists, but neither of these were as common incentives as way of life, party loyalty, or policy. Both friendship with a candidate and attraction to power each received twenty or more responses from Republicans and Democrats. All 37

TABLE 2.1: MOTIVATIONS FOR INVOLVEMENT IN THE CAMPAIGN *

Republican Democratic %** N %N

Way of life 54.9 (112) 40.1 (79) Influencing policies 23.0 (47) 40.1 (79) Party loyalty 45.5 (93) 34.0 (67) Personal friendship with candidate 12.2 (25) 18.2 (36) Being close to power 9.8 (20) 12.1 (24) Personal ambition 6.3 (13) 2.5 (5) Social contacts 5.3 (11) 3.5 (7) Fun and excitement 2.4 (5) 3.0 (6) Business contacts 3.9 (8) 1.0 (2) Community obligation 3.9 (8) 1.5 (3) Community recognition 2.9 (6) 2.0 (4) Other 57.3 (117) 67.0 (132)

* The item used to measure motivation was "What, specifically, led you to become active in this campaign?" There were two alternate measures of motivation in the data. One asked why the local leader got involved in his first campaign, and the second tapped ongoing motivations, or what was keeping the local leader active. These three items allow for an analysis of changing motivations in a local leaders career. However, this research focused on the current motive since the objective was to study the campaign organizations in 1972. This item, more than the others, forced the respondent to focus on the campaign at hand.

** Three responses were possible to the question and all responses are in the table, so of course the N is greater than the sample size, and the percentages add up to more than 100%. 38

other motives were minor, mentioned by fewer than 5% of each party.

A word should be said here about the "Other" category since so

many answers could not be categorized. These categories were designed

to identify amateurs and professionals, and excluded incentives not

associated with one type or the other. A substantial number of the

"Other" motives were attraction to Nixon or attraction to McGovern.

These motives were quite different from what Eldersveld found in

1963. Perhaps the basic distinction is between a national sample and

the characteristics of activists in one local area. Social contacts

and attraction to power, motives common in the earlier sample ranked

quite low on the list for both parties in 1972. There is also a

contrast in the importance attached to policy. Only 3% of the Detroit

Democrats and 17% of the Republicans mentioned issues as a motive. ^

However, in 1972 this was cited by 23% and 40% of the county leaders.

We might expect county leaders in a presidential campaign to be more

concerned with policy since they play more of an elite role than precinct leaders. But beyond what might be explained by position in

the party we should consider the attention given to issues in the campaigns in the 1960's and the particular appeals of the 1972 candi­ dates. Numerous studies have indicated an increase in issue awareness

14 since 1960. If events are making issues more salient to the general population, then surely elites such as party leaders are even more sensitive to such events. Also the appeals of McGovern probably enhanced issue awareness. Goldwater's emphasis on the conservative ideology in 1964 was accompanied by an increased ideological sophisti­ cation in the electorate.^ So, too, we might expect the candidacy 39

of a left wing Democrat to raise the level of issue thinking among

activists. Considering these events it is not surprising so many

county leaders were concerned with policy.

Returning to the amateurs and professionals, two operationaliza­

tions were used in an attempt to make a meaningful distinction and

test the applicability of these types to the 1972 county leaders.

Operationalization 1 : Amateurs were those whose first response

to the question of why they got involved was to influence policies.

Professionals were those whose first response was a professional res­

ponse (way of life, party loyalty, social or business contacts, fun,

ambition, being close to power, community obligation or recognition).

Operationalization 2: Amateurs were those whose first response

was influencing policies (same as above).

Semi-professionals were those whose second or third response was

influencing policies.

Professionals gave no policy response.

Table 2.2 shows the distribution of amateurs and professionals by

the first definition. The most notable result is the absence of

amateurs even among Democrats. The explanation may lie in the roles

performed by county leaders. A considerable amount of organizational

responsibility is placed upon these activists, and the party is not

likely to risk these responsibilities with a novice. These local

leaders are experienced in partisan politics (see Table l A ) and motives tend to become more professional over time. We should expect

such veteran activists to mention motives other than policy first. 40

TABLE 2.2: AMATEURS AND PROFESSIONALS AMONG THE 1972 COUNTY CAMPAIGN LEADERS*

Republicans Democrats % N % N

Professionals 69.1 (141) 54.3 (107)

Amateurs 5.9 (12) 11.7 (23)

Other 25.0 (51) 34.0 (67)

Total 100% (204) 100% (197)

* Amateurs are those whose first response to the question of why they got involved in the 1972 campaign was to influence policies. Pro­ fessionals are motivated by habit, party loyalty, social or business contacts, fun, ambition, being close to power, community obligation or recognition. 41

However, policy was the most popular second motive. Issues, then, are

in the minds of most county leaders. It is possible that amateur-

professional distinction is less meaningful at this level of party

elites.

The data in Table 2.3 confirm this suspicion. Everyone is inter­

ested in policy, and in committed candidates, be they amateurs or pro­

fessionals. The amateur-professional distinction was not developed on

the basis of a large sample of party leaders, but rather on rare types

of organizations at both ends of the spectrum such as the Lexington

Avenue Democratic Club and the Daley Organization. These concepts may

not be very useful, particularly at the level of county leader and with the increasing interest in issues generally.

The second operationalization proved to be no more useful (Table

2.4) for the same reason. The widespread concern with issues precludes

dividing the sample on that basis.

Motives for being in this particular campaign were examined because it was thought they would affect attitudes toward the organiza­

tion and behavior. Apparently there is not much variety in county leaders motives, and nearly all county leaders are concerned with policy.

A division on the basis of motives may be telling us how long the local leader has been in party work. Or it may be identifying local leaders who are in the campaign for one issue in particular. But it does not indicate an orientation toward policy in general. 42

TABLE 2.3: AMATEURS' AND PROFESSIONALS' INTEREST IN ISSUES*

Amateurs Professionals % N% N

Obligation to select a nominee who is strongly committed on a variety of issue positions

Strong obligation to do 61.8 (21) 60.4 (148) Some obligation to do 14.8 (5) 24.9 (61) No obligation either way 17.6 (6) 12.7 (31) Some obligation to avoid 2.9 (1) 1.6 (4) Strong obligation to avoid 2.9 (1) 0.4 (1)

Total 100.0 (34) 100.0 (245)

Obligation to hold strong beliefs about a number of different issues

Strong obligation to do 70.3 (24) 63.0 (155) Some obligation to do 11.8 (4) 24.0 (59) No obligation either way 11.8 (4) 8.9 (22) Some obligation to avoid 0.0 (0) 1.6 (4) Strong obligation to avoid 5.6 (2) 2.5 (6)

Total 100.0 (34) 100.0 (247)

* Amateurs are those 1972 county campaign leaders who are motivated by policy to participate. Professionals are those motivated by habit, party loyalty, social or business contacts, fun, ambition, being close to power, community obligation or recognition. 43

TABLE 2.4: AMATEURS', SEMI-PROFESSIONALS', AND PROFESSIONALS' INTEREST IN ISSUES *

Amateurs Semi-Prof. Prof. % N %N % N

Obligation to select a nominee who is strongly committed on a variety of issues

Strong obligation to do 61.8 (21) 67.8 (61) 65.6 (21) Some obligation to do 14.7 (5) 24.4 (22) 12.5 (4) No obligation either way 17.7 (6) 7.8 (7) 18.8 (6) Some obligation to avoid 2.9 (1) 0 3.1 (1) Strong obligation to avoid 2.9 (1) 0 0

Total 100% (34) 100% (90) 100% (32)

Obligation to hold strong personal beliefs about a number of different issues.

Strong obligation to do 70.5 (24) 75.6 (68) 67.7 (21) Some obligation to do 11.8 (4) 12.2 (11) 25.8 (8) No obligation either way 11.8 (4) 5.6 (5) 3.2 (1) Some obligation to avoid 0 2.2 (2) 0 Strong obligation to avoid 5.9 (2) 4.4 (4) 3.2 (1)

Total 100% (34) 100% (90) 100% (31)

* Amateurs are those 1972 county campaign leaders whose first response to the question of why they got involved in the campaign was influen­ cing politics. Semi-professionals are those whose second or third response was influencing politics. Professionals gave no policy response. Intra-Party Issue Consensus

This section focuses on agreement between the presidential candi­

dates and their workers, and consensus among the workers themselves.

Just how important policy consensus is has been a subject of contro­

versy for students of parties. One view holds that parties are able to mobilize individuals with diverse purposes to support common positions

(Holcombe). If parties encouraged value conflicts they would destroy

the balance needed to maintain their group support. Another view is

that parties are mechanisms solely for the election of candidates, not proponents of policy. Obviously, if policy consensus is not crucial or even significant for the conduct of party activities there is little point in including it in a study of parties as organizations. Why then consider issue consensus? First and most important is because of the siginificance attached to policy by activists themselves. A third mentioned the desire to influence the policies of government as a motive for their involvement in the campaign. Moreover. 86% felt an obligation to select a nominee who is strongly committed on a variety of issue positions, and 86% felt an obligation to hold strong personal beliefs about a number of issues. Herbert McClosky et al in their study of policy consensus give us one explanation. Leaders of the parties have more incentive to hold opposing views on political question

They are articulate, informed, and involved compared to the general public, and their involvement makes it ieasier for them to be rewarded or punished for nonconformity. In light of these findings we should not belittle the role of issues in party involvement. 45

Second, the new politics described in Chapter 1 is moving the focus

in campaigns from party to candidates and images, and issues are a part

of those images. Mass media increases the awareness of issues by

nationalizing them, i.e., we are all made aware of the problems of

blacks, the underprivileged, the aged, etc., making it difficult for

political activists to ignore issues.

Finally, the importance of policy consensus has to do with the purpose of the campaign organization, which is to elect the presidential

candidate. A candidate espousing unattractive issue positions will simply not receive the potential of party support. Put another way, local leaders have resources to offer, and they will be reluctant to offer them to a candidate they perceive to be out of step with them­ selves and their constituents. The history of the McGovern campaign is replete with examples of this sort of conflict.

Agreement with the candidate is considered to be an incentive for commitment to the campaign. Since candidates' stands are not always clear and can at times be interpreted by one's preference, we will first examine just what issue positions the activists think the candi­ dates were taking. As Table 2.5 indicates, there was some disagree­ ment between Democrats and Republicans as to who would do what. Ken­ dall's Tau^ gives a summary measure of disagreement. But there is a problem in that Tau^ is capable of taking on equally high values if the party disagreement is ordered one magnitude apart or six magnitudes apart. However, the percentage tables contain the information lost in the statistic. 46

TABLE 2.5: 1972 COUNTY CAMPAIGN LEADERS' PERCEPTIONS OF THE CANDIDATES' ISSUE POSITIONS

Republicans Democrats % N % N

McGovern would stop the use of busing Tau = .02 * b

Strongly Disagree 34 (63) 29 (53) Disagree 8 (15) 10 (19) Slightly Disagree 38 (69) 45 (84) Not Sure 13 (23) 9 (17) Slightly Agree 2 (3) 4 (8) Agree 2 (4) 2 (3) Strongly Agree 3 (6) 1 (2)

Total 100 (183) 100 (198)

Nixon would stop the use of busing Tau = -.14 b Strongly Disagree 3 (5) 12 (22) Disagree 6 (11) 13 (24) Slightly Disagree 4 (7) 6 (12) Not Sure 8 (14) 10 (18) Slightly Agree 14 (26) 11 (20) Agree 48 (87) 28 (53) Strongly Agree 17 (31) 20 (38)

Total 100 (181) 100 (187) 47

Table 2.5 (cont'd)

Republicans Democrats % N % N

McGovern.would spend whatever necessary to have a strong military .53 Tav Strongly Disagree 45 (82) 8 (16) Disagree 33 (60) 11 (20) Slightly Disagree 9 (17) 8 (16) Not Sure 5 (9) 4 (7) Slightly Agree 0.5 (1) 10 (19) Agree 5 (9) 34 (63) Strongly Agree 3 (6) 25 (46)

Total 99.5 (184) 100 (187)

Nixon would spend whatever necessary to have a strong military Taub= .04

Strongly Disagree 1 (2) 5 (10) Disagree 3 (6) 4 (7) Slightly Disagree 3 (6) 3 (5) Not Sure 1 (2) 3 (5) • Slightly Agree 10 (18) 6 (12) Agree 49 (92) 36 (67) Strongly Agree 33 (62) 43 (79)

Total 100 (188) 100 (185) 48

Table 2.5 (cont'd)

Republicans Democrats %N % N

McGovern would bring peace in Vietnam Tau, = .69 b

Strongly Disagree 25 (45) 1 (2) Disagree 22 (40) 1 (2) Slightly Disagree 11 (20) 0 (0) Not Sure 12 (22) 0.5 (1) Slightly Agree 7 (12) 3 (6) Agree 17 (31) 18 (34) Strongly Agree 6 (11) 76 (143)

Total 100 (181) 99.5 (188)

Nixon would bring peace in Vietnam Tau = -.67 b Strongly Disagree 2 (3) 25 (46) Disagree 1 (2) 23 (42) Slightly Disagree 0 (0) 13 (25) Not Sure 2 (3) 19 (35) Slightly Agree 4 (8) 9 (17) Agree 43 (82) 8 (15) Strongly Agree 48 (93) 3 (6)

Total 100 (191) 100 (186) Table 2.5 (cont'd)

Republicans Democrats % N % N

McGovern would increase welfare payments Tau = -.30 b Strongly Disagree 4 (8) 7 (12) Disagree 3 (6) 10 (18) Slightly Disagree 2 (3) 5 (9) Not Sure 5 (9) 13 (24) Slightly Agree 8 (14) 16 (30) Agree 43 (81) 37 (68) Strongly Agree 35 (65) 12 (23)

Total 100 (186) 100 (184)

Nixon would increase welfare payments Tau = -.23 b Strongly Disagree 8 (14) 25 (45) Disagree 24 (43) 36 (66) Slightly Disagree 17 (32) 10 (18) Not Sure 17 (32) 6 (11) Slightly Agree 20 (36) 12 (22) . Agree 11 (20) 8 (14) Strongly Agree 3 (6) 4 (7)

Total 100 (183) 100 (183) 50

Table 2.5 (cont'd)

Republicans Democrats %______N %_____ N

McGovern would give the police more authority Tau = .14 b Strongly Disagree 30 (54) 20 (37) Disagree 34 (63) 28 (51) Slightly Disagree 11 (20) 20 (36) Not Sure 17 (31) 14 (26) Slightly Agree 2 (4) 8 (15) Agree 3 (6) 8 (14) Strongly Agree 3 (5) 2 (4)

Total 100 (183) 100 (183)

Nixon would give the police more authority Tau = .16 b Strongly Disagree 3 (6) 3 (6) Disagree 7 (12) 5 (9) Slightly Disagree 5 (8) 3 (5) Not Sure 8 (14) 10 (18) Slightly Agree 18 (32) 10 (18) Agree 49 (88) 38 (70) Strongly Agree 10 (18) 31 (56)

Total 100 (178) 100 (182)

* Higher coefficients mean greater differences between the parties' perceptions of the candidates on issues. The greatest difference in party perceptions concerns which candi­

date would bring peace in Vietnam. This is understandable since both

candidates were committed to ending the war as soon as possible. In

a sense, a high Tau correlation reflects the extent to which county

leaders projected their own ideas onto candidates. Two processes

known as the assimilation and dissociation effects seem present in

activists' perceptions of the candidates. These are tendencies to

perceive people that we agree with as being closer to us than they

actually are, and those with whom we disagree as being further from

our own position than is actually the case. For instance, in face of

the overwhelming public pressure to end the war, over two-thirds of

each party were unwilling to acknowledge that the other candidate would

bring peace as president. In another example, McGovern's proposal to

dramatically cut the military budget brought strong reactions from

Republicans. They did not perceive McGovern as willing to spend

enough to keep the military strong. Democrats, on the other hand, could not accept the idea that their candidate would endanger national secur­

ity, so they responded that he would spend whatever necessary.

There was a slight party disagreement over what the candidates would do concerning welfare. The demogrant program of McGovern to guarantee a $1,000 annual income to every American probably influenced these results. 86% of the Republicans agreed that McGovern would increase welfare payments and considering their view on increased wel­ fare, this high proportion reflects a perception of McGovern irrespon­ sibly increasing payments. Democrats reacted in a similar fashion to 52

Nixon on welfare, making his position more extreme than his own party members perceived it.

The issues where parties agreed on candidates' stands were busing

and police authority. Consensus indicates, first, that the activists

are themselves divided along partisan lines, and second, that they are fairly sure of what the candidates' positions are, thereby making pro­ jection of their own opinions less likely.

Perhaps these results speak well for the party organizations. The activists are clearly psychologically involved in the struggle for office, seeing their own and the opposition candidates through the screen of their own predispositions. Next we will see how united the activists are in these perceptions.

The measure of intra-party consensus is a measure used by John

Kessel in The Domestic Presidency. ^ It compares the actual distribu­ tion to the distribution that would exist if respondents were distri­ buted equally across the scale. Using the example of a seven-point scale the formula would be:

f(obs) - f (3/7) f(all) - f (3/7) where f(obs) is the observed frequencies within one magnitude on either side of the mode, f (3/7) is the frequency that would exist within one magnitude of the mode if cases were distributed equally and f(all) is the total frequency. This measure ranges from 0.0 meaning no consensus

18 to 1.0 meaning complete consensus. 53

TABLE 2.6: PERCEPTIONS OF THE CANDIDATES: A MEASURE OF INTRA-PARTY CONSENSUS *

Republicans Democrats

McGovern would stop the use of busing .65 ,72

Nixon would stop the use of busing .64 .29

McGovern would give police more authority .56 .43

Nixon would give police more authority .61 .63

McGovern would spend whatever necessary to have a strong military .68 .45

Nixon would spend whatever necessary to have a strong military .85 .71

McGovern would bring peace in Vietnam .26 .92

Nixon would bring peace in Vietnam .88 .27

McGovern would raise welfare payments ,75 .40

Nixon would raise welfare payments .10 .48

* High coefficients mean greater intra-party consensus. As was the case with candidate perceptions, consensus depends on

the issue involved. Republicans are in near total agreement that Nixon

would spend on the military and that he would bring peace in Vietnam.

They also agreed on McGovern's welfare position. Recalling Table 2.5,

these are the same issues on which the candidates were perceived so

differently. Perhaps the high visibility of these issues made it more

imperative that the activist perceived his candidate to be on the

"right" side, and the same process encouraged intra-party consensus.

The Democrats, as we would expect, exhibited less overall consensus and

the scores were much more erratic. Only two Republican scores were less than .50, but six of the Democrats' scores were below that level. Again there is near total agreement that their own candidate would bring peace in Vietnam. But on the other salient issue, military spending, only

Nixon's position was clear to Democrats; McGovern's proposed defense budget cut provoked some confusion in his party. His economic program had a similar effect. His own workers were not quite sure what he meant. Many questions arose from his income equalization program that neither McGovern nor his academic advisors could answer. It was unclear just what kind of taxes would be required and who would pay what.

There was virtually no consensus among the Democrats as to whether

Nixon would stop busing or bring peace in Vietnam. Some Democrats may have felt that busing was inevitable, beyond Nixon's power to stop.

Regarding Vietnam, we have noted the reluctance of some partisans to admit that the opposing candidate might bring peace. Also some county leaders may have expected the war to end no matter who was president.

Combining these activists with those who would not admit that the 55

opposition would bring peace results in low consensus.

In sum, Republicans generally agreed as to what Nixon's positions were on the major issues of Vietnam and military spending. However,

Democrats were confused about McGovern on military spending and were generally more divided on their perceptions of the candidates. It is likely that agreeing on what candidates stand for is tied to the acti­ vist's own issue position, i.e., when activists themselves agree on an issue, they will tend to have the same perceptions of the candidate.

The issues on which the parties differ most concern social welfare, civil liberties and a few miscellaneous social change issues. Some prior research would lead us to expect members of opposing parties to diverge on the traditional liberal-conservative, governmental inter­ vention issues. McCloskey concluded that the opinions of the party elite are linked by a common ideology. The Democrats favored using legislation to redistribute wealth and supported a wide variety of social services. Republicans, supported a more laissez-faire approach 19 to social problems and favored continuance of status distinction. In two other studies using multi-dimensional scaling techniques on the

1968 and 1970 SRC election surveys, Weisberg and Rusk found two dimen­ sions of candidate evaluation: (1) traditional party affiliation and

(2) the issues of the day. The first traditional dimension, on which the major party candidates were evaluated, consisted of party identifi­ cation and social welfare issues. Although the two dimensions were not as distinct in 1970 as they were in 1968, the social welfare questions that realigned the parties in the 1930's still formed the basis of party 56

TABLE 2.7: THE ISSUE POSITIONS OF THE 1972 COUNTY CAMPAIGN LEADERS

Republicans Democrats %N % N

The government should aid foreign countries Tau, = .15* b Strongly Disagree 9.8 (18) 9.7 (18) Disagree 26.1 (48) 15.7 (29) Slightly Disagree 19.6 (36) 11.4 (21) Not Sure 7.1 (13) 8.6 (16) Slightly Agree 19.0 (35) 23.8 (44) Agree 12.0 (22) 16.8 (31) Strongly Agree 6.5 (12) 14.1 (26)

Total 100% (184) 100% (185)

American troops should be brought home Tau, = .30 D

Strongly Disagree 18.7 (35) 3.7 (7) Disagree 36.9 (69) 19.5 (37) Slightly Disagree 10.2 (19) 16.3 (31) Not Sure 7.5 (14) 11.1 (21) Slightly Agree 14.4 (27) 16.8 (32) Agree 5.3 (10) 11.1 (21) •Strongly Agree 7.0 (13) 21.6 (41)

Total 100% (185) 100% (190)

America should spend whatever is necessary to have a strong military Taub~ -.39

Strongly Disagree 2.6 (5) 21.8 (41) Disagree 6.3 (12) 16.5 (31) Slightly Disagree 3.7 (7) 9.6 (18) Not Sure 1.6 (3) 4.8 (9) Slightly Agree 16.3 (31) 14.4 (27) Agree 41.6 (79) 26.1 (49) Strongly Agree 27.9 (53) 6.9 (13)

Total 100% (190) 100% (188) 57

Table 2.7 (cont'd)

Republicans Democrats %N % N

We must have peace in Vietnam Tau, = .25 b

Strongly Disagree 0.5 (1) 2.6 (5) Disagree 1.6 (3) 1.6 (3) Slightly Disagree 1.6 (3) 0 0 Not Sure 2.7 (5) 1.1 (2) Slightly Agree 7.4 (14) 1.6 (3) Agree 29.8 (56) 11.1 (21) Strongly Agree 56.4 (106) 82.0 (155)

Total 100% (188) 100% (189)

Government spending should be cut -.25 Tav Strongly Disagree 2.6 (5) 3.7 (7) Disagree 3.2 (6) 14.4 (27) Slightly Disagree 3.2 (6) 11.8 (22) Not Sure 2.6 (5) 9.1 (17) Slightly Agree 13.7 (26) 14.4 (27) Agree 37.9 (72) 22.5 (42) Strongly Agree 36.8 (70) 24.1 (45)

Total 100% (190) 100% (187)

The federal government is getting too powerful Taub= -.08

Strongly Disagree 5.3 (10) 4.8 (9) Disagree 7.9 (15) 14.8 (28) Slightly Disagree 8.5 (16) 11.6 (22) Not Sure 1.6 (3) 4.8 (9) Slightly Agree 14.8 (28) 14.3 (27) Agree 32.8 (62) 21.7 (41) Strongly Agree 29.1 (55) 28.0 (53)

Total 100% (189) 100% (189) 58

Table 2.7 (cont'd)

Republicans Democrats % N % > N

The government should control wages and prices Tau, = -.09 b Strongly Disagree 4.8 (9) 10.3 (19) Disagree 24.5 (46) 26.1 (48) Slightly Disagree 20.2 (38) 19.6 (36) Not Sure 6.9 (13) 10.3 (19) Slightly Agree 15.4 (29) 13.6 (25) Agree 19.1 (36) 11.4 (21) Strongly Agree 9.0 (17) 8.7 (16)

Total 100% (188) 100% (184)

The government should help pay people's medical bills Tau, = .54 b

Strongly Disagree 31.6 (59) 3.8 (7) Disagree 29.4 (55) 9.9 (18) Slightly Disagree 15.0 (28) 3.8 (7) Not Sure 7.5 (14) 7.1 (13) Slightly Agree 7.5 (14) 18.7 (34) Agree 8.6 (16) 23.6 (43) Strongly Agree 0.5 (1) 33.0 (60)

Total 100% (187) 100% (182)

Social Security benefits should be increased Tau = .42 b

Strongly Disagree 2.7 (5) 1.1 (2) Disagree 11.2 (21) 0 (0) Slightly Disagree 8.0 (15) 0.5 (1) Not Sure 15.0 (28) 5.9 (11) Slightly Agree 24.1 (45) 11.8 (22) Agree 27.8 (52) 39.6 (74) Strongly Agree 11.2 (21) 41.2 (77)

Total 100% (187) 100% (187) Table 2.7 (cont'd)

Republicans Democrats fr'S % N N

Government should give aid 00 to the underprivileged Tau, = b

Strongly Disagree 1.1 (2) 2.1 (4) Disagree 4.8 (9) 0 (0) Slightly Disagree 7.4 (14) 0 (0) Not Sure 6.9 (13) 1.6 (3) Slightly Agree 22.8 (43) 3.7 (7) Agree 45.0 (85) 35.4 (67) Strongly Agree 12.2 (23) 57.1 (108)

Total 100% (189) 100% (189)

Welfare payments should be increased Tau, = .53 b Strongly Disagree 18.9 (35) 2.8 (5) Disagree 32.4 (60) 6.2 (11) Slightly Disagree 14.1 (26) 5.6 (10) Not Sure 15.1 (28) 11.2 (20) Slightly Agree 14.1 (26) 17.4 (31) Agree 3.8 (7) 35.4 (63) .Strongly Agree 1.6 (3) 21.3 (38)

Total 100% (185) 100% (178)

I favor letting Negroes move into white neighborhoods Tau = .45 b Strongly Disagree 4.9 (9) 0 (0) Disagree 5.9 (11) 1.1 (2) Slightly Disagree 5.9 (11) 1.6 (3) Not Sure 8.6 (16) 2.7 (5) Slightly Agree 17.3 (32) 5.9 (11) Agree 40.5 (75) 26.1 (49) Strongly Agree 16.8 (31) 62.8 (118)

Total 100% (185) 100% (188) 60

Table 2.7 (cont’d)

Republicans Democrats % N % N

Busing should be used to desegregate schools Taub= -.55

Strongly Disagree 5.4 (10) 30.9 (58) Disagree 4.3 (8) 29.3 (55) Slightly Disagree 1.1 (2) 17.6 (33) Not Sure 2.7 (5) 4.3 (8) Slightly Agree 9.7 (18) 2.1 (4) Agree 40.9 (76) 9.6 (18) Strongly Agree 36.0 (67) 6.4 (12)

Total 100% (186) 100% (188)

Police should be give more authority Taub- -.49

Strongly Disagree 4.8 (9) 27.3 (51) Disagree 11.8 (22) 35.3 (66) Slightly Disagree 5.9 (11) 17.1 (32) Not Sure 5.9 (11) 7.0 (13) Slightly Agree 21.9 (41) 4.8 (9) Agree 34.8 (65) 4.8 (9) . Strongly Agree 15.0 (28) 3.7 (7)

Total 100% (187) 100% (187)

The government should act to stop pollution Tau, = .36 b

Strongly Disagree 0.5 (1) 0 (0) Disagree 0 (0) 0 (0) Slightly Disagree 1.6 (3) 0.5 (1) Not Sure 2.1 (4) 1.1 (2) Slightly Agree 18.0 (34) 6.3 (12) Agree 50.8 (96) 26.8 (51) Strongly Agree 27.0 (51) 65.3 (124)

Total 100% (189) 100% (190) 61

Table 2.7 (cont'd)

Republicans Democrats % N % N

Marijuana should be legalized Taub= .50

Strongly Disagree 50.3 (96) 8.2 (15) Disagree 25.1 (48) 12.5 (23) Slightly Disagree 5.8 (11) 6.5 (12) Not Sure 5.2 (10) 15.8 (29) Slightly Agree 5.8 (11) 13.6 (25) Agree 3.1 (6) 25.0 (46) Strongly Agree 4.7 (9) 18.5 (34)

Total 100% (191) 100% (184)

A woman's place is in the home Taub= -.47

Strongly Disagree 15.7 (29) 60.0 (114) Disagree 28.6 (53) 26.3 (50) Slightly Disagree 11.9 (22) 5.8 (11) Not Sure 8.6 (16) 1.6 (3) Slightly Agree 14.1 (26) 4.2 (8) Agree 11.4 (21) 1.6 (3) Strongly Agree 9.7 (18) 0.5 (1)

Total 100% (185) 100% (190)

The law should allow anyone who wants to get an abortion Tau, = .38 b

Strongly Disagree 22.2 (40) 8.1 (15) Disagree 20.6 (37) 5.4 (10) Slightly Disagree 11.1 (20) 4.9 (9) Not Sure 6.7 (12) 5.9 (11) Slightly Agree 9.4 (17) 9.2 (17) Agree 20.6 (37) 22.7 (42) Strongly Agree 9.4 (17) 43.8 (81)

Total 100% (180) 100% (185) 62

Table 2.7 (cont'd)

Republicans Democrats % N % N

I favor farmers being guaranteed a good income ^aub= *07

StronglyDisagree 4.0 (7) 1.8 (3) Disagree 15.9 (28) 7.6 (13) Slightly Disagree 9.7 (17) 13.5 (23) Not Sure 15.3 (27) 20.5 (35) SlightlyAgree 26.7 (47) 22.8 (39) Agree 23.3 (41) 25.1 (43) Strongly Agree 5.1 (9) 8.8 (15)

Total 100% (176) 100% (171)

* Higher coefficients indicate greater party differences on issue positions. 63

20 identification.

Party differences in 1972 do not emerge on a broad liberal-conser- vation dimension, but rather on the more specific social welfare issues.

Both parties agree that government spending should be cut, and opinion

is fairly divided as to whether or not the federal government is getting too powerful or should control wages and prices. These are the issues of a broad left-right nature measuring approval of big government or government intervention.

Social welfare issues like medical care, social security benefits, aid to the underprivileged, and welfare payments turned up significant party differences, confirming a continuing party alignment on these issues.

Civil liberties was the second issue area where parties diverged.

Busing brought out the greatest difference of opinion with 77.8% of the Democrats approving and 86.6% of the Republicans disapproving.

Although both parties approved of integrated neighborhoods, Democratic approval was more emphatic. When police authority is the issue, the

Democrats clearly think they have enough authority while 71.7% of the

Republicans favored more. Civil liberties has been a major issue in elections now since 1964. Just how long it takes for an issue area to become a permanent part of the partisan alignment is unknown, but distinct party positions along with the changing partisan patterns in the South indicate an assimilation of civil liberties into party loyalties. 64

The three issues remaining on which parties differ, women's role,

the legalization of marijuana, and abortion do not belong to any recog­ nizable category, although they may represent a dimension of social change. The abortion issue, which could on substantive grounds be paired with women's role, did not show as much difference between the parties as the latter item. Perhaps the Catholic Democrats diluted the results by expressing a conservative view on abortion.

Foreign and military affairs present no clear picture of agreement or disagreement. Opinion was divided in both parties on foreign aid.

There was a solid consensus that we must have peace in Vietnam, but the

Republicans were less willing to bring American troops home. Military spending, even though it received so much publicity in the campaign was not the scene of sharp party differences. These results reflect the split in the Democratic party over foreign affairs and defense spending.

Henry Jackson exemplifies those favoring staunch anti-Communism and a strong defense and McGovern represents the liberal side. Also the norm of preserving national security and the wording of the question might have inclined more Democrats to agree with the Republicans.

To conclude, it is interesting that in a campaign where foreign and military affairs were so prominent, the parties were found to divide on the traditional social welfare policies. The notion of liberalism-conservatism in foreign affairs has undergone change and just what constitutes liberal foreign policy depends on whether one refers to the 1950's or the 1970's. Changing world and domestic situations have resulted in enough party shifts and splits over these issues to make difficult the identification of Republican or Democratic 65

foreign policy. Foreign and military affairs are more likely to be

issues of the day than enduring bases of partisanship.

Republicans and Democrats were nearly identical in their opinions

on a guaranteed income for farmers. A majority of both parties suppor­

ted such guarantees.

In Table 2.8 the measure of consensus is applied to activists' own

issue position. Again, the parties agree on different issues. Vietnam

and pollution are the only issues uniting both parties. Beyond that,

the Democrats agree on the issues that reflect the group bases of their

support, integration and aid to the underprivileged and women's role,

and Republicans agree on a tenet of conservatism, that government spen­

ding should be cut. The low consensus measures reflect coalition in

the parties. Republicans are split on the social welfare and social

change issues uniting Democrats and on foreign aid. Democrats are split

on foreign, military and farm issues. Control of wages and prices di­

vides both parties, probably due to a Republican president invoking the

controls, and interventionist move. Civil liberties items produced some

negative reactions. Republicans united in their opposition to busing

and Democrats did not want to give the police more authority. If we were to eliminate the effects of region, or race, or sex, the indices

of agreement would be greater. But since the party organization cannot

apply controls in real life, it has to deal with the mixture of consen­

sus and conflict. These data tend to support the notion of parties as groups of individuals with diverging purposes. TABLE 2.8: INTRA-PARTY ISSUE CONSENSUS*

Republicans Democrats

Government should aid foreign countries .22 .11

American troops should be brought home .40 .06

America should spend whatever necessary to have a strong military .75 .07

We must have peace in Vietnam .80 .90

Government spending should be cut .80 .26

The federal government is getting too powerful .59 .30

Government should control wages and prices .11 .23

Government should help pay people's medical bills .46 .39

Social security benefits should be increased .35 .73

Government should give aid to the underprivileged .65 .89

Welfare payments should be increased .39 .55

I favor letting Negroes move into white neighborhoods .56 .84

Busing should be used to desegregate schools .76 .45

Police should be given more authority .50 .64 67

Table 2.8 (cont'd)

Republicans Democrats

Government should act to stop pollution .92 .89

Marijuana should be legalized .66 .24

A woman's place is in the home .23 .81

Law should allow anyone who wants to get an abortion .20 .53

I favor farmers being guaranteed a good income .39 .11

* High coefficients mean greater intra-party consensus. Another way of viewing consensus is to examine the extent to which

individual party workers agree with their party. Figures 2.1 and 2.2

present this composite picture of issue consensus. A local leader was

considered outside his party's consensus on a given issue if he was not 21 within one point of the mode. These results modify the issue by issue analysis. Although there are a number of issues on which agreement is

low, any one activist will probably contribute to the dissensus on only a few issues. Out of eighteen issues, the average activist was out of line on five, less than one third. As we would expect, Republicans de­ monstrated the most consensus with 87% agreeing on about two-thirds of the issues. Consensus among Democrats was only slightly lower.

Reconciling the issue by issue analysis with this composite picture requires more exploration into the coalition nature of American parties.

Although there are a few persons who disagree with the party position on more than one-half of the issues, the "average" person disagrees on only five and agrees on thirteen. While there is low consensus on several issues, the extent of agreement is such that most party activists have enough issue "pay-off" to remain loyal.

Up until now we have been studying consensus among the county leaders. Agreement with the presidential candidate more directly assesses a motivation to work for his election and cooperate with the national party. Agreement with the candidate was measured by the dis­ tance between a local leader's self-placement on a seven-point agree­ ment/disagreement scale and his placement of the candidate. This distance measuring is a simple form of the spatial model. Applications of spatial models to American politics have been more theoretical, Percentage of Activists 20 15 10 5 FIGURE 2.1: NUMBER OF ISSUES ON WHICH 1972 REPUBLICAN COUNTY CAMPAIGN LEADERS CAMPAIGN COUNTY REPUBLICAN 1972 WHICH ON ISSUES OF NUMBER 2.1: FIGURE 1 3 5 7 9 0 1 2 3 4 5 6 7 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 * Being within the issue consensus means being within one magnitude of the mode of mode the of magnitude one within being means consensus issue the within Being * n' prys itiuino n su. te pstos r otie h issue the outside are positions Other issue. an on consensus. distribution party's one's ARE OUTSIDE THEIR OWN PARTY ISSUE CONSENSUS* ISSUE PARTY OWN THEIR OUTSIDE ARE Number of Issues X = 4.7 = X Issues of Number

Percentage of Activists 20 0 FIGURE 2.2: NUMBER OF ISSUES ON WHICH 1972 DEMOCRATIC COUNTY CAMPAIGN LEADERS ARE LEADERS CAMPAIGN COUNTY DEMOCRATIC 1972 WHICH ON ISSUES OF NUMBER 2.2: FIGURE en ihn h ise osnu mas en ihnoemgiueo temd of mode the of magnitude one within being means consensus issue the within Being * n' prys itiuino n su. te pstos r otie h issue the outside are positions Other issue. an on distribution party's one's consensus. 1 3 2 OUTSIDE THEIR OWN PARTY ISSUE CONSENSUS* ISSUE PARTY OWN THEIR OUTSIDE 4 6 5 ubr fIse X 5.1 = X Issues of Number 7 8 1 1 1 1 1 1 1 1 18 17 16 15 14 13 12 11 10 9

71

searching for dimensions on which voters and parties can be placed.

Donald Stokes has ably pointed out why it is inaccurate to apply a

spatial model to American party competition. One of the reasons he

cites is the lack of a common dimension on which the voters view poli- 22 tics. In this study the specific issue questions provide the common dimension although on a scale less grand than the left-right continuum.

The distance measure is perceived agreement with the candidate rather than an objective measure of agreement. A local leader's atti­

tude toward his candidate is more dependent on his perceptions than an objective measure of agreement. If an objective measure had been used it is possible that many local leaders would have been unaware of their positions vis a vis their candidate, an undesirable situation from the standpoint of this study.

The data in Table 2.9 show both groups of activists to be quite close to their candidates on all five issues. Workers in both parties were closest to their candidates on the Vietnam issue, a sign of the overwhelming desire of everyone to end the war. Republicans perceived themselves as closest to Nixon on Vietnam, military spending, and police authority, and a bit further from him on busing and welfare payments. These latter issues were probably the ones on which Republi­ can activists had reservations about their candidate. After Vietnam the Democrats perceived themselves as closest to McGovern on police authority. As we might have predicted, the Democrats were farthest from McGovern on military spending, the issue on which there was confusion about his position. 72

TABLE 2.9: MEDIAN DISTANCES* OF ACTIVISTS FROM CANDIDATE ON ISSUES

Republicans Democrats from from from from Nixon McGovern Nixon McGovern

Busing 0.7 4.0 2.9 0.7

Military Spending 0.4 4.1 2.6 1.0

Vietnam Peace 0.3 3.3 3.9 0.1

Welfare Payments 0.9 3.6 2.9 0.7

Police Authority 0.5 2.9 3.6 0.5

* Respondents were asked to place themselves and the candidates on a 7-point scale of agreement/disagreement with issue positions. The distance measure is the absolute value of the difference between the respondent's self-placement and his placement of a candidate. Ranges from 0-6. 73

Republicans perceived themselves far from McGovern on all issues.

Nixon's being a moderate helped his standing with the Democrats; they did not oppose Nixon to the degree that Republicans opposed McGovern.

The reaction to McGovern reflects his representing the left of his party, presumably far from a moderate Republican position. These re­ sults speak well for the party organizations; county leaders believed themselves to be in close agreement with their national leaders.

More detail is provided by the use of an individual index of agree­ ment. The index is the mean of an activist's distances from a candidate on all five issues, and ranges from 0.0 to 6.0. The lower scores indicate high levels of agreement. The index was constructed to give each activist a score representing how close or far he perceived him- slef to be from his candidate. Again we can observe the high intra­ party consensus and the greater opposition to McGovern from Republicans.

These index scores will be the measure of agreement used in coming chapters.

Obligations to the Party

The extent to which activists regard the party as a legitimate authority is the focus of this last section. How much power should the top elites have over the cadre? Should the party hold public officials accountable? Should party service be an important criterion in nomina­ tions? Because these are people with prior party service we would ex­ pect more support of party power from them than from other political actors. But the localism and decentralization of the system makes one 74

TABLE 2.10: THE INDIVIDUAL INDEX OF AGREEMENT*

Republicans Democrats Republicans Democrats Index Score with Nixon with McGovern with McGovern with Nixon

0 - .99 54.4% (111) 50.1% (99) 7.4 % (15) 7.1% (14)

1 - 1.99 38.7 (79) 41.1 (81) 8.5 (17) 9.0 (18)

2 - 2.99 4.4 (9) 6.5 (13) 21.1 (43) 28.5 (56)

3 - 3.99 2.0 (4) 1.5 (3) 31.4 (64) 27.4 (54)

4 - 4.99 0.5 (1) 0.5 (1) 23.5 (48) 20.2 (40)

5 - 5.99 0 0 7.4 (15) 5.0 (10)

6 0 0 1.0 (2) 2.5 (5)

Total 100% (204) 100% (197) 100% (204) 100% (197)

* Respondents were asked to place themselves and the candidates on a seven-point scale of agreement/disagreement with issue positions. The index is the mean of an individual's distances from the candi­ date on five issues. It ranges from 0-6. 75

question just how much authority they are willing to place in party

leaders above the local level.

Obligation to the party is important to the organization as a

motive to continue efforts under adverse conditions. Democrats were

urged to fulfill their duty to the party even though they were dis­

pleased with McGovern as the nominee. The greater the feeling of obli­

gation to the party, the less effort would be required to launch a

presidential campaign after divisive primaries or conventions.

The problem with the concept "obligation to the party" is the meaning of the word "party." If local leaders interpret it to mean

their local organization then their feeling of obligation will be higher.

As the data indicate, many local leaders do interpret the term party to mean their own organization and only think of the national organization when asked to.

Obligation to party is measured by four attitudes:

1. Support of patronage 2. Belief that party service should be a criterion for nominations 3. Belief that public officials should be held accountable to the party 4. Belief that one should always follow party leaders even when one disagrees

On face value one would expect the county leaders to be most sup­ portive of patronage since this gives local organizations greater in­ fluence. After patronage, support of party service in nominations and holding public officials accountable would seem to be most acceptable.

However, the fourth item, the obligation to follow party leaders even when one disagrees, will probably receive the least support. It is this last item that does not allow local leaders to respond in terms of their

own organizations.

The result in Table 2.11 shows that the meaning of the term "party" does influence the degree of obligation activists feel. Overall the activists supported party power over nominations, public officials and patronage, and Republicans felt more obliged than Democrats. On the subjects of patronage and considering party service in nomination, both parties tended to support increased party influence although support was stronger among Republicans. Republicans also felt public officials should be accountable to the party, an issue which split the Democrats.

The last item, which taps the authority of higher levels in the party, received the least support from Republicans. 23.7% said they felt some obligation or a strong obligation to avoid following decisions of the party leaders when they disagreed. Democrats were even less inclined

(30.5%) to say they would follow party leaders.

In an effort to obtain a summary measure of obligation to the 23 party, a Guttman scale of the four items was created. The easiest item to pass was patronage. This, more than the other items, clearly en­ hanced the power of the local leaders. Emphasis on party service and accountability of public officials were second and third. They, too, contribute to local leaders' influence. The most difficult item to pass was the obligation to follow party leaders. It seems local leaders do feel a commitment to increasing the local party's influence, but, when asked about higher levels, their sense of obligation drops. 77

TABLE 2.11: OBLIGATION TO THE PARTY FELT BY THE 1972 COUNTY CAMPAIGN LEADERS

Republicans Democrats % N % N

Obligation to see to it that those who work for the party get help in the form of jobs and other things they need

Strong obligation to do 26.9 (54) 17.4 (34) Some obligation to do 32.3 (65) 28.7 (56) No obligation either way 29.8 (60) 34.4 (67) Some obligation to avoid 5.5 (11) 8.7 (17) Strong obligation to avoid 5.5 (11) 10.8 (21)

Total 100% (201) 100% (195)

Obligation to keep elected public officials strictly accountable to the party organization

Strong obligation to do 20.9 (42) 19.2 (37) Some obligation to do 33.3 (67) 28.0 (54) No obligation either way 29.4 (59) 22.2 (43) Some obligation to avoid 7.5 (15) 16.6 (32) Strong obligation to avoid 8.9 (18) 14.0 (27)

Total 100% (201) 100% (193)

Obligation to weigh prior service to the party very heavily in selecting candi­ dates for nomination

Strong obligation to do 28.2 (57) 17.4 (34) Some obligation to do 33.7 (68) 31.3 (61) No obligation either way 27.2 (55) 29.2 (57) Some obligation to avoid 4.5 (9) 9.8 (19) Strong obligation to avoid 6.4 (13) 12.3 (24)

Total 100% (202) 100% (195) 78

Table 2.11 (cont'd)

Republicans Democrats % N % N

Obligation to follow decisions of party leaders even when you disagree

Strong obligation to do 13.4 (27) 10.4 (20) Some obligation to do 38.1 (77) 30.6 (59) No obligation either way 24.8 (50) 28.5 (55) Some obligation to avoid 8.9 (18) 11.4 (22) Strong obligation to avoid 14.8 (30) 19.1 (37)

Total 100% (202) 100% (193) 79

Three criteria were used to establish the validity of the scale:

(1) the degree to which the items were ordered as expected, (2) the correlations among the items and, (3) the coefficient of reproducibility.

As mentioned previously, on face value the items seemed to be appro­ priate for Guttman scaling. The final item ordering on the scale sup­ ported the expectation. The order of the items was consistent with the expected ordering.

Turning to the second criterion, we find that all coefficients

(Yule's Q) are above .60 which Clausen says is the appropriate level to minimize the possibility of including items that are conceptually dif-

2 / ferent from the concept being measured (Table 2.12). On the other hand, Johan Galtung warns that correlations that are too high may mean too similar items are being used. The goal then is to find dissimilar 25 items tapping the same dimensions.

Finally, the coefficient of reproducibility was used to check the validity of the scale. The coefficient was .949 which is higher than the normal level of acceptability of .90. The coefficient of reproduci­ bility measures the ratio of successful reproductions to total responses.

Herbert Menzel has criticized this coefficient on the grounds that there is always a minimum number of successful reproductions stemming from extreme items and extreme individuals. He introduced an alternate, the coefficient of scalability, which, by computing ceilings on the number of reproduction errors possible, takes into account extreme items and extreme individuals. As a result it is lower than the coefficient of 26 reproducibility. While the coefficient of reproducibility is ques­ tionable, and therefore is not sufficient by itself to establish the 80

TABLE 2.12: CORRELATION MATRIX FOR OBLIGATION ITEMS (YULE'S Q)

Favor Party Patronage Service Accountability

Favor Patronage *

Emphasize Party Service .90

Hold Public Officials Accountable .70 .80

Always Follow Party Leaders .86 .80 .67

* These are obligations which the 1972 county campaign leaders felt toward their party. 81

TABLE 2.13 GUTTMAN SCALE OF OBLIGATIONS TO THE PARTY FELT BY THE 1972 COUNTY CAMPAIGN LEADERS*

Republicans Democrats % N % N

Four Passed 48.9 (85) 34.8 (57)

Three Passed 27.0 (47) 26.2 (43)

Two Passed 13.2 (23) 9.1 (15)

One Passed 4.0 (7) 9.8 (16)

None Passed 6.9 (12) 20.1 (33)

Total 100% (174) 100% (164)

* The order of the items, from easiest to most difficult: Favoring patronage Emphasizing party service in nominations Accountability of public officials to party Obligation to always follow party leaders Coefficient of reproducibility = .95. 82 validity of the scale; it is substantially higher than the normal accep­

tance level and is corroborating evidence for the other criteria dis­

cussed above.^

The distributions on the Guttman scale show more vividly the dif­

ferences between parties. Many more Republicans than Democrats passed all of the items and only 5.9% of the Republicans compared to 16.8% of the Democrats failed to pass any. More than anything else these data show that the kinds of obligations activists feel toward the party are those that expand their own sphere of influence.

Conclusions

The findings on motivations present a mixture of consequences for the party organizations. On the positive side, many activists are moved by a sense of party loyalty and by habit to participate in the campaign.

Also beneficial is the high interest shown in policy, indicating a sub­ stantive goal in their party activity. Although party disagreements over policy can be disruptive, concern with issues keeps the party con­ temporary. An absence of policy interest could be more destructive because the parties would become less relevant to an increasingly issue-oriented public.

Activists seem to be psychologically involved in the election and projected certain of their own ideas onto the candidates. Disagreements about where the candidates stood were in part a result of intra-party conflict on the issue itself. When it came to specific issues the local leaders displayed widely varying degrees of consensus. Disagreements occurred over different issues in each party and reflected the existing 83 party coalitions. In general, Democrats were more united on issues concerning social welfare and civil liberties, issues that are impor­ tant to the party's group support. Republican consensus was on issues representing conservatism, governmental spending and military spending.

In spite of the party splits over several issues, there was a clear per­ ception of agreement with the presidential nominee. The issue agreement among activists themselves also bodes well for the parties. Although a few more Democrats than Republicans were policy deviants, the majority of activists only deviated on a small number of issues. Considering the publicity given to squabbles in the Democratic party, the amount of consensus they show is a bit surprising.

Along with consensus with their candidate and among themselves, local leaders had an eye toward the interests and independence of their local organizations. They favored increased party influence in politi­ cal affairs, but were not so willing to vest authority in party leaders. FOOTNOTES

1. C. Richard Hofstetter, "The Amateur Politician: A Problem in Construct Validation," Midwest Journal of Political Science, Vol. 15, No. 1 (February, 1971), p. 48. James Q. Wilson, The Amateur Democrat (Chicago: University of Chicago Press, 1966), Chapter 1. Peter B. Clarke and James Q. Wilson, "Incentive Systems: A Theory of Organiza­ tion," Administrative Science Quarterly, VI (September, 1961), pp. 135- 137.

2. James Q. Wilson, Political Organizations (New York: Basic Books, Inc., 1973), p. 96.

3. Ibid., p. 106.

4. C. Richard Hofstetter, "The Amateur Politician: A Problem in Construct Validation," pp. 47-49.

5. James Q. Wilson. The Amateur Democrat, Chapter 1. James Q. Wilson. Political Organizations, pp. 110-111.

6. Samuel J. Eldersveld. Political Parties: A Behavioral Analysis (Chicago: Rand McNally, 1964), pp. 277-292.

7. Dennis S. Ippolito, "Political Perspectives of Suburban Party Leaders," Social Science Quarterly, Vol. 49 (March, 1969), pp. 800-815, reprinted in David W. Abbott and Edward T. Rogowsky. Political Parties: Leadership, Organization, Linkage (Chicago: Rand McNally, 1971), p. 71.

8. M. Margaret Conway and Frank B. Feigert, "Motivation, Incen­ tive Systems, and the Political Party Organization," American Politi­ cal Science Review, Vol. 62 (December, 1968), p. 1168.

9. Joseph A. Schlesinger, "Political Party Organization" in Handbook of Organizations, ed. by James G. March (Chicago: Rand McNally, 1965), pp. 767-769.

10. James Q. Wilson. Political Organizations, p. 107.

11. Ibid., p. 110.

84 85

12. Samuel J. Eldersveld, Political Parties: A Behavioral Analysis, p. 278.

13. Ibid.; p. 278.

14. Richard W. Boyd, "Popular Control of Public Policy: A Normal Vote Analysis of the 1968 Election," American Political Science Review 66 (June, 1972), pp. 429-449. David E. Repass, "Issue Salience and Party Choice," American Political Science Review 65 (June, 1971), pp pp. 389-400. Gerald Pomper, "From Confusion to Clarity: Issues and American Voters, 1956-1968," American Political Science Review 66 (June, 1972), pp. 415-428. Mark A. Schulman and Gerald M. Pomper, "Variability in Electoral Behavior: Longitudinal Perspectives from Causal Modeling," American Journal of Political Science, Vol. 14, No. 1 (February, 1975), pp. 1-18.

15. John Field and Ronald Anderson, "Ideology in the Public's Conceptualization of the 1964 Election," Public Opinion Quarterly (Fall, 1969), pp. 380-398.

16. Herbert McClosky, P. Hoffman and R. O'Hara, "Issue Conflict and Consensus Among Party Leaders and Followers," American Political Science Review (June, 1960), p. 426.

17. John Kessel, The Domestic Presidency (North Scituate, Mass: Duxbury Press, 1974), pp. 128-130.

18. Under certain circumstances this measure can take on negative values. When there is a bimodal distribution and the modes are several magnitudes apart, the numerator of the consensus measure could be negative.

19. McClosky et al., p. 426.

20. Herbert Weisberg and Jerrold Rusk , "Dimensions of Candidate Evaluation," American Political Science Review, Vol. 64 (December, 1970), pp. 117, 1184. Herbert Weisberg and Jerrold Rusk, "Perceptions of Presidential Candidates: Implications for Electoral Change," Midwest Journal of Political Science (August, 1972), pp. 400-403.

21. The issues are those from Table 2.7 where local leaders placed themselves on a seven-point scale of agreement/disagreement. The dis­ tance of one magnitude from the mode was chosen because it had been used successfully in the consensus measure taken from The Domestic Presidency and used earlier in this chapter. 86

22. Donald E. Stokes, "Spatial Models of Party Competition," in Elections and the Political Order ed. by Angus Campbell, Philip Converse, Warren Miller, and Donald Stokes (New York: John Wiley & Sonse, Inc.; 1966), pp. 165-170.

23. Johan Galtung lists ten requirements of indices or scales. One is that the index should give as much information as possible. The loss of information in any index is important. The Guttman scale is one index that satisfies this requirement. Knowing the scale score enables us to make inferences about the pattern of responses. See Johan Galtung, Theory and Methods of Social Research (New York: Columbia University Press, 1967), pp. 241, 265-270.

24. Aage R. Clausen, "A Comparative Analysis of Senate House Voting on Economic and Welfare Policy: 1953-1964," American Political Science Review, Vol. 64, No. 1 (March, 1970), pp. 139-140.

25. Johan Galtung, Theory and Methods of Social Research (New York: Columbia University Press, 1967), pp. 259-260.

26. Herbert Menzel, "A New Coefficient for Scalogram Analysis," Public Opinion Quarterly (Summer, 1973), pp. 269-280.

27. John P. Robinson, "Toward a More Appropriate Use of Guttman Scaling," Public Opinion Quarterly, Vol. 37, No. 2 (Summer, 1973), pp. 260-267. CHAPTER THREE COMMUNICATIONS AND CANDIDATE IMAGES

Introduction

This chapter focuses on communication between local leaders and the national party, and on the local leaders' images of their presidential candidates. These two variables are more directly related to the cam­ paign at hand, whereas the motivations, issue opinions, and obligations to the party were probably developed before the campaign. This tem­ poral sequence, however, would be more true for McGovern than Nixon who had been active for so long that his image was well-formed before

1972. I make this point now because later in this chapter hypotheses will be tested relating the variables Chapter Two to communication and candidate images.

Communications channels to local leaders are important from an organizational perspective. What sources do they rely on the most?

The least? What difference does the information source make? Commun­ ication in a party as in any organization can influence its ability to coordinate activity and maintain unity.

Attitudes toward the leadership are similarly important in any campaign. In a presidential campaign attention is focused on the nominee who is at least the temporary leader of the party. Many of the local workers were motivated to join the campaign by attraction to

87 88

McGovern or Nixon. What makes up that attraction? What particular

characteristics are most appealing? Or least appealing? If the

nominee is considered competent and qualified for the presidency, will

it be easier to obtain support from local leaders? Why do activists

think someone would vote for their candidate? Does the anticipation

of victory or defeat play a role in motivating activists?

Communications Channels

Communications has been considered a prominent factor in organ­

izational behavior. The effects of communications have been studied

in many ways: the semantic content of messages, the preservation of meaning, sources of systemtic error, accuracy, and the effects of feed­

back. Harold Guetzkow has integrated a substantial amount of liter­

ature on the effects of communications on organizations.'*' Three

points seem particularly applicable to political parties. The first

concerns the interference of external environment in organizational communications. The notion is that the greater the extent to which

the environment provides a basis for information, the greater the differences in perceptions within the organizations. But when the information originates within the organization, contradictions are reduced. In a campaign organization the amount of information from the environment is tremendous, encouraging contradictory perceptions among members. A local leader's reliance on information sources out­ side the party could well affect his ideas about what is happening as well as his behavior. 89

A second notion applicable to political parties concerns the

extent to which the recipient of information understands what he is being told. Individuals are more inclined to speculate themselves when they do not receive feedback from higher levels in the organiza­

tion. Eldersveld concluded from his examination of party communica­ tions at lower levels that local leaders had conflicting perceptions 2 of the state of the organization. However, these contradictions were not associated with the absence of communication with top elites.

Rather, the precinct leaders acted autonomously regardless of the frequency of communication with district leaders.

A third idea has to do with the individual’s use of counter­ biases to adjust for biases he anticipates in the information he re­ ceives. The local loyalties of activists would incline them to be well aware of local implications of some national campaign strategy.

One of the functions of local activists is to tailor the campaign to their constituents. Consequently, it is possible that those counter­ biases serve a useful function for the campaign.

Our primary concern in this section will be the internal vs. external sources of information on the assumption that reliance on internal sources leads to closer cooperation with party leaders.

Intra-party communication, in contrast to the mass media, provides opportunities for two-way communication. Without such opportuni­ ties, the activists will be more likely to behave independently 3 and come to their own conclusions. 90

Campaign communication in this study refers to information about

the national campaign only. The period of the presidential campaign

is a time when tremendous amounts of information about national politics

are being disseminated. The local leader receives information about

the national campaign from national headquarters and state politicians

through letters, phone calls, and meetings. He is also subject to the same media barrage as is the general public. Because a local leader's source of information may well color the picture he has of the na­ tional campaign we need to know just which sources he relies on most.

Table 3.1 provides detailed data on the information sources of county leaders. It seems that friends and newspapers are quite impor­ tant followed by directors' meetings. The importance of friends as sources of information confirms prior notions about the importance of interpersonal networks in political communications.

This list of sources includes both media and sources internal to the party organization. The party sources are directors' meetings, state and national headquarters and friends in politics, and TV and newspapers are the media sources. Although state and national head­ quarters are external to the county organization, this is a study of the nationwide party, so state and national party leaders will be categorized as being within the party organization of which the county leaders are a part.

To confirm this dichotomy, the relationships between all the sources were examined. The pattern of relationships in Table 3.2 suggests the existence of a party audience and a media audience. Meetings of 91

TABLE 3.1: LOCAL PARTY LEADERS' SOURCES OF INFORMATION ABOUT THE NATIONAL CAMPAIGN

"I'd like you to tell me whether you learn a great deal, a fair amount, or not too much in each way.'

Great Deal Fair Amount Not Much % % %

Telephone 18.7 30.7 50.6

Directors' Meetings 37.7 42.9 19.0

Polls 25.2 40.4 34.2

TV News 36.9 49.4 13.2

Newspapers 47.9 44.1 7.5

Headquarters Letters 29.4 42.4 27.7

Friends in State and National Politics 48.4 36.9 13.7

Nixon's Speeches 28.2 32.7 37.2

McGovern's Speeches 37.7 36.7 24.4 92

TABLE 3.2: BIVARIATE RELATIONSHIPS BETWEEN ALL SOURCES OF INFORMATION*

Gamma

TV - Newspapers .62

Headquarters Letters - Friends .45

Meetings - Headquarters Letters .43

Meetings - Friends .43

McGovern Speeches - Nixon Speeches .42

TV - McGovern Speeches .36

Telephone - Directors' Meetings .36

Polls - Newspapers .29

Telephone - Friends .26

Meetings - Nixon Speeches .26

Polls - Nixon Speeches .26

Headquarters Letters - Newspapers .08

Meetings - Newspapers .07

Friends - Newspapers .07

Headquarters Letter - TV .06

Telephone - TV .05

Polls - Headquarters Letters .05

Polls - Friends .05

Meetings - TV .03

Telephone - Polls .02

Friends - TV .02 * County campaign leaders were asked how much they learned from each source. 93

directors, headquarters letters and friends show much closer relation­

ships to each other than they do to the media variables. Similarly,

TV and newspapers produced the highest coefficient, a gamma of .62.

Neither TV nor newspapers were associated with any other source to nearly that degree. To see which of these two broad categories is more important, a second question, "What would you say is your most important source of information about the national campaign?" is perhaps a better point of departure.

The distribution of media, party, and other sources is given in

Table 3.3. One-third cited the party, and about 45% said the media was their most important source. One interesting response which could not be classified as party or media was "the people" or "interest groups." Intuitively we would not think of this as a source of infor­ mation about the national campaign. But some local leaders apparently see national politics through their constituents. At any rate, the communications channels divide nicely into party and media sources.

In using this as a dichotomous variable, it would be helpful to identify respondents who say that party is an important source while the media is not, or, who learn a great deal from the media and not much from party sources. By using the nine separate rankings of infor­ mation sources and defining a "party audience" and a "media audience" several different ways, 36% of the county leaders could fit into a purely media or purely party audience. So over 60% are actually rely­ ing on both sources. We cannot say that those who reply that their most important source is newspapers are not also receiving information from national headquarters or some other party source. But in 94

TABLE 3.3: MOST IMPORTANT SOURCE OF INFORMATION ABOUT THE NATIONAL CAMPAIGN FOR 1972 COUNTY CAMPAIGN LEADERS

Media Sources N %

Television 59 14.7

Newspapers 84 21.0

TV and Newspapers 23 5.7

Magazines and Radio 14 3.5 N % Total Media Sources 180 44.9

Party Sources

Meetings with state and regional directors 11 2.7

Letters from state or national headquarters 74 18.4

Friends in state or national politics 26 6.5

Local party 6 1.5

Political leaders/party in general 8 2.0

Other party sources 11 2.7 N % Total Party Sources 136 33.9

Other Sources

Media and Party 21 5.2

The People and Interest Groups 16 4.0

Contacts with Individuals 9 2.2

Polls 14 3.5

Other 18 4.5

Missing Data 7 1.8 N % Total Other Sources 78 19.4 95

response to the question about the most important source, almost all

chose either media or the party, and did not mention anything from

the other in response to probes.

It may be that the question about the respondents' most important

source taps his most reliable source. That is, an activist who regards

friends at state headquarters as his most trusted source will also say

that they are his most important source, while at the same time he

reads newspapers and watches the TV news. While this dichotomy has an

intrinsic meangfulness, it will be even more important if it has an

impact on attitudes and behavior related to the campaign.

A variety of background characteristics were investigated to see if

any type of activist was more likely to rely on the media or the party.

Few had any notable relation to the use of one source or the other. The

politically experienced and the inexperienced both relied on media and

party to the same degree. Those with lower incomes and especially those with less education did cite the media more often, but they only com­

prised about 10% of the sample. Activists from the East Coast also

relied on the media more, possibly a result of the two prestigious newspapers in that area. Although Democrats and Republicans both

cited media sources more frequently than party sources, the Democrats did so more often than Republicans. These results are consistent with other evidence that the Republican organization was more cohesive in

1972. From this brief analysis it seems that use of the media or party as the main information source depends more upon the activist's close­ ness to his party organization at a particular time than it does on 96

TABLE 3.4: BACKGROUND CHARACTERISTICS OF THE COUNTY CAMPAIGN LEADERS AND COMMUNICATIONS CHANNELS

Media* Party Total Party Experience Experienced in Party Politics 57.4% 42.6% 100% (124) (92) (216)

Inexperienced in Party Politics 55.6 44.4 100 (55) (44) (99) Gamma = -.04

Education High school graduates 71.0 29.0 100 (22) (9) (31)

Some college or a college degree 54.4 45.6 100 (148) (124) (272) Gamma = .37

Income Incomes below $6000 64.2 35.8 100 (18) (10) (28)

Incomes above $10,000 56.9 43.1 100 (141) (107) (248) Gamma = -.04

Region From the East Coast 64.2 35.8 100 (43) (24) (67)

From the Midwest 55.6 44.4 100 (60) (48) (108)

From the South 57.5 42.5 100 (50) (37) (87)

From the West 49.1 50.9 100 (26) (27) (53) Gamma = .09 97

Table 3.4 (cont'd)

Media Party Total Party Democrats 61.4 38.6 100 (94) (59) (153)

Republicans 52.5 47.5 100 (85) (77) (162) Gamma = -.18

* Media or Party refers to their most important source of information about the national campaign. 98

any prior inclination to use one or the other.

Communications Channels and Other Variables

It may be said that communications are both a cause and an effect.

Let us first consider the choice of, or trust in, a communications

channel as an effect of local leaders other attitudes. If, for example,

reliance on party sources indicates a closer relationship with the

party, we would expect those with a strong sense of obligation to the

party to be using party channels. Likewise, activists in closer agree­

ment with their candidate may well be paying more attention to the media.

The justification for this hypothesis lies in the nature of media com­

munications . It is a more direct communication from the candidate to

activists than are the party channels. Candidates' activities and

speeches are daily reported in the media while party communiques more

often concern vote-getting tactics and other practical matters. What

has been defined as party channels includes many state, regional, and

local sources of information, sources which involve considerable local

interpretations. Activists strongly attracted to McGovern or Nixon

either for issue or other reasons, should find the media a more satis­

factory source of information.

A similar reasoning leads us to predict that amateurs will also prefer the national media. They, too, will choose the communication

from the candidate which is more likely to be issue-oriented than party communications. 99

Still another variable affecting communications channels might be

choice of agreement with one's fellow activists. We would expect policy

deviants to find the media a more comfortable source of information

than the party with whom they are at odds.

Contrary to expectations neither obligations to the party nor

agreement with the candidate had the predicted relationship to reliance

on media or party channels. (Tables 3.5 and 3.6). There was a slight

curvilinear relation between agreement and communications, but the

small number of cases in the bottom category precludes any conclusions.

Amateurism was related to communications channels differently in

the two parties (Table 3.7). Republican amateurs relied heavily (70%)

on the media, but we must be wary of this conclusion since there were

only twelve amateur Republicans. Among the Democrats, it was the pro­

fessionals who preferred the media more than the amateurs, contradicting

the expectation that amateurs would be more media-oriented.

Agreement or disagreement with other activists does not have a

consistent relationship with one's reliance on the media or party channels

(Table 3.8). We might expect policy deviants to find the media a more comfortable source of information than the party with which they are at odds. This is true with Democratic county leaders; those who are out­ side the consensus on at least half the issues are more likely to rely on the media. But with the Republicans the pattern is reversed. The county leaders who disagree with their peers on a greater number of issues are more likely to rely on party sources. 100

TABLE 3.5: OBLIGATION TO PARTY AND COMMUNICATIONS CHANNELS

Republicans

Media * Party Total

Sense of Obligation to Party **

Low 0 45.5% 54.5% 100% (5) (6) (11)

1 40.0% 60.0% 100% (2) (3) (5)

2 63.2% 36.8% 100% (12) (7) (19)

3 50.0% 50.0% 100% (16) (16) (32)

High 4 50.7% 49.3% 100% (35) (34) (85) Gamma = .02

Democrats

Sense of Obligation to Party

Low 0 60.9% 39.1% 100% (14) (9) (23)

1 92.3% 7.7% 100% (12) (1) (13)

2 63.6% 36.4% 100% (7) (4) (11)

3 52.9% 47.1% 100% (18) (16) (34)

High 4 62.5% 37.5% 100% (30) (18) (48) Gamma = .08 101

Table 3.5 (cont'd)

* Media or Party refers to the county leaders' most important source of information about the national campaign.

** Sense of Obligation to Party is a Guttman scale of four items. 102

TABLE 3.6: AGREEMENT WITH CANDIDATE AND USE OF COMMUNICATIONS CHANNELS

Republicans and Nixon

Media Party Channels* Channels Total Individual Index of Agreement **

High 00-.99 57.4% 42.6% 100% (50) (37) (87)

1.00-1.99 46.0% 54.0% 100% (29) (34) (63)

Low 2.00 or greater 50.0% 50.0% 100% (6) (6) (12) Gamma = .17

Democrats and McGovern

Individual Index of Agreement

High 00-.99 67.9% 32.1% 100% (55) (26) (81)

1.00-1.99 53.3% 46.7 % 100% (32) (28) (60)

Low 2.00 or greater 58.3% 41.7% 100% (7) (5) (12) Gamma — .17

* Media Channels or Party Channels refers to the county leaders' most important source of information about the national campaign.

** Agreement with the Candidate is a mean of the distances from the candidate on five issues. 103

TABLE 3.7: AMATEURISM AND THE USE OF COMMUNICATIONS CHANNELS

Republicans

Media * Party Total

Professionals ** 56% 44% 100% (79) (62) (141)

Amateurs 70% 30% 100% (8) (4) (12) Gamma = -.30

Democrats

Professionals 67% 33% 100% (72) (35) (107)

Amateurs 57% 43% 100% (13) (10) (23) Gamma = .20

* Media or Party refers to the county leaders' most important source of information about the national campaign.

** Amateur or professional refers to the motivation to participate in the campaign. 104

TABLE 3.8: BEING WITHIN THE ISSUE CONSENSUS AND THE USE OF COMMUNICATIONS CHANNELS

Republicans

Media* Party Total Number of issues on which activist is outside his party's issue consensus **

0-3 58% 42% 100% (25) (18) (43)

4-7 51% 49% 100% (51) (49) (99)

8-11 45% 55% 100% (9) (11) (20)

12-15 none none Gamma = .11

Democrats

Number of issues on which activist is outside his party's issue consensus

0-3 62% 38% 100% (18) (11) (29)

4-7 59% 41% 100% (57) (39) (96)

8-11 70% 30% 100% (16) (7) (23)

12-15 too few cases to compute meaningful percentages Gamma = -.11______* Media or Party refers to county leaders' most important source of information about the national campaign. ** Being within the issue consensus means being within one magnitude of the mode of one's party's distribution on an issue. 105

Communications channels could also influence the images local

leaders had of their presidential candidate. Since the individual index

of agreement was constructed from perceptions of the candidates' issue positions, a member of the media audience in close agreement may be

that way due to media messages that are different from party messages.

To test this possibility the media and party audiences were compared as to where they thought the candidates stood on issues (Table 3.9).

There was no evidence of differences between the two audiences on most issues. The Republican party audience was more likely to strongly agree that Nixon would stop the use of busing, and the Republican media audience was more likely to see Nixon against higher welfare. McGovern was seen as more pro-busing by his media audience, but the party audience was more likely to say he would give the policy more authority.

None of these differences represents opposite perceptions. They are slightly different distributions in the same direction. We have not found any evidence that activists come to different conclusions about the candidates because of their use of one communications channel or the other.

In summary, choice of a communications channel was affected only by amateurism and had no bearing upon a local leader's perceptions of his candidate's issue positions. Activists receive information about the campaign from party and media sources. Furthermore, most reported that both types of sources provided them with a great deal of infor­ mation. Under these conditions perhaps activists don't "choose" a source of information; they are simply receiving it from a variety of 106

TABLE 3.9: COMMUNICATIONS CHANNELS AND PERCEPTIONS OF THE CANDIDATES' ISSUE STANDS

Republicans

Media Audience * Party Audience* %N % N

Nixon would stop the use of busing Disagree 9.7 (6) 16.2 (11) Not Sure 4.1 (3) 13.2 (9) Agree 86.3 (63) 70.6 (48) Total 100.0 (72) 100.0 (68) Gamma = -.06 Nixon would spend whatever necessary to have a strong military Disagree 7.9 (6) 8.4 (6) Not Sure 0 0 2.9 (2) Agree 92.1 (70) 88.7 (63) Total 100.0 (76) 100.0 (71) Gamma = .15 Nixon would bring peace in Vietnam Disagree 3.9 (3) 2.7 (2) Not Sure 0 4.1 (3) Agree 96.1 (74) 93.2 (68) Total 100.0 (77) 100.0 (73) Gamma = .20 Nixon would increase welfare payments Disagree 53.4 (40) 43.4 (30) Not Sure 16 (12) 18.9 (13) Agree 30.6 (23) 37.7 (26) Total 100.0 (75) 100.0 (69) Gamma = .15 Nixon would give the police more authority Disagree 16.7 (12) 16.9 (11) Not Sure 8.3 (6) 6.2 (4) Agree 75.0 (54) 76.9 (50) Total 100.0 (72) 100.0 (65) Gamma = .02 107

Table 3.9 (cont'd)

Democrats

Media Audience Party Audience % N % N

McGovern would stop the use of busing Disagree 85.4 (76) 75.9 (41) Not Sure 10.1 (9) 11.1 (6 ) Agree 4.5 13.0 _IZ1 Total 100.0 (89) 100.0 (54) Gamma = .16 McGovern would spend whatever necessary to have a strong military Disagree 32.7 (29) 25.5 (14) Not Sure 2.2 (2 ) 5.5 (3) Agree 65.1 (58) 69.1 (38) Total 100.0 (89) 100.0 (55) Gamma = -.03 McGovern would bring peace in Vietnam Disagree 2.2 (2 ) 1.9 (1) Not Sure 1.1 (1) 0 Agree 96.7 (89) 98.1 (52) Total 100.0 (92) 100.0 (53) Gamma = .19 McGovern would increase welfare payments Disagree 17.1 (15) 26.0 (14) Not Sure 13.6 (12) 9.2 (5) Agree 69.3 (61) 64.8 (35) Total 100.0 (88) 100.0 (54) Gamma = -.14 McGovern would give the police more authority Disagree 73.5 (64) (33) Not Sure 9.2 (8) (8) Agree 17.1 (15) (13) Total 100.0 (87) 100.0 (54) Gamma = * The Media Audience are the county leaders who relied on the media most for information, and the Party Audience relied on party s ources. 108 sources. This would help explain the lack of relationships to activists' attitudes. Furthermore, since a great deal of information is being received from both sources, neither will be unimportant in shaping activists' perceptions. This may be why it has been difficult to un­ cover differences of opinion associated with the party or the media.

Candidate Images

Image is an elusive and fragile goal which candidates spend millions pursuing. Much of the campaign machinery is geared to creating favorable candidate images. Yet with all the expertise and technology available, campaign history abounds with single events doing irreparable damage to candidate image, and there often is a deep gulf between the people's view of a candidate and the image he seeks to project. Barry

Goldwater was misunderstood as being too eager to use nuclear weapons.

Adlai Stevenson had to content with his image as an aloof intellectual.

Richard Nixon went to great lengths to project the "new Nixon," an image to replace one of coldness and deception. McGovern never over­ came an image of being indecisive.

The creation and maintenance of a favorable image influence a whole range of campaign activities, but particularly important is the effect on those working for a candidate's election. What activists think of their candidates has consequences for the campaign. If the nominee is perceived to be a loser, it will be difficult to spur party workers to greater efforts. If he is objectionable on a salient issue, 109

local leaders may be caught between loyalty to the party and to their

constituencies, inclining them to be less enthusiastic. Perception of

the nominee as a capable leader is also crucial; a candidate thought

to be incompetent will aggravate the problem of party unity.

From studies of elites we would anticipate local leaders' eva­ luations to be different from mass evaluations in terms of sophitica-

tion and discrimination. Local leaders could be expected to offer more detailed and lengthy comments about the candidates, and to make more evaluations on the basis of policy than the general public.

Candidate image was operationalized as the number of positive and negative comments made by the local leaders about Nixon and McGovern on a number of topics. This measure was used in The American Voter to assess the public's perceptions of Eisenhower and Stevenson in

1952 and 1956.^ To measure each activist's own image, the number of positive comments about Nixon or McGovern was counted. This ranged from 0 to 5.

Tables 3.10 and 3.11 present a summary of the elites' candidate images. These images do not directly reflect the activists' own likes and dislikes, but rather their ideas about Nixon or McGovern appealing to voters. Local leaders were asked, "Is there anything in particular about George McGovern/Richard Nixon that would lead people to vote for/ against him?" Fortunately for both party organizations, out of all four categories of comments, the largest number of comments were posi­ tive comments about one's own candidate, and the second largest number were negative ones about the opposition. Contrary to expectation, 110

TABLE 3.10: REPUBLICAN ASSESSMENT OF CANDIDATE IMAGES: AVERAGE NUMBER OF POSITIVE AND NEGATIVE COMMENTS ABOUT CANDIDATES BY TOPIC *

McGovern Nixon

Negative Positive Negative Positive

Party Representative .299 .122 .161 .068 Personal Attraction .710 .622 .676 .490 Leadership Qualities .647 .220 .107 .372 Experience and Ability .421 .151 .220 1.070 Government Management .058 .024 .147 .107 Economic Policy .073 .034 .142 .200 Civil Rights .009 .000 .034 .053 Social Welfare .102 .132 .034 .034 Other Domestic Policies .038 .027 .007 .043 Philosophy of Govern .313 .161 .127 .008 Foreign Policy .029 .027 .058 .774 Military Issues .112 .029 .009 .002 Vietnam .088 .161 .176 .313 Group Connections .029 .107 .122 .019 Other .063 .004 .009 .181

Total 2.991 1.821 2.029 3.734

* County campaign leaders were asked why people would vote for and against Nixon and McGovern. Ill

TABLE 3.11: DEMOCRATIC ASSESSMENT OF CANDIDATE IMAGES: AVERAGE NUMBER OF POSITIVE AND NEGATIVE COMMENTS ABOUT CANDIDATES BY TOPIC*

McGovern Nixon

Negative Positive Negative Positive

Party Representative .335 .116 .152 .126 Personal Attraction .492 1.172 .598 .314 Leadership Qualities .314 .289 .314 .228 Experience and Ability .187 .375 .289 .456 Government Management .035 .050 .395 .086 Economic Policy .060 .203 .192 .050 Civil Rights .040 .040 .111 .076 Social Welfare .071 .111 .137 .045 Philosophy of Government .390 .335 .147 .238 Other Domestic Policies .035 .045 .086 .025 Foreign Policy .015 .050 .040 .380 Military Issues .071 .035 .001 .020 Vietnam .091 .253 .309 .106 Group Connections .050 .233 .213 .111 Other .035 .167 .106 .071

Total 2.221 3.474 3.090 2.520

------— ..... 33------* County campaign leaders were asked why people would vote for and against Nixon and McGovern. 112

TABLE 3.12: THE GENERAL PUBLIC’S IMAGE OF THE CANDIDATES

Average Number of Likes and Dislikes About Candidates and Parties By Topic

McGovern Nixon Negative Positive Negative Positive

Party Representative .098 .105 .053 .033 Character Background .265 .060 .066 .093 Economic Policy .130 .130 .224 .158 Philosophy of Government .119 .026 .018 .021 Civil Rights .009 .007 .012 .010 Civil Liberties .000 .000 .000 .000 Foreign Policy .178 .140 .198 .348 Groups .007 .095 .106 .032 Experience and Ability .218 .152 .185 .660 Personal Attraction .435 .287 .248 .393 Other .030 .033 .022 .021 Total Candidate 1.487 1.036 1.132 1.769

Source: C. Richard Hofstetter, Director, Television and Election News Coverage Project, The Ohio State University 113

Republicans did not make more negative statements about McGovern than

Democrats did concerning Nixon. Both parties were equally negative

about the opposing candidate, mentioning an average of three reasons

why people would not vote for him. But the Democrats did acknowledge more reasons to vote for Nixon than Republicans did to vote for McGovern,

pointing to the not surprising conclusion that Democrats saw Nixon as a more formidable opponent.

Thus far we have only been discussing the direction of the comments.

Their content reveals more detail about the elites' images. Personal attraction is McGovern's most appealing quality according to his own party members. There was an average of over one comment per Democrat about his personal appeal. Republicans, too, acknowledge McGovern's attractive personal qualities. But personal qualities were also consi­ dered a major reason people would vote against McGovern. He was consi­ dered to have both attractive and unattractive personal traits. On the positive side were his honesty, kindness and high ideals. But the qualities working against him were his dishonesty and lack of common sense. There was obviously some disagreement as to whether McGovern was an honest man.

Turning away from personal qualities, let's consider the extent to which Democrats perceived McGovern to be qualified for the office of

President. These perceptions can be measured by looking at qualities such as experience, ability, and leadership. Democrats regarded their candidate as appealing for his experience, but oddly they thought Nixon was even more appealing for his experience. Similarly, on leadership, 114

Democrats viewed their own candidate as negatively as they did Nixon.

In fact his own workers saw McGovern lacking in leadership qualities, inclining us to expect commitment problems in the Democratic organiza­ tion. Still another indicator of the lack of enthusiasm for McGovern as a presidential candidate was the fact that he was unappealing as a representative of the Democratic Party.

On policy matters McGovern was seen as most attractive for his economic and Vietnam policies, and he was considered more appealing than Nixon for these policies. Nixon's only advantage was on foreign policy other than Vietnam such as dealing with Red China and Russia.

To sum up McGovern's image among his own party workers his best feature was his personal attraction. However, on leadership and compe­ tence dimensions he was viewed as no better than Nixon. His own party workers thought McGovern had an advantage over Nixon on policy matters, not ability, and on economic and Vietnam policies in particular.

Republicans were more positive about their candidate as a leader and President than were Democrats, even though Nixon was less personally attractive. Nixon's best feature was his experience and ability, a point with which Democrats agreed. On leadership qualities, too, he had an advantage. Republicans made many comments about the personal attraction of both candidates, but they gave McGovern the edge by making more positive comments about him. Nixon had an overwhelming advantage on foreign policy issues in his own party, and his Vietnam policies also worked in his favor. McGovern, on the other hand, was 115

not particularly attractive for any policy, foreign or domestic. Twice

as many remarks about police were made by Republicans than Democrats

and the difference consists of comments on Vietnam and foreign policy.

All in all, the data on candidate images confirms other findings

leading us to expect more organizational problems in the Democratic party. They have doubts about their candidate’s leadership capabili­

ties and are dissatisfied with him as a representative of the party.

If he were considered very appealing on policy, that might have compen­ sated for his other disadvantages. But even on domestic policy he had no clear advantage. Republicans are quite confident about the leadership abilities of their candidate, and while they are not parti­ cularly drawn to his domestic policies, they consider him to be quite competent in handling foreign affairs.

It was predicted that elites would make more comments than the public and this turned out to be the case; they had two to three times as much to say about the candidates. Since the elites were requested to estimate what was appealing to voters it would be interesting to see whose estimates, the Democrats' or Republicans', were more accurate.

Like elites, the public had most to say on the subject of personal attraction. It seems that the Democrats, who believed McGovern was so personally attractive, were in error. A more accurate assessment of how the public felt about McGovern was made by Republicans who, recognizing McGovern's negative features, did not overrate his positive ones."* The Republican estimate of experience and ability was 116

also more consistent with the public who gave a decided advantage to

Nixon. Nixon, however, had a net disadvantage in the public's eye on

his economic policies, as the Democrats predicted. But, contrary to

the Democratic assessment, McGovern had no advantage. In fact, the

Democratic candidate was not particularly appealing for any of his policies, but Nixon was viewed very favorably in foreign affairs, as both groups of elites anticipated.

Why did Republican party workers evaluate the popular images more accurately? Perhaps their own images of the candidates were closer to

the public's. Because McGovern represented a minority wing of his party, workers were probably brought into the campaign who shared his more leftist tendencies. Recall that an important motive for partici­ pation was attraction to McGovern. The comparison of perceptions shows an interplay between elites' own ideas and their estimates of the public images.

In present day politics, polls and other devices are employed to obtain an accurate measure of public opinions and attitudes. Democratic local leaders undoubtedly saw many polls, from state and national head­ quarters and from the national media. Yet they misjudged their candi­ date's popular image. The information provided by opinion polls was in competition with information provided from other sources, most notably friends and co-workers, who in the Democratic case would give cues contrary to the polls. In conversations with each other the more optimistic images of McGovern emerged. 117

On balance these misperceptions were probably an advantage for

the organization. For voluntary organizations like parties to maintain

enthusiasm and effort for the months of a campaign, the candidate needs

to be considered at least a possible winner. To the extent that a can­

didate is popular or clearly advantaged, his workers have no need to mis­

judge his popular image. But in the case of McGovern, an accurate

assessment of his unpopularity would have been very discouraging.

Even at the high levels of the Democratic campaign there was a

tendency to take poll results and interpret them so as to boost morale.

In September 1972, Gary Hart was desperately looking for some good news

to counteract the results of a Harris poll showing McGovern to be 28

percentage points behind. Pat Caddell, pollster for the Democrats, had been surveying continuously and his results showed a steady narrowing

of the margin and the possibility of a close race. This was publicized

to workers and supporters (Hart, 1972, p. 298). So selective perception

and information from friends and co-workers operated to minimize bad

news in the Democratic campaign. This may have had positive short­

term results, but it also prevented accurate estimates of popular atti­

tudes with serious consequences. The organization with the best esti­ mates of voter opinion has an advantage in knowing what strategies to employ, and what characteristics and issue positions to emphasize.

Although national campaign leaders make most of these kinds of decisions, the evidence indicates that selective perception was operating through­ out. So those strategic decisions that local leaders did make were probably affected. 118

Candidate Images and Other Variables

Why do some activists have a more positive image of their candidate than others? This section examines the effects of the other variables in this study upon the attractiveness of a candidate to his local workers. The local leaders' images will be measured by the number of positive comments he makes about his candidate and can range from zero to five.

The distribution of this measure of candidate appeal is shown in

Table 3.13. Many activists made the total of five comments about their own candidate, but few were able to be so complimentary toward the opposition. Consistent with our other data, the Republicans had more praise for Nixon than Democrats had for McGovern, and Democrats acknow­ ledged Nixon to have more appeal than the Republicans gave to McGovern.

Turning to the first effect on candidate images, we anticipate that the closer an activist is to his candidate on issues, the more attractive he will perceive that candidate to be. This, however, is not the case in either party (Table 3.14). Those with low distance scores made four and five positive comments in about the same propor­ tion as activists with greater distance scores. Perhaps the content of the comments we are counting is affecting this relationship. Most of the positive comments were not about policy; mentions of personal and leadership qualities and experience were much more common.

While agreement with the candidate has no impact on images, agree­ ment with one's fellow activists does influence images (Table 3.15).

The proportion of local leaders making four or five positive comments 119

TABLE 3.13: CANDIDATE IMAGES

Democrats Republicans Democrats Republicans and McGovern and Nixon and Nixon and McGovern % N %N % N % N Number of positive comments about the candidate * 0 1.5 (3) 1.5 (3) 16.2 (32) 24.5 (50)

1 6.1 (12) 3.4 (7) 14.2 (28) 21.6 (44)

2 13.2 (26) 10.8 (22) 23.4 (46) 20.6 (42)

3 22.8 (45) 14.7 (30) 24.4 (48) 20.1 (41)

4 32.5 (64) 32.8 (67) 14.7 (29) 7.4 (15)

5 23.9 (47) 36.8 (75) 7.1 (14) 5.9 (12)

100% (197) 100% (204) 100% (197) 100% (204)

* County campaign leaders were asked why people would vote for and against Nixon and McGovern. 120

TABLE 3.14: ISSUE AGREEMENT AND CANDIDATE IMAGES

The Republicans and Nixon

Individual Index of Agreement*

High Agreement Lower Agreement 00-.99 1.00-1.99 2.00 and greater % N % N % N Number of positive comments about Nixon ** 0 (0) Number of (3)

1 5 (5) 1 (1) cases too (1)

2 10 (11) 13 (10) small to (1)

3 16 (18) 14 (11) compute (1)

4 32 (36) 33 (26) meaningful (5)

5 37 (41) 39 (31) percentages.(3)

Total 100 (111) 100 (79) (14)

Gamma = .00

The Democrats and McGovern

00-.99 1.00-1.99 2.00 and greater % N % N % N Number of positive comments about McGovern 0 2 (2) (0) Number of (1)

1 6 (6) 2 (2) cases too (4)

2 11 (11) 17 (14) small to (1)

3 23 (23) 20 (16) compute (6)

4 30 (30) 36 (29) meaningful (5)

5 27 (27) 25 (20) percentages(0)

Total 100 (99) 100 (81) (17) Gamma = -.10 121

Table 3.14 (cont'd)

* The Index is a mean of the distances from the candidate on five issues. Ranges from 0-6.

** County campaign leaders were asked why people would vote for and against Nixon and McGovern. 122

TABLE 3.15: BEING WITHIN THE PARTY'S ISSUE CONSENSUS AND HAVING A POSITIVE IMAGE OF THE CANDIDATE

Republicans

Number of positive comments about Nixon** 2 - 3 4 - 5 Total Number of issues on % N % N % N which activist is outside his party's issue consensus * 0 - 3 28 (16) 72 (40) 100 (56)

4 - 7 25 (28) 75 (85) 100 (113)

8 - 1 1 32 (8) 68 (17) 100 (25)

12 - 15 no cases Gamma = -.07

Democrats

Number of positive comments about McGovern 2 - 3 4 - 5 Total % N % N % N

Number of issues on which activist is outside his party's issue consensus 0 - 3 34 (11) 66 (21) 100 (32)

4 - 7 35 (41) 65 (76) 100 (117)

8-11 59 (17) 41 (12) 100 (29)

12 - 15 too few cases to compute meaningful percentages Gamma = -.20

* Being within the issue consensus means being within one magnitude of the mode of one's party distribution on an issue. ** County campaign leaders were asked why people would vote for and against Nixon and McGovern. 123

about McGovern sharply declines once one reaches the group disagreeing

on half the issues. In the Republican party the pattern is much weaker.

Why does agreeing with other activists lead to a more positive

image of the candidate and agreeing with the candidate does not? One

explanation may lie in the distributions of the two agreement variables.

Agreement with the candidates is heavily skewed in the direction of

close agreement, whereas there seems to be less agreement with other

activists. Recalling that agreement with the candidate is perceived

agreement and party consensus is an objective measure, it is logical

that candidate agreement would be higher. But for the purpose of test­

ing relationships, candidate agreement is a poor independent variable because of its lack of variation.

Amateurism should also incline county leaders to have a positive

image of their candidate, an expectation that is supported in Table 3.16.

Amateurs were more likely to make four or five positive comments than were the professionals. However, only the Democratic party patterns support the relationships. If amateurs are motivated to participate in

the campaign for one particular issue or issue area it makes sense that they would choose the candidate in line with their position and be very attached to that candidate. The professional motives, on the other hand, are much less dependent on these characteristics of the nominee.

Finally, one's obligation to the party was expected to affect his image of the candidate. A strong commitment to increasing party in­ fluence should be associated with a strong identification with the party,

inclining one to be positively disposed toward the presidential 124

TABLE 3.16: AMATEURISM AND CANDIDATE IMAGES

Republicans

2 - 3 4 - 5 Total % N% N % N

Professionals* 29 (39) 71 (96) 100 (135)

Amateurs 9 (1) 91 (10) 100 (11)

Gamma = .01

Democrats

Number of positive comments about McGovern 2 - 3 4 - 5 Total % N % N % N

Professionals 40 (39) 60 (58) 100 (97)

Amateurs 26 (6) 74 (17) 100 (23)

Gamma = .28

* Amateur or Professional refers to the motivation to participate in the campaign.

** County campaign leaders were asked why people would vote for and against Nixon and McGovern. 125 candidate. However, likd issue agreement, a feeling of obligation to

the party had no bearing upon a local leader's candidate image (Table

3.17). Republicans who passed all the items on the obligation scale were no more likely to offer positive mentions about Nixon than those lower on the scale. Oddly, Democrats at the lower end of the obligation scale offered slightly more positive comments about McGovern. It appears that activists' assessment of their candidate's appeal is maintained independent of their commitment to the party. They are making a dis­ tinction between opinions about a temporary leader and loyalty to their own more permanent organization.

Earlier in this chapter communications channels were shown to be unrelated to perceptions of the candidates' issue positions. Neither does a reliance on party or media sources seem to affect the image an activist holds of his candidate (Table 3.18). Communications channels were thought to be an important variable because they could influence ideas about what was going on in the campaign. That influence is absent on at least two major points, ideas about the candidates and issues and ideas about candidate appeal. The results suggest that we may have been exagerating the importance of using different communica­ tions channels. Party activists are able to perceive and discriminate some important things about the campaign no matter how they get their information. 126

TABLE 3.17: OBLIGATION TO THE PARTY AND CANDIDATE IMAGES

Republicans and Nixon

Guttman Scale of Obligation to the Party (0-4)** 2 * 3 4 % N % N % N Number of positive comments about Nixon *** 2 or 3 30 (7) 28 (12) 28 (23)

4 or 5 70 (16) 72 (31) 72 (59)

Total 100 (23) 100 (43) 100 (82)

*Lower categories were eliminated because cases were clustered in the middle and upper ranges of both variables.

Democrats and McGovern

Guttman Scale of Obligation to the Party (0-4) 0 1 2 3 4 % N% N % N % N % N Number of positive comments about McGovern 2 or 3 28 (8) 47 (7) 33 (5) 41 (16) 41 (22)

4 or 5 72 (21) 53 (8) 67 (10) 59 (23) 59 (32)

Total 100 (29) 100 (15) 100 (15) 100 (39) 100 (54)

** Obligation to the Party is a Guttman scale of four items.

*** County campaign leaders were asked why people would vote for and against Nixon and McGovern. 127

TABLE 3.18: COMMUNICATIONS CHANNELS AND CANDIDATE IMAGE

Republicans and Nixon

Party Channels* Media Channels % N % N Number of positive comments about Nixon ** 2 or 3 29 (23) 23 (17)

4 or 5 71 (56) 77 (57)

Total 100 (79) 100 (74) Gamma = .08

Democrats and McGovern

Party Channels Media Channels % N % N Number of positive comments about McGovern 2 or 3 38 (32) 44 (25)

4 or 5 62 (52) 56 (32)

Total 100 (84) 100 (57) Gamma = -.04

* Party or Media refers to county leaders most important source of information about the national campaign.

** County campaign leaders were asked why people would vote for and against Nixon and McGovern. 128

Coneluslons

The findings about communications channels showed the parties to be very open organizations. The use of party or media sources failed

to influence ideas about the campaign. The flow of information from the external environment affects nearly every activist; two-thirds of them said they received a great deal of information from both party and media. We can presume that the intra-party communications incorporate much of what the media is saying. That is, friends talk to friends about what the newspapers are reporting. The two sources are inextri­ cably mixed to the point that the real source of a message is hard to identify. Organization theory concepts of internal and external sources are not very helpful in studying parties. That theory has not been used very often to study parties. Now it seems that parties may present a further departure from the types of organizations that this particular communications theory was designed to study.

Understandably, the two parties held different images of the can­ didates . They were quite positive about their own nominee and negative about the opposition. That in itself is good, but not surprising news for the parties. More significant information lies in the content of those images.

Republicans were confident of Nixon's ability as a leader, and they believed his foreign policy to be particularly appealing. Demo­ crats, too, acknowledged Nixon's leadership abilities and even ranked them above McGovern's. Although McGovern was perceived to be very personally attractive, he also had some undersireable personal qualities. 129

There was a substantial disagreement among Democrats about their candi­

date's honesty and sincerity. These findings confirm others in this

study about the dissension in Democratic ranks.

The Democrats were also less capable of assessing the popular

image of their candidate. They overrated McGovern's personal attrac­

tion and his appeal for certain policy positions. While this could have

undesireable consequences for campaign strategy, it helps maintain

morale among the activists; and as was mentioned before, this is parti­

cularly important in voluntary organizations. Because Nixon clearly was the advantaged candidate his workers had no need to misperceive

his popular image.

Relations between candidate images, communications channels and

the other variables were examined primarily because they represent the

sources of organizational unity or disunity derived from the theories and frameworks of Chapter One. The results were mixed, differing from party to party.

In neither party did the obligation scale have any bearing on the use of communications channels or the image of the candidates. Issue consensus and amateurism behaved differently among Republicans and Demo­ crats. Amateurism was associated with reliance on the media among Re­ publicans, but Democratic amateurism was related to citing the party as the important information source and having a more positive image of

McGovern. Being within their party's issue consensus also led Demo­ crats to a more positive image of their candidate. FOOTNOTES

1. Harold Guetzkow, "Communication in Organizations," in James G. March (ed.) Handbook of Organizations, pp. 534-573 (Chicago, Rand McNally & Co., 1965).

2. Samuel Eldersveld. Political Parties: A Behavioral Analysis (Chicago: Rand McNally, 1964), pp. 98-117.

3. Harold Guetzkow, "Communication in Organizations," in James G. March (ed.) Handbook of Organizations, pp. 534-573 (Chicago: Rand McNally & Co., 1965).

4. Angus Campbell, Philip Converse, Warren E. Miller, Donald Stokes. The American Voter (New York: John Wiley & Sons, 1964), pp. 24-25.

5. In a study of representation to the 1972 Conventions Jeane Kirkpatrick found the Democratic elite and rank and file on opposite sides of issues and the Republican elite held views that were more re­ presentative of the views of rank-and-file Democrats than were the views of the Democratic delegates. See Jeane Kirkpatrick, "Representation in the American National Conventions," British Journal of Political Science, Vol. 5, Part 3 (July 1975), pp. 286-293.

130 CHAPTER FOUR THE ISSUE AGENDAS: A SUMMARY MEASURE OF CAMPAIGN UNITY

Introduction

Throughout this study we have been examining variables that pre­

sumably incline an activist to be cooperative or uncooperative with the national campaign. Professionalism, issue agreement with the candidate and one's colleagues, and a feeling of obligation to the party all were expected to promote organizational cohesiveness, and their absence would

lead to problems in managing the party. So too, it was reasoned that relying on party information sources and having a positive image of the candidate would incline an activist to follow national guidelines.

This chapter introduces a measure of the extent to which local leaders follow the national campaign. It is, more specifically, the degree to which the issues emphasized by local leaders are the same as those emphasized by their presidential candidate. These two sets of emphases will be referred to as the local issue agenda and the national issue agenda.

By comparing the national and local issue agendas we are in effect asking how closely the local campaigns resemble the national campaign.

However, there are several reasons why we would expect the two campaigns not to resemble each other. First, the nature of a local leader's

131 132

constituency will influence the way the campaign is run. One function

of these local leaders is to adapt the national campaign to local needs.

An agricultural community will want to hear about farm policy, a south­ ern community about integration and so forth. In some sense the local

issue agenda is set by local needs. Second, the traditional view of

American parties as decentralized mitigates against a unified campaign.

The data on obligations demonstrated how reluctant local leaders were to compromise their power and interests. The description of parties as collections of state and local parties leads us to be skeptical about similarities across localities.

However, issues are becoming more nationalized. The new style of campaigning with its extensive use of mass communications diffuses issues throughout the country. Everyone who is informed knows about busing in Boston and pollution in Pittsburgh. Not only are issues nationalized through communications, but many issues are affecting large segments of the population directly. The war and the economy are the best examples. National issues will be important in the local leader's activities as well as in the strategies of the national party leaders.

The National Issue Agenda

The national issue agenda can be conceived of as a plan or a strategy to win votes. That is, Nixon and his advisors believed it would be to his advantage to emphasize foreign policy issues. Similarly the McGovern plan was not to emphasize one issue in particular, but to speak to several issues important to voters. 133

The two issue agendas are presented in Table 4.1. The data base

was a content analysis of the two candidates' speeches from the conven- 1 tions to election day. One variable in the analysis was the issue area

to which a sentence belonged, and this variable provided the basis for

the national issue agendas. The entries in the table are the propor­

tions of all statements about issues that were made concerning that one

issue area. For example, 36.6% of all issue statements made by Nixon were about international involvement. Use of the candidates' speeches

to measure the issue emphases of the national campaign can be supported

on several grounds. First, the candidate is the voice of the national campaign. What he says reflects a combination of his own views, the views of his advisors, and strategic decisions by both as to what would be advantageous. In other words, his speeches are what the pinnacle of the organization wants to communicate. Second, the candidates' speeches are a constant flow of communication. They reflect trends taken by the national campaign such as the increasing negativism of McGovern toward the end. Finally, although many speeches are geared toward regional, ethnic, or occupational audiences, with today's media coverage of cam­ paigns candidates are mindful of the national exposure of their speeches regardless of where they are campaigning. In sum, the speeches are communication from the elites of the organization, and they continue throughout the campaign to a national audience.

Although the ordering was different the top three issues are the same in both agendas, evidence of their significance to the public.

Both campaigns agreed on what these issues were, however they differed TABLE 4.1: NATIONAL ISSUE AGENDAS; 1972

Republican

Issue Area Proportion *

International Involvement 36.6% Economic Management 23.2% Social Benefits 23.0% Government 22.1% Civil Liberties 11.0% Agriculture 5.3% Natural Resources 1.0%

Democratic

Economic Management 27.1% Social Benefits 23.1% International Involvement 21.8% Natural Resources 12.4% Civil Liberties 11.4% Government 11.2% Agriculture 4.2%

* The proportion of the presidential candidate's issue statements that were made about a particular issue area. 135 in the weights they assigned to them. The dominance of international

involvement in the Nixon campaign fits with the strategy to present

Nixon as the President doing his job. It is also consistent with

Nixon's own wish to make his contribution in foreign affairs. The top two issue areas of the Democratic agenda are what we would expect of a

Democratic campaign, and the third reflects McGovern's criticism of

Nixon's handling of the Vietnam War.

But McGovern was hardly a typical Democratic candidate. Much of his material was not proposals of his own for the economy or in social welfare, but attacks on a government of special interest groups, vetoes of welfare programs and on Nixon himself. The conduct of the war was another favorite target for McGovern. He denounced it on basically moral grounds and said that expenditures on the war were depriving

Americans of needed social services. His speeches took an even more negative turn toward the end becoming more moralistic and attacking

Nixon on the full range of issues. Although this is not reflected in the national issue agenda measure, the thread that ties the top three issues together is an extremely critical approach to Nixon and his

Administration.

The position of the governmental power issue in the Republican agenda illustrates their strategy of presenting Nixon as President.

Also, over the last twenty years Republicans have been viewed more favorably than Democrats by the public in the area of government 2 management. 136

Civil liberties received equal emphasis, but the candidates spoke

to different specific issues. McGovern, to keep the support of blacks

in his coalition, stressed the equal rights aspect of civil liberties, such as busing. But law and order and police authority are also in­ cluded in this issue area, and these issues belonged to Nixon.

Agriculture, low on both agendas, is the only issue area with an obvious geographic influence. The candidates spoke to agricultural issues when they were in farming areas and ignored those issues else­ where.

The Local Issue Agendas

Local leaders were asked "We are interested in what issues you have been emphasizing so far this year in the campaign. Tell me whether you see these issues as being very important, important, somewhat important, or not very important in your campaign activities?" There are three major influences on the local agenda. First are the constituency factors described above. Second, the national issue agenda tells local leaders what issues are most important to their candidate. Local leaders will tend to follow the national campaign since, as we have seen, they agree with the candidate on issues and hold a very positive image of him. In the sense that the national issue agenda is a strategy for winning votes, local leaders will want to go along with it. Also, much of the litera­ ture and other campaign material that is used at the local level is designed at the national level. So through the activists' own desire to go along and through the mechanics of campaigning, the local agendas are influenced by the national agenda. Third, local leaders themselves 137

exercise some discretion over the issues in their campaigns. We would

guess that the greatest deviation from the national agenda would occur

where constituency needs influence the agenda and where the local leader

is himself at odds with the national party. But the prior descriptions

of these activists indicate few are in sharp disagreement with the tem­

porary party leader. Constituency influences are thus a more likely

source of deviation from the national agenda than are activists* own

preferences.

There is a tendency for activists to respond that nearly all the

issues are important in their campaigns (Table 4.2). Only four out of

eleven issues elicited a "not very important" response from more than

10% of the activists. Since these issues were chosen as the most pro­ minent issues of the day, this finding is not surprising. Vietnam

received by far the most emphasis in local campaigns. A related issue, bringing troops home was also high on both parties' agendas. Wage and price controls, too, ranked high in both parties. Among the less impor­

tant issues were farmers' incomes, police authority, and busing, issues which, as we will see later, become important as a result of consti­ tuency characteristics.

Major differences between the parties are on two issues reflecting their traditions. Democrats attributed more importance to social secu­ rity payments, and Republicans did the same to government spending. The civil liberties issues, busing and police authority, were also more sig­ nificant to Republicans, illustrating their use of the negative public reaction to busing and the increase in crime. Bringing home the troops, 138

TABLE 4.2: 1972 LOCAL ISSUE AGENDAS: ISSUES IN ORDER OF IMPORTANCE BY PARTY*

Republicans

Very Impor­ Somewhat Not Very Important tant Important Important Total % % % % %

Peace in Vietnam 78 17 3 2 100 Government Spending 47 36 13 4 100 Bringing troops home 44 22 16 17 100 Wage-Price Control 40 37 18 5 100 Pollution 39 32 22 7 100 Welfare Payments 38 30 24 8 100 National Defense 34 38 22 6 100 Busing 38 23 17 22 100 Social Security Payments 28 38 28 6 100 Police Authority 22 30 30 18 100 Farmers1 incomes 14 26 30 29 100

Democrats

Very Impor- Somewhat Not Very Important tant Important Important Total % % % % %

Peace in Vietnam 87 11 1 1 100 Bringing troops home 59 28 9 5 100 Wage-Price Control 50 34 13 3 100 Social Security Payments 42 36 16 5 100 Pollution 37 40 16 6 100 Welfare Payments 34 39 16 11 100 National Defense 42 19 18 21 100 Government Spending 35 27 22 16 100 Farmers' income 22 28 24 25 100 Busing 15 26 25 34 100 Police Authority 6 13 16 65 100

* The Local Issue Agendas are the responses of county campaign leaders as to how much emphasis was put on a particular issue in their campaign. 139

an extension of the Vietnam issue, was emphasized more by Democrats.

So the local agendas reflected both the national party traditions and their particular strategies in the 1972 campaign.

Constituency Influence on the Local Agenda

The local environment probably influences issue emphasis in local campaigns. Local leaders are under some pressure to fashion their campaigns to the needs in their constituencies. One such environmental characteristic is regionalism, the effects of which are presented in

Table 4.3. Regional forces were associated with variation in emphasis on only four issues, and all of these variations occurred in the Re­ publican Party. Farmers' incomes, an issue predictably affected by section of the country, was less significant in campaigns in the East.

Busing received more emphasis in the South and social security in the

West. Finally, wage and price controls were much less important in the West than in the rest of the country. Beyond these four issues there is little evidence of regional influence among Republicans.

The Democrats were remarkably consistent in their issue emphases across regions. Even the busing and agriculture issues failed to produce deviations.

Another relevant constituency characteristic is urbanization.

Only three issue emphases were related to urbanization. Republican leaders placed more emphasis on social security in the small cities 140

TABLE 4.3: REGIONAL INFLUENCES ON THE 1972 LOCAL ISSUE AGENDA

Republicans

East Midwest South West

Peace in Vietnam 3.6* 3.6 3.7 3.8 National Defense 2.8 2.8 2.8 2.9 Bringing troops home 2.7 3.0 2.9 2.7 Government Spending 3.2 3.4 3.2 3.1 Wage-Price Controls 2.9 3.3 3.1 2.0 Welfare Payments 3.0 2.9 3.0 3.1 Social Security 2.9 2.9 2.8 3.4 Police Authority 2.8 2.4 2.6 2.3 Busing 2.5 2.5 3.1 2.8 Pollution 3.0 3.0 3.1 3.0 Farmers' Incomes 1.9 2.3 2.4 2.3

Democrats

East Midwest South Wesl

Peace in Vietnam 3.8 3.8 3.8 3.9 National Defense 2.9 2.8 2.7 2.8 Bringing troops home 3.4 3.4 3.4 3.3 Government Spending 2.7 2.8 2.8 2.8 Wage-Price Controls 3.3 3.3 3.3 3.3 Welfare Payments 2.8 3.0 3.0 2.8 Social Security 3.1 3.2 3.1 3.1 Police Authority 1.7 1.6 1.6 1.4 Busing 2.2 2.2 2.3 2.1 Pollution 3.1 3.1 3.1 3.2 Farmers' Income 2.2 2.6 2.6 2.3

* Entries are the average (mean) level of importance in the local campaigns. The higher the average the more significant was the issue. The entries ranged from 1-4. 141

TABLE 4.4: URBAN-RURAL INFLUENCES ON THE 1972 LOCAL ISSUE AGENDA

Republicans Rural areas Cities over 100,000 10,000 and small 500,000 to 500,000 to 100,000 towns

Peace in Vietnam 3.7 * 3.8 3.6 3.7 National Defense 3.0 2.9 3.0 3.0 Bringing troops home 2.8 3.2 2.9 2.8 Government Spending 3.0 3.1 3.2 3.4 Wage-Price Controls 3.1 3.1 2.9 3.1 Welfare Payments 3.0 3.0 3.0 3.0 Social Security 2.5 3.0 2.7 3.0 Police Authority 2.5 2.4 2.3 2.7 Busing 2.9 2.7 2.8 2.7 Pollution 2.9 2.9 3.1 3.0 Farmers’ Incomes 1.8 2.1 2.1 2.5

Democrats Rural areas Cities over 100,000 10,000 and small 500,000 to 500,000 to 100,000 towns____

Peace in Vietnam 3.9 3.9 3.8 3.8 National Defense 2.7 2.7 3.0 2.7 Bringing troops home 3.5 3.3 3.3 3.4 Government Spending 2.6 2.8 2.9 2.8 age-Price Controls 3.4 . 3.5 3.3 3.2 Welfare Payments 3.0 3.1 2.8 3.0 Social Security 3.3 3.1 3.1 3.2 Police Authority 1.4 2.0 1.7 1.6 Busing 2.4 2.4 2.2 2.1 Pollution 3.2 3.0 2.9 3.2 Farmers' Incomes 2,4 2.6 2.2 2.6

* Entries are the average (mean) level of importance in the local campaigns. The higher the average the more significant was the issue. The entries ranged from 1-4. 142

and in rural, areas and small towns. The third issue emphasis related

to urbanization was the Democratic extra stress on police authority

in small cities. Again, beyond these three the issues show about the

same level of importance in all categories of the urban-rural variable.

Most of the issues appear to be national in scope. Only two

issues, social security and farmers’ income were related to both

constituency characteristics in both parties. And only seven issues

differed in emphasis by more than one-eighth of the total possible

variation (one-half of a point on a four point scale). These seven

were out of all 44 issues created by categorizing by party, region,

and urbanization. Overall the impact of these constituency charac­

teristics was small.

Measuring Deviation from the National Issue Agenda

Similarity between the local and national campaigns was measured

by the distance between the two issue agendas. Each activist had

his own local agenda, operationalized by his response to the questions

about the importance of eleven issues. However, all Democrats had

the same national agenda, as did all Republicans, operationalized as

.the proportion of statements concerning an issue in the candidate's

speeches. 143

First, the local agenda constructed of specific issues had to be matched to the general issue areas of the national agenda. They were matched topically as follows:

National: International Involvement

Local: Peace in Vietnam Bringing troops home National Defense

National: Social Benefits

Local: Social Security Payments Welfare Payments

National: Civil Liberties

Local: Police Authority Busing

National: Economic Management

Local: Wage and Price Controls Government Spending

National: Natural Resources

Local: Pollution

National: Agriculture

Local: Farmers' Incomes

The general areas include more specific issues than those listed, so the match is not perfect. The specific issues were a subset of issues mentioned by the candidates in an issue area. But each of the specific items reflects an important dimension within the more general issue area. 144

Each local leader's own agenda was composed of his scores on the

eleven specific items. Each score ranges from 1 to 4, 1 meaning not

very important and 4 meaning very important. An activist's local

agenda might look like this:

33244134334

All Democratic activists have the same national agenda composed

of weights assigned to the proportions in McGovern's speeches. These weights are approximations of the proportions created to make the national agenda more comparable to the local agendas. In the Demo­ cratic case:

Issue Area Proportion Weight

Economic Management 27.1% 3

Social Benefits 23.0 3

International Involvement 21.8 3

Natural Resources 12.4 2

Civil Liberties 11.4 2

Agriculture 4.2 1 and similarly with Republicans,

Issue Area Proportion Weight

International Involvement 36.3% 4

Economic Management 23.2 3

Social Benefits 23.0 3

Civil Liberties 11.0 2

Agriculture 5.3 1

Natural Resources 1.0 1 145

So if our hypothetical activist is a Democrat, his national agenda is

3 3 3 2 2 1

To obatin the summary measure of deviation from the national issue agenda,

the absolute differences between each specific issue area score and its

general issue area weight were summed. For example, social benefits, which received a 3 weighting in the Democratic agenda, was subtracted

from the local activists' score on social security and on welfare pay­ ments. The absolute value of the differences was added to the sums of

the differences in the other issue areas. For the hypothetical acti-

measure would be obtained as follows:

= Economic Management | 3 - 3 I 0

= and its two issues I3 - 3 I 0

= Social Benefits and I2 - 3 I 1

its two issues 14 " 3 I = 1

International Involvement 14 - 3 I = 1

and its three issues |l - 3 I = 2

I3 " 3 I = 0

Natural Resources and 14 - 2 I = 2

its one issue

Civil Liberties and its I3 - 2 I = 1

= two issues I3 - 2 I 1

= Agriculture and its |l - *1 3

one issue

Sum of the differences = 12 146

This local leader received a score of 12 on the summary measure of

deviation from the national issue agenda.

Figures 4.1 and 4.2 show the distribution of this distance measure

for both parties. Democrats and Republicans have similar distributions,

centering around a score of 9 or 10. Both parties also have the same

proportion (about 46%) clustered about the mode. However, the Demo­

crats have slightly lower distance scores overall.

The distributions show that local leaders are reflecting their national issue agendas, although imperfectly. A modal score of 9 indi­

cates a moderate amount of congruence. Such a score could have been obtained a number of ways. An activist could have been two points dis­ tant on four issues, one point distant on one issue and no points dis­ tant on the remaining six issues. To deviate very far from the national agenda on a few issues, the activist with a score of 9 would have to be right with the national agenda on the remaining issues. In using this measure as a dependent variable we will be looking at those acti­ vists who deviate from the national agenda more than their fellow party members and at those who are especially close. This division is oper­ ationalized as those more than one point above or below the mode. For the Republicans the low category includes those with scores of 8 or less, and the high category contains those with scores of 12 or greater. For

Democrats the low category includes scores of 7 or less, and the high category contains scores of 11 or greater. The higher scores indicate that a local leader deviated from the national agenda across a number of issues. Percentage of Local Leaders 20 10 0 IUE41 UMR ESR FDSAC RMTENTOA ISEAED: REPUBLICANS AGENDA: ISSUE NATIONAL THE FROM DISTANCE OF MEASURE SUMMARY 4.1 FIGURE 1 *Distance from the national issue agenda refers to the degree to which issues which to degree the to refers agenda issue national the from *Distance mhszdb tepeieta addt ee mhszd nlcl campaigns. local in emphasized were candidate presidential the by emphasized 2 91 1 1 1 14 13 12 11 10 9 8 3 5 6 5 4 7 ae 9 css 5 cases 65 cases 91 cases 37 5 6 7 8 19 18 17 16 15 7 Distance From the National Agenda* National the From Distance

147 Percentage of Local Leaders 20 15 10 5 0 IUE42 UMR ESR FDSAC RMTENTOA ISEAED: DEMOCRATS AGENDA: ISSUE NATIONAL THE FROM DISTANCE OF MEASURE SUMMARY 4.2 FIGURE 1 *Distance from the national issue agenda refers to the degree to which issues which to degree the to refers agenda issue national the from *Distance 2 mhszdb h peieta cniaewr mhszd nlcl campaigns. local in emphasized were candidate presidential the by emphasized 3 5 ae 8 css 5 cases 75 cases 86 cases 25 5 7 6 Distance from the National Agenda* National the from Distance 8 1 1 1 1 14 13 12 11 10 9 5 6 7 184 17 16 15

19 148 149

Organizational Influences

The independent variables (presented in detail in Chapters Two and

Three) presumably affect party behavior. As such we expect them to have an impact on the extent to which local leaders follow the national cam­ paign. The anticipated relationships can be expressed as five hypo­ theses :

1. Amateurs will be more likely to deviate from the national

party.

2. Activists in less agreement with their nominee on issues

will be more likely to deviate.

3. Activists in less agreement with the rest of their party

will be more likely to deviate.

4. Activists with a lower sense of obligation to the party

will be more likely to deviate.

5. Activists relying on the media sources will be more

likely to deviate.

6. Activists with a less positive image of their candidate's

appeal will be more likely to deviate.

Hypothesis 1: The literature indicates that a predominance of amateurs in a party organization provokes splintering over policy dif­ ferences. Amateur organizations have problems keeping their members motivated, while the more social motivations of the professionals incline them to be consistent and longer-lasting party workers. The gamma coefficients in Table 4.5 show this relationship to be true in both parties, but stronger in the Democratic organization. The lower 150

Republican coefficient could be due to the lower number of Republican

amateurs. There were nearly twice as many Democrats in the amateur

category. Since amateurs are working in the campaign for policy reasons,

they may well be deviating from the national agenda on the issue(s)

that motivated them. Someone for whom the war in Vietnam is the most

salient issue will probably reply that it is also very important in his campaign activities.

By using congruence between issue agendas as the measure of organi­ zational unity, we are "stacking the deck" in favor of independent var­ iables with an issue component. In this study the issue-related inde­ pendent variables are amateurism, agreement with the candidate, and agreement with fellow activists. Two of these show relationships with the dependent variable, in Table 4.5, amateur-professional (.34 and .56) and issue agreement with the candidate among Democrats (.24).

Hypothesis 2: It is anticipated that the further a local leader perceives himself to be from the candidate, the more his local agenda will deviate from the national. However, there is no such relationship in_the Republican organization, and the Democratic association is a weak one. Recall that very few local leaders disagreed with their can­ didate over a number of issues. Most perceived themselves to be in quite close agreement. Much of the variance in issue agreement was being less than one or between one and two points distant from the can­ didate. This lack of variance prevents relationships from appearing. 151

TABLE 4.5: RELATIONSHIPS BETWEEN ORGANIZATIONAL INFLUENCES AND DISTANCE FROM THE NATIONAL AGENDA IN THE 1972 CAMPAIGN ORGANIZATIONS

Republicans Democrats

Amateur-Professional .34* .56

Issue Agreement with Candidate -.05 .24

Issue Consensus Among Party Activists .04 -.08

Obligation to the Party -.09 .07

Communications Channels -.21 .11

Candidate Image -.15 -.11

* The gamma coefficients are based on trichotomized versions of distance from the national agenda, issue agreement, and candidate image. 152

Hypothesis 3: It was predicted that being in agreement with one's

party would lead to following the national agenda. But the data show no such relationship. Activists outside the issue consensus were as

likely to be in line with the national agenda as were activists in close agreement. Perhaps one reason these were not related is that issue agenda (emphasis) is measuring something quite different than issue agreement. One refers to how much attention and resources one devotes to the issue, and the other reflects one's position on it. It is worth noting, though, that disagreement with fellow activists does not lead one to depart from the national issue agenda.

Hypothesis 4: The scale of obligation to the party is the most direct measure of closeness to the party. As such it should be related to following the national campaign. However, there is no such rela­ tionship in either party. The weakness of the obligation scale is that it dealt mainly with expanding the role of the local party organization.

Only one question required that the local leader interpret the word,

"party," to mean the national organization.

Hypothesis 5: Reliance on the party, or trust in the party's in­ formation, is another indicator of closeness to the organization. It was expected to lead to congruence with the national agenda. However, this variable does not exhibit the predicted association among the

Democrats, and only weak association in the Republican Party (a positive coefficient means that use of the media is related to closer congruence with the national agenda, a negative coefficient means reliance on the party is associated with congruence). Since McGovern represented the 153

left wing of his party, it could be that Democrats relying on party

channels were receiving cues closer to traditional Democratic position

rather than those taken by McGovern and represented in the national

agenda. The communications variable has been particularly impotent

throughout this study. It had no bearing on where local leaders thought

the candidates stood on issues or on their images of the candidates.

Since using the media or the party made no difference in perceptions, it

makes sense that it had no association with following the national

campaign.

Hypothesis 6: Having a positive image of one's candidate should

be related to following the national issue agenda. But the gammas of

-.15 and -.11 although they are in the right direction (meaning positive

imagery is associated with congruence) are too small to provide very

strong evidence for this proposition. The image variable suffers the weakness of the agreement variable. Nearly all the activists have quite

positive images of their candidates. It is difficult to assign any sub­

stantive meaning to the difference between three and four positive

comments. Again, a lack of variance in the independent variable is keeping the measures of association low.

Conclusion

This chapter introduced a summary measure of campaign unity based on the issue emphases in the local and national campaigns. Acknowled­ ging that local conditions affect the conduct of campaigns, the effect of constituency characteristics on local issue emphases was examined.

Constituencies influences varied from party to party, with Republican 154

emphases more affected. Social security was emphasized to a greater

degree in the West and in small cities and rural areas. Farmers' incomes

were less important in the East than in all other regions, and busing

received more attention in the South. Among the Democrats only police

authority showed as much variation across constituencies, and it was more important in small cities. Most issues seemed to be national.

Their emphasis was not greatly affected by region or urbanization.

Influences within the party organizations were represented by the

six independent variables from Chapters Two and Three, most of which

showed little relation to campaign unity. Only amateurism was clearly related to deviation from the national party in both parties. Other relationships were only slight and true for one party.

Why were these relationships so slight or non-existent? First, in view of the lack of previous empirical tests of these presumed relation­ ships, our results should not be too discouraging. Second, we can look to both the dependent and independent variables. The summary measure, in addition to being experimental, had methodological and conceptual flaws. It was weak in that it measured only issue emphasis, not issue position, Issue positions are probably more central to the idea of a unified party, but we had no objective measure of the candidates' issue positions, only the preceptions of the local leaders. The summary measure was also methodologically weak in that it was constructed of approximations. Specific issues were equated with general issue areas, and the national agenda was given rough weights to make it compatible with the local variables. The final summary measure varied (for each 155

Issue) only from 1 to 4. However, the measure does do well in view of

the strong national effects shown in Figures 4.1 and 4.2.

In spite of this critique most of the blame for the lack of rela­

tionships does not lie with the summary measure. The independent var­

iables were stronger both conceptually and methodologically, Yet they, too, failed to show expected relationships with each other. For example, local leaders in close agreement with their candidate did not have a more positive image of him. And activists with a strong feeling of ob­ ligation to the party were no more likely to have a positive image of the candidate.

It seems that instead of relationships we have found a picture of national parties, a conclusion that will be expanded in the following chapter. The lack of variance was a major problem in looking for rela­ tionships among the independent variables. It was particularly acute for issue agreement and candidate image. There simply were not enough activists disagreeing with the candidate or having a mildly positive image of him. The skewed distributions undoubtedly affected relation­ ships with the dependent variable as well.

The strong national effects on the issue agenda were consistent with the high level of issue agreement with the candidate. Local leaders perceived themselves to be in very close agreement with their nominess, and they responded that what was important in the national campaign was also significant in their own campaigns. This similarity in issue emphasis is an interesting contrast to the issue by issue study of consensus. There were several issues in both parties on which there 156 was very little agreement among the activists. It was this issue by

issue analysis that showed the least intra-party consensus. What is

important for the parties as organizations is that the presence of

these divisive issues did not prevent local leaders from perceiving

themselves close to their candidates on issues nor did it prevent the local campaigns from resembling the national campaigns in issue emphases. FOOTNOTES

1. Under the direction of John H. Kessel at the Ohio State University, a systematic sample of sentences from Nixon's and McGovern's speeches from the Conventions to election day were content analyzed into the following categories: (1) Subject matter — this included the traditional substantive areas, international involvement, economic management, social benefits, civil liberties, agriculture, natural resources and government; (2) the time that the statement occurred — either at the conventions, between conventions, or week 1, 2, 3, etc. of the campaign; (3) type of action — regulatory, allocative or symbolic; (4) the group referred to in the statement — Republican Party, Democratic Party, Presidency, Congress, Judiciary, the bureaucracy and others; (5) the number of groups referred to, and (6) the valence of the statement — positive, neutral, or nega­ tive. This dissertation only utilized the statements that con­ cerned issues, i.e. those that were coded as one of the above seven issue areas. For Nixon this sample totaled 8,779 sentences and for McGovern the total was 14,656 sentences. For further details see John H. Kessel, "The Parameters of Presidential Politics," Social Science Quarterly, June 1974, pp. 8-24.

2. William H. Flanigan and Nancy H. Zingale, Political Behavior of the American Electorate (3rd edition), (Allyn and Bacon, Inc., 1975), p. 142.

157 CHAPTER FIVE DISCUSSION AND CONCLUSIONS

Introduction

This study was undertaken to examine attitudes and characteristics of local party activists which would affect their behavior in relation to the national party. The focus was upon what factors inclined these activists to be in step with the national campaign and what factors inclined them to diverge from the national party.

The literature on parties over and again proclaims the lack of unity and centralization of American parties. However, there are several ways in which any organization can be held together. There may be an authoritative hierarchy that can dispense rewards and punishments. Or cohesion could be the result of effective leadership.

A third possibility is that members of an organization have enough shared attitudes that they work well together without leadership.

Which of these patterns, or others, obtains in American parties?

Dimensions of Decentralization

When parties scholars refer to a lack of national parties, or their decentralization, they usually mean the absence of structure or an authoritative hierarchy. David Truman's essay on federalism and

158 159

the party system is illustrative of this concept of decentralization.

He describes the party as "a tremendous variety of elements capable of

showing remarkable separatism and autonomy."^ Federalism, according to

Truman, is a major cause of the localization of parties because the local parties exercise so much power over nominations. Since the lea­ dership of the presidential party does not control the structure which nominates the legislative candidates, fractionated parties can only be expected. Truman doubts there will be responsible parties until there

O is a centralization of sanctions over candidate selection. Federalism also contributes to the dispersion of party power by creating indepen­ dent clusters of offices, each attracting party competitors. Parties will gravitate to centers of political power and federalism, by defini­ tion, means many such centers. Truman foresees that any changes in this pattern might come about because of the rising importance of issues 3 and problems of an urbanized society.

Decentralization of the nominating procedure was the particular concern of Austin Ranney's study of parties.^ Almost all statutes regulating nominations are state laws, eg. primary laws and laws regulating the conduct of state and local conventions. Although there has been a movement toward more national control over delegate selec­ tion to the national convention, it remains to be determined whether the national party has any constitutional status over state legisla­ tures in party matters."* If not, the local and state control will be perpetuated. 160

Structure and leadership considerations were central to Samuel 6 Eldarsveld’s description of American parties. Whereas Truman and

Ranney looked at the organization from the vantage point of the larger

political system, one of Eldeisveld's concerns was the meaning of the

organization in the minds of the party leaders. He sought answers to

questions such as were party leaders aware of the structure they were working in? Were role perceptions congruent from one level to the next? His findings indicated conflicting and different perceptions about the organization from party workers in the same district. Also, using an "index of organizational awareness," 12-24% of the precinct 7 leaders were extremely ignorant of the state of party structure.

And finally, due to confusion over roles, precinct leaders seemed quite 8 free to define their own roles in accordance with their preferences.

His conclusions about party organization led him to adopt a term previously used by Lasswell and Kaplan, "stratarchy," meaning the en­ largement of the ruling group of an organization so that many people 9 are involved in group decision-making.

Frank Sorauf too dealt with the leadership dimension of party de- 10 centralization. Few people think of the national committee people when asked to identify party leaders. Yet, the national chairman and his staff comprise what Sorauf calls "the uncertain summit" of American parties. In no way does this committee or its chairman exercise control

1 1 over the state and local party organizations. Beyond their main function of planning the national convention, their role is very flexible. The National Committees have ongoing structures and 161 activities, but these organizations do not speak for the party's office holders. When the party is "in," the President is its spokesman.

When a party is "out," the leadership usually comes from Congress.

Hence few look to the National Committee.

The reasons for the weakness of the national party cited by Sorauf are structural. It has no base of mass membership, nor does it have its own cadre of workers. It depends upon financial levies on state parties, 12 and lacks the fulltime attention of its own members. So we are to interpret the weakness of the national party as the lack of the centra­ lization of power in the national committee or its staff.

Discipline and control from the top are what makes centralization of a large scale organization. This leadership dimension was also the concern of Gordon Black in a paper presented at the 1975 American

Political Science Association Annual Meeting. Again the notion of party unity is one of centralized control by a few elites. Specifically

Black says "a centralized party structure depends upon a homogeneity among the top political elites within the party; a steady flow of incen­ tives to maintain discipline and control within the party, especially incentives which can be used to control the nomination processes; and 13 a monopoly over the flow of those incentives by party elites." Black predicted a trend toward nationalization due to new campaign laws, more amateur activists, and the democratization of the parties. But what he calls nationalization is a system of hierarchical authority. 162

The Psychological Unity of American Parties

The evidence introduced in this study points to another dimension

of nationalization, and that is one of shared attitudes and beliefs, a

psychological unity. This evidence suggests a degree of integration

that we would not expect after having read most of the literature on

parties. The literature examined structure, sanctions, and control

whereas the results of this study indicate another kind of cohesion

that has existed in spite of the structural barriers. This unity is

in terms of shared attitudes toward policy, the symbolic leadership,

and the purpose of the organization.

The activists exhibited a strong sense of party loyalty. About

two-thirds had held previous party offices and 50% said party activism was a way of life for them. This party loyalty was buttressed by a widespread concern for policy. 80% felt that they should select a

candidate who was committed on issues, and an even larger percentage believed that they, too, should hold strong personal issue beliefs.

This mix of personal satisfaction and an interest in public issues is

healthy for the parties because it combines a sense of purpose and

achievement with personal gratifications.

Literature on activism cites the increase in policy-motivated activists as one of the symptoms of the weakening of party machinery.

But large numbers of these types were not found at the level of county

leader. Only 12% of Democrats and 6% of Republicans mentioned policy

first as their motive for activism. Instead, policy was a motive held along with party loyalty and habit. 163

Party unity was also enhanced by the tendency of activists to pro­ ject their own ideas onto the candidates. Perceptions about who would bring peace were the most extreme example of this projection. Democrats were unwilling to believe the accusations against McGovern, but Republi­ cans accepted such accusations quite readily. Neither party would acknow­ ledge the other candidate's ability to end the war. 23% of the Republi­ cans said McGovern would end the war, and only 11% of the Democrats said the same about Nixon. These party workers had invested considerable time and efforts toward the election of their candidate, and so they had a real stake in believing him to be the best. The projections are most interesting among Democrats. Here we might anticipate that party workers would be less able to see their candidate as they wanted to see him, simply because some of his proposals were widely publicized. It was true that the question of where McGovern stood on military spending and welfare provoked some confusion in his party. However, most of his workers perceived him to be on the "right" side. That is, about 60% agreed that McGovern would maintain a strong military and increase welfare payments. It would have been interesting to know Democratic workers' perceptions of McGovern before the campaign was underway.

Some of them probably favored another candidate for the nomination. If so, did their image of McGovern change over time? Because they accepted him as the party leader and worked in his behalf, they may have viewed him more favorably as the campaign went on.

Psychological involvement was also apparent in the degree to which activists thought they agreed with their candidate and disagreed with the opposing candidate. About 83% of Republicans were in close 164 agreement with Nixon while only 15% perceived themselves agreeing with

McGovern. Similarly, 91% of Democrats believed they agreed with

McGovern, but only 16% felt close to Nixon. A high level of perceived agreement is a strong motivation to work for a candidate's election. It also suggests that local leaders do have an affective tie to the national party leader, and that it is important to them to agree with him.

The significance of identification with a personality was noted by

Arthur Holcombe in 1950. He cited the power of personality as one means of unifying the party. Furthermore he said the pull of this personality might upset nice rational predictions about the behavior of voters (and 14 presumably activists) based on economic interests. More recently,

Joseph Schlesinger pointed to the need of a candidate to have support 15 without demands for specific benefits. The identification of county leaders found in this study reflects Holcombe's statements about person­ ality holding parties together. Together with Schlesinger, these findings show how this type of party unification is in the candidate's interest once he holds office. Followers attached to a personality are less likely to make specific demands on the office-holder.

In this research identification with national party leadership is also demonstrated by the very positive images local leaders held of their candidates. A majority of activists in both parties made four or more positive comments about their nominees. It is important to morale to perceive one's candidate as capable and popular. Both parties were quite confident of their own candidate's popularity, but Republicans attributed more leadership ability to Nixon than Democrats did to 165

McGovern. These positive images were held in spite of some contrary evidence from a mass survey. Selective perception seemed to be opera­

ting among Democrats who clearly misjudged their candidate's, popular

image. They thought he was more popular than was actually the case.

Against the psychological involvement of these local leaders in their candidate's election was working to create a sense of unity in the party.

Furthermore, the source of an activist's information about the cam­ paign had no bearing on these images. Those relying on the party or the media were equally as positive toward their candidate. The national media was not able to alter Democrats' image of McGovern even though that source was probably presenting a less optimistic picture. The

Democrats needed to feel they had a chance to win and contrary informa­ tion was ignored or explained away.

The fact that these interviews occurred during a presidential cam­ paign enhances these nationalizing effects. At other times there is no one leader for the out-party to identify with. The activities of the campaign itself make local leaders feel a part of a national organiza­ tion. With attention focused on the Presidential contest, it is natural that local leaders would identify with their candidate.

The positive attachment to the presidential candidate did not carry over to other party leaders. When asked about their obligations to the party and its leadership most supported party power over nominations, public officials, and patronage, with Republicans advocating a stronger party role than Democrats. But neither party felt a strong obligation to follow party leaders when they disagreed; 51% of Republicans and 41% of the Democrats felt no such obligation. Probably the phrase "party 166

leaders" is interpreted to mean people like state chairmen or, perhaps,

regional coordinators. These are the party leaders who would likely

issue directives and make suggestions to the county level. Recall that

in examining communications patterns we found that state and national

headquarters were the second largest source of campaign information for

local activists. The presidential nominee is far enough removed from

county politics that they can feel a strong identification with him.

And best of all, he doesn't interfere.

Still another indication of campaign unity was the congruence between the local and national issue agendas. The county leaders generally agreed with their presidential candidates on issues. Only a few described their local campaign as very different from the national one in terms of issues. The Republican campaign emphasized heavily the questions of international involvement, while the Democrats spread their emphasis fairly equally between economic management, social bene­ fits, and international involvement.

Previous literature on intra-party consensus leads us to expect 16 local leaders to agree on issues. This expectation is modified by our findings on policy consensus. One must consider the issue when asking how much consensus exists. The measure of consensus on indivi­ dual issues show very little agreement on several issues, and the parties disagree on different issues. Republicans are split on the social welfare issues, women's rights, and abortion, and Democrats are divided on military issues and farmers' income. Neither party could agree on wage and price controls or foreign aid. 167

However, consensus was such that the average activist agreed with his fellow party members enough to feel at home in the party. 87% of the Republicans agreed with fellow workers on two-thirds of the issues, and Democratic consensus was only slightly lower. Another way of putting it is that any one party worker only contributed to the dissensus on a few issues. The parties were divided on some policies, but the division cut across issues in a variety of ways. These types of cleavages miti­ gate against party factions based on a number of policy disagreements.

David Truman predicted that party centralization, in the structural sense, would come about because of the issues of an urbanized society.

The issues examined in this study are such national issues. It was on these issues that the activists wanted to be in agreement with their candidate, and to perceive him on the correct side. It was also these issues on which there was agreement as to their relative emphasis in the campaign. The trend toward the nationalization of issues may, as

Truman foresees, enhance the probability of structurally centralized parties. But it contributes today to the psychological nationalization of the parties.

To conclude, previous findings about the decentralization and localism of American parties have been modified by this study. Much of the modification arises from an alternate definition of centralization or nationalization as psychological instead of structural. The decen­ tralized party structure and dispersed authority patterns belie a strong sense of national identification at the local level. And this identification is not only with some entity called "the party." We 168

already know that these activists have strong partisan attachments.

Rather they felt it important to be in step with a specific national

leader, and they wanted to regard this leader as capable and popular.

American parties become more nationalized when the contest is

national. In explaining the localized parties, scholars usually cite

aspects of the political system such as federalism and the fact that most elections are local ones. But another side of that argument is

also true. The one time that there is a national contest, local party workers exhibit a strong attachment' to the national level. A consider­ able number of them had had prior experience in local politics so they were subject to localizing influences. But during the presidential cam­ paign they focused on that contest and wanted to follow the national leadership.

The findings suggesting unity and nationalization are more impres­ sive than those suggesting a lack of it. It is not a structural unity or one based on a system of authority. Rather it is psychological unity buttressed by strong attachments to party symbols such as presidential candidates. Yes, there are real policy disagreements within each party.

These activists are educated, geographically dispersed, and politically aware. On a wide range of important issues, dissensus should be expec­ ted. But during the presidential campaign local leaders seem to regard themselves as members of a national organization and exhibit strong effective ties to the national leadership. 169

Comments on Future Research

For further research in this area I would suggest a similar study

with the sample changed from county campaign leaders to the regular

party county chairmen whether they were actively participating in the

presidential campaign or not. The findings of psychological nationali­

zation were probably enhanced by the particular type of activist used

for this study. Although many party regulars were in this population

it would be instructive to know about those regular party people who were not. By including only party regulars one would obtain some dis­

illusioned or angry activists who were "sitting out" the campaign, par­

ticularly if an extreme left or right wing candidate were nominated.

These are some questions to be asked about the party regulars. Do they show the same tendency to project their ideas onto the candidates?

Do they agree as closely as campaign leaders? What are the differences in the two groups' images of their candidate? The party regular approach would be more consistent with most studies on party activists and therefore more comparable to such studies.

Another area for future research involves the refinement and expansion of a measure of campaign unity. One change I would suggest is the use of congruence of Issue positions rather than, or in addition to, issue emphases. That is, are the issue positions taken by the candidates the same as those projected in local campaigns? A problem with this would be discerning the presidential candidates' positions on a wide range of issues. A second approach to campaign unity would be to expand the operationalization beyond just issues. This might 170 involve the study of state and regional coordinators as well as the county leaders, and possibly the use of sociometric methods. The goal would be to determine who gives directives to whom, to what extent are these directives carried out, from whom and how often do county leaders receive instruction, and how do they respond. One drawback to this method would probably be the loss of a measure of issue or ideological unity since most intra-party communications are concerned with campaign tactics such as getting out the vote, precinct organization and the like. This points again to the problem of what is meant by unity or nationalization. As in other research, conclusions will depend on the researcher's own definition. FOOTNOTES

1. David B. Truman, "Federalism and the Party System," in Arthur W. MacMahon, Federalism, Mature and Emergent (Doubleday & Co., N.Y.) 1955, pp. 118.

2. Ibid., p. 122.

3. Ibid., p. 134.

4. Austin Ranney and Willmoore Kendall, Democracy and the American Party System (New York: Harcourt, Brace and World, Inc., 1956), Chapters 10-12. Austin Ranney, The Governing of Men (New York: Henry Hold and Co., 1958), pp. 269, 312-327. Austin Ranney, "Changing the Rules of the Nominating Game," in James D. Barber (ed.) Choosing the President (Englewood Cliffs, New Jersey: Prentice-Hall, 1974), pp. 71-93.

5. The recent Supreme Court Case, Buckley v. Valeo (1976), in which the Court ruled that the Federal Election Commission was impro­ perly constituted and suggested that Congress re-write the 1974 campaign finance law probably represented a step toward federal jurisdiction over parties.

6. Samuel J. Eldersveld, Political Parties: A Behavioral Analysis (Chicago: Rand McNally, 1964), Chapter Five.

7. Ibid., p. 104.

8. Ibid., pp. 111-113.

9. Ibid., p. 99.

10. Frank Sorauf, Party Politics in America, 2nd edition (Boston: Little-Brown, 1972), Chapter Five.

11. Ibid., p. 113.

12. Ibid., p. 130.

171 172

13. Gordon Black, "A Rational Choice Analysis of Alternative Party Scenarios" presented at the American Political Science Association Annual Meeting, 1975, in San Francisco.

14. Arthur Holcombe, Our More Perfect Union (Cambridge: Harvard University Press, 1950), pp. 91-92.

15. Joseph A. Schlesinger, "The Primary Goals of Political Parties: A Clarification of Positive Theory," American Political Science Review, Vol. LXIX (September, 1975), pp. 848-849.

16. Herbert McClosky, P. Hoffman, and R. O'Hara, "Issue Conflict and Consensus Among Party Leaders and Followers," American Political Science Review, June, 1960. LIST OF REFERENCES

Abbott and Rogowsky, (eds.) Political Parties; Leadership, Organiza­ tion, Linkage (Chicago: Rand-McNally, 1971).

Agranoff, Robert. The New Style in Election Campaigns (Boston: Holbrook Press, 1972).

Anderson, Lee F., "Organization Theory and State and Local Parties," in William J. Crotty (ed.) Approaches to the Study of Party Organization (Boston: Allyn & Bacon, 1968), pp. 375-403.

Black, Gordon S., "A Rational Choice Analysis of Alternative Party Scenarios," presented at American Political Science Association Annual Meeting, San Francisco, California, 1975.

Bone, Hugh A. American Politics and the Party System, 4th edition (New York: McGraw-Hill, 1971).

Bowman, Lewis and G.G. Boynton, "Recruitment Patterns Among Local Party Officials: A Model and Some Preliminary Findings in Selected Locales," American Political Science Review, Vol. 60, (June, 1966), pp. 667-676.

Boyd, Richard, "Popular Control of Public Policy: A Normal Vote Analysis of the 1968 Election," American Political Science Review Vol. 66 (June, 1972), pp. 429-449.

Clarke, Peter B. and James Q. Wilson, "Incentive Systems: A Theory of Organization," Administrative Science Quarterly, VI, (Septem­ ber, 1961), pp. 129-166.

Clausen, Aage R. and Richard Cheney, "A Comparative Analysis of Senate House Voting on Economic and Welfare Policy: 1953-1964," American Political Science Review, Vol. 64, No. 1 (March, 1970), pp. 138- 152.

Conway, M. Margaret and Frank B. Feigert, "Motivation, Incentive Systems and the Political Party Organization," American Political Science Review, Vol. 62 (December 1968), pp. 1159-1173.

173 174

Cotter, Cornelius P. and Bernard C. Hennessy Politics Without Power (New York: Atherton, 1964).

Crotty, William J., (ed.) Approaches to the Study of Party Organiza­ tion , (Boston: Allyn and Bacon, 1963).

Dahl, Robert, (ed.) Political Oppositions in Western Democracies (New Haven, Conn.: Yale University Press, 1966).

Eldersveld, Samuel Political Parties: A Behavioral Analysis (Chicago: Rand-McNally, 1964).

Etzioni, Amitai Modern Organizations (Englewood Cliffs, N.J.: Prentice- Hall, Inc., 1964).

Field, John and Ronald Anderson, "Ideology and the Public's Conceptual­ ization of the 1964 Election," Public Opinion Quarterly (Fall, 1969), pp. 380-398.

Flanigan, William H. and Nancy H. Zingale Political Behavior of the American Electorate, 3rd edition (Boston: Allyn and Bacon, Inc., 1975).

Galtung, Johan Theory and Methods of Social Research (New York: Colum­ bia University Press, 1967).

Guetzkow, Harold, "Communication in Organizations," in James G. March (ed.) Handbook of Organizations, pp. 534-573 (Chicago: Rand- McNally & Co., 1965).

Haas, J. Eugene and Thomas E. Drabek Complex Organizations: A Sociological Perspective (New York: The Macmillan Company, 1973)

Hart, Gary W. Right From the Start (New York: Quadrangle, 1973).

Hirschfield, Robert S., Bert E. Swanson, and Blanche D. Blank, "A Profile of Political Activists in Manhattan," Western Political Quarterly, Vol. 15, 1962, pp. 489-507.

Hofstetter, C. Richard, "The Amateur Politician: A Problem in Construct Validation," Midwest Journal of Political Science, Vol. 15 (February, 1971), pp. 34-50.

, "Organizational Activists: The Bases of Participation in Amateur and Professional Groups," American Politics Quarterly, Vol. 1, No. 2 (April, 1973), pp. 244-276. 175

Holcombe, Arthur Our More Perfect Union (Cambridge: Harvard University Press, 1950).

Ippolto, Dennis S., "Political Perspectives of Suburban Party Leaders," Social Science Quarterly, Vol. 49 (March, 1969), pp. 800-815.

Kaufman, Herbert Administrative Feedback (Washington: Brookings Insti­ tution, 1973).

Kessel, John H. The Domestic Presidency (North Scituate, Mass.: Duxbury Press, 1974).

, "Strategy for November," in Choosing the President by James Barber (Englewood Cliffs, N.J.: Prentice-Hall, 1974).

Kirkpatrick, Jeane, "Representation in the American National Conven­ tions: The Case of 1972," British Journal of Political Science, Vol. 5, Part 3 (July, 1975), pp. 265-322.

Kirkpatrick, Samuel A., "Political Attitude Structure and Component Change," Political Opinion Quarterly (Fall, 1970), pp. 403-407.

Klapper, Joseph T. The Effects of Mass Communications (New York: Free Press, 1963).

Lamb, Karl A. and Paul A. Smith Campaign Decision Making (Belmont, California: Wadsworth, 1968).

Marvick, Duaine, "The Middlemen of Politics," in William J. Crotty (ed.) Approaches to the Study of Party Organizations (Boston: Allyn and Bacon, 1968), pp. 341-374.

May, Ernest and Janet Fraser (eds.) Campaign *72, The Managers Speak (Cambridge: Harvard University Press, 1973).

McClosky, Herbert, "Consensus and Ideology in American Politics," American Political Science Review (June, 1964), pp. 361-382.

____, P. Hoffman and R. O'Hara, "Issue Conflict and Consensus Among Party Leaders and Followers," American Political Science Review (June, 1960), pp. 406-427.

McCombs, Maxwell E. and Donald L. Shaw, "The Agenda-Setting Function of Mass Media," Public Opinion Quarterly, Vol. 36, (Summer, 1972), pp. 176-187. 176

Menzel, Herbert, "A New Coefficient for Scalogram Analysis," Public Opinion Quarterly (Summer, 1953), pp. 269-280.

Miller, Arthur, Warren Miller, Alden Raine, and Thad Brown, "A Majority Party in Disarray: Policy Polarization in the 1972 Election," American Political Science Association Annual Meeting in New Orleans, September 4-8, 1973.

Nexon, David, "Asymmetry in the Political System," American Political Science Review, LXV (September, 1971), pp. 716-730.

Nimmo, Dan The Political Persuaders (Englewood Cliffs: Prentice-Hall, 1970).

Ogden, Daniel M., Jr. and Arthur L. Peterson Electing the President, revised edition (San Francisco: Chandler, 1968).

Patterson, Samuel C., "Characteristics of Party Leaders," Western Politics Quarterly, Vol. 16, No. 2 (June, 1973), pp. 332-352.

Pomper, Gerald, "From Confusion to Clarity: Issues and American Voters 1956-1968," American Political Science Review, Vol. 66 (June, 1972), pp. 415-428.

Ranney, Austin, "Changing the Rules of the Nominating Game," in James D. Barber Choosing the President (Englewood Cliffs: Prentice- Hall, 1974), pp. 71-93.

The Governing of Men (New York: Henry Holt & Co., 1958), pp. 269, 312-327.

and Willmoore Kendall Democracy and the American Party System (New York: Harcourt, Brace & World, Inc., 1956), Chapters 10-12.

Repass, David E., "Issue Salience and Party," American Political Science Review, Vol. 65 (June, 1971), pp. 389-400.

Robinson, John P., "Toward a More Appropriate Use of Guttman Scaling," Public Opinion Quarterly, Vol. 37, No. 2 (Summer, 1973), pp. 260- 267.

Rosenbloom, David Lee The Election Men: Professional Campaign Managers and American Democracy (New York: Quadrangle, 1973).

Salisbury, Robert H., "The Urban Party Organization Member," Public Opinion Quarterly, Vol. 29 (Winter, 1965-66), pp. 550-564. 177

Schlesinger, Joseph A., "Political Party Organization," in Handbook of Organizations by James G. March (ed.) (Chicago: Rand McNally, 1965) pp. 764-801.

, "The Primary Goals of Political Parties: A Clarification of Positive Theory," American Political Science Review, Vol. 69 (September, 1975), pp. 840-849.

Schulman, Mark A. and Gerald M. Pomper, "Variability in Electoral Behavior: Longitudinal Perspectives from Causal Modeling," American Journal of Political Science, Vol. 14, No. 1 (February, 1975), pp. 1-18.

Selznick, Philip TVA and the Grass Roots (New York: Harper & Row, 1966).

Sorauf, Frank Party Politics in America, 2nd edition (Boston: Little- Brown, 1972).

Soule, John W. and James W. Clarke, "Amatuers and Professionals: A Study of Delegates to the 1968 Democratic National Convention," American Political Science Review, LXIV (September, 1970), pp. 888- 898.

Stokes, Donald E., "Spatial Models of Party Competition," in Elections and the Political Order ed. by Angus Campbell, Philip Converse, Warren Miller and Donald Stokes (New York: John Wiley & Sons, Inc., 1966), pp. 161-179.

Truman, David B., "Federalism and the Party System," in Arthur W. MacMahon Federalism, Mature and Emergent (New York: Doubleday . & Co., 1955), pp. 115-136.

Verba, Sydney and Norman Nie Participation in America: Political Demo­ cracy and Social Equality (New York: Harper and Row, 1972).

Weisberg, Herbert and Jerrold G. Rusk, "Dimensions of Candidate Evalua­ tion," American Political Science Review, Vol. 64 (Decmeber, 1970) pp. 1167-1185.

, "Perceptions of Presidential Candidate Implications for Elec­ toral Change," Midwest Journal of Political Science (August, 1972), pp. 388-410.

White, Theodore The Making of the President, 1972 (New York: Bantam, 1973). 178

Wilson, James Q. The Amateur Democrat (Chicago: University of Chicago Press, 1966).

Political Organizations (New York: Basic Books, Inc., 1973).