Political Culture and Dimensions of Variation in the Content of Governorsí Speeches.

55th Annual Meeting of the Midwest Political Science Association, Chicago IL, April 1997.

Daniel DiLeo, Penn State Altoona

ABSTRACT: This study analyzes the content of governorsí speeches from forty-four states to conduct a two-part test of Elazarís typology of political culture. In the first part of the test, seven factors are extracted from twenty-one frequently-occurring nouns that function as political symbols. Elazarís typology helps to account for the statesí factor scores. In the second part of the test, Elazarís typology helps to explain the frequencies of words that communicate the basic goals and rhetorical strategies of the moralistic and individualistic political cultures, but has less success in accounting for the goals and strategies of the traditionalistic type.


Thirty-five years after the publication of Almond and Verbaís seminal work,1 the definition of political culture continues to be an object of contention.2 3 Even if we limit ourselves to the study of political culture in the United States, fundamental disagreement persists about the meaning of the concept. There are those who hold that political culture is a set of preferences that citizens choose4 and those who hold that it is a set of attitudes instilled in citizens through processes of political socialization,5 which may depend primarily on levels of economic well-being.6 Others hold that political culture resides in the hearts and minds of political elites7 or in the nature of key political institutions, such as party systems.8 It would appear that vexing questions about freedom and determinism, structure and super-structure, and the role of elites in contemporary political regimes have precluded agreement as to where to look for political culture at work. Such is not the case.

All of the conflicting understandings of political culture mentioned above share the view that political culture manifests itself in political action. This tells us to look for political culture by looking at people engaged in institutionalized processes where possible courses of government action are proposed, struggled over, chosen, and carried out. It is worth noting that all of these processes entail communication between political actors.9 Therefore, variations in political culture should be manifest in variations in the content of political communication. We should find that actors in different political cultures emphasize different goals and employ different rhetorical strategies when they communicate with other actors in their respective political cultures.

Such an approach is not without precedent. Studies that use speeches and texts to classify culturally-defined values and strategies of political actors are numerous and diverse.10 Furthermore, the political import of rhetoric is well recognized.11 Therefore, the use of political rhetoric as an instrument for studying political culture has considerable theoretical appeal.12 Nevertheless, despite continuing calls for measuring manifestations of political culture in ways that are justifiable on theoretical grounds,13 there has been only one published study testing for the presence of distinctive types of rhetoric in Elazar's moralistic, traditionalistic, and individualistic political cultures. That was Joslyn's study of political advertising.14 The lack of studies analyzing rhetoric across Elazar's political cultures is particularly noteworthy in view of his explicit call for such examinations.15

This study is an attempt to fill that gap through two separate analyses of the content of political rhetoric. The first analysis will determine whether the principal dimensions of variation in the content of governorsí speeches correspond to variations in the predominant political cultures in their states as described by Elazar. The second analysis will determine whether the governors tend to use language that espouses the central goals and employs the characteristic rhetorical strategies of the types of political culture that Elazar identified as predominant in their states.

DATA In order to serve as a measure of political culture, the rhetoric examined had to be roughly comparable across states, politically significant, and addressed to a cross-section of important political actors. The forty-two state-of-the-state speeches that the governors delivered in 1996 fit these specifications. In two of the remaining eight states, the governors gave budget addresses that presented the governorsí overall agendas.16

The speeches share a number of characteristics that make them useful and roughly comparable. The governors used them to present their agendas at the start of the legislative sessions in their respective states and to present themselves and their ideas in a favorable light to other significant political actors.17

METHODS OF ANALYSIS A comparison of the three most commonly used methods of content analysis has found that the most efficient method of discriminating between thought patterns of speakers is word retrieval. This method counts pre-selected words in a text and divides that number by the total number of words, deriving a frequency count. Gains in validity obtained by elaborate programs designed to distinguish between ambiguous uses of words or to detect unexpected usages were found to be minimal. The use of trained human coders to read, interpret and classify portions of the texts, or whole texts, was both more costly and less effective at discriminating between individuals whose thought patterns were known to differ.18 It is generally believed that the reason for the effectiveness of word retrieval is that "the relative frequencies with which particular words are used is an indication of their cognitive centrality or importance."19

This study consisted of two replicable word retrieval analyses of the content of the speeches in the data set. The analyses were designed to test hypotheses derived from Elazar's typology. The first was an inductive study that examined the content and structure of the governors' rhetoric without the prior imposition of theoretical constructs.20 It began with a factor analysis with varimax rotation of a set of words selected because they referred to broad classes of objects and had consistently positive or negative connotations. In other words, they served as symbols in the sense described by Lowell Dittmer.21 Each of the words was among the twelve most frequently occurring nouns in two or more of the forty-four speeches in the data set.22

Once the factor scores were generated,23 they were regressed on dummy variables for the political culture types and some economic and political variables that served as benchmarks and controls in a set of multivariate OLS regressions.

The political variables were public opinion liberalism and party of the governor. The indicator for public opinion was the liberalism index calculated by Erikson, Wright and McIver.24 The high correlation between public opinion liberalism and elite liberalism25 provides a likely path through which this variable might determine the content of the governors' speeches. The party of the governor was included as a dummy variable, scored 1 for Democrats and 0 for the Republicans and the Independent governor of Maine, Angus King.

The economic variables were state per capita income and growth in personal income during the preceding year. Inglehart and Abramson found that higher current per capita income is associated higher proportions of postmaterialists, people concerned with freedom, self-expression, and quality of life and a lower proportion of materialists, people concerned with economic and physical security.26 Postmaterialism might affect governors' rhetoric through a variety of paths, any one of which might link per capita income to certain combinations of words found in the speeches. Per capita income might also affect the rhetoric of the governors in other ways, such as through its impact on interest group systems, prevailing levels of education, the availability of resources and the professionalism of the various institutions of government.

Growth in personal income was included because the lack of it can create distinctive challenges for governors.27 It is likely that governors might use the state-of-the-state speeches to respond to these challenges. Economic growth was measured as percent change in personal income from 1994 to 1995.28 Other variables, such as characteristics of the stateís bureaucracies, political parties, legislatures, interest groups, governorsí staffs, or of the governors as individuals were not included because some of their characteristics are manifestations of political culture as described by Elazar.

The hypotheses for this phase of the analysis were that the strength of the relationships between the factors and the political culture dummy variables should compare favorably to the strength of the relationships between the factors and the political and economic variables.

The second part of the analysis was a deductive test of Elazar's theory. A careful reading of Elazar's theory of political culture and his descriptions of the three principal types produced a set of twenty-eight words and phrases that the author believed indicative of culturally characteristic rhetorical strategies or goals. Table 1 contains a list of the words and phrases.

(Table 1 about here)

It is not necessary for every occurrence of every word or phrase on the lists to convey the intended message. As long as occurrences of the listed words are even slightly more likely than unlisted words to pick up messages characteristic of the intended cultures and as long as they are more likely to communicate the messages that are intended than messages that directly contradict the intended messages, the lists will pick up the appropriate biases in meaning. However, occurrences of words on the lists that communicated unintended messages, less than five percent of the occurrences, were excluded from the frequency counts.29

As in the first analysis, a set of multivariate OLS regressions of the frequency counts on the political culture dummies and the economic and political variables provided an assessment of Elazarís typology as an explanation for the content of the speeches. In order to provide a preliminary assessment of the validity of the frequency counts, this phase of the analysis concluded with a brief discussion of key passages in the speeches with the highest frequency counts for each of the three political culture indicators.

The hypotheses in this phase of the analysis were that the relationships between the frequencies of words communicating the core goals and strategies of a given political culture and the dummy variable for that political culture should be positive in direction and at least comparable in strength to their relationships to the political and economic control variables. FINDINGS In the first analysis, the scree method suggested the extraction of seven factors. They accounted for 68% of the variance. All of them had eigenvalues greater than 1.28. Table 2 displays the symbols used in the factor analysis, their eigenvalues and their loadings on the rotated factors.

(Table 2 about here)

Table 3 displays the standardized coefficients produced by regressing the factor scores on the political culture dummies, the political variables and the economic variables.

(Table 3 about here)

The political culture dummies outperformed the economic and political variables on factors 1, 4, 5 and 7. This is a fairly good performance, considering that public opinion liberalism, party of the governor, per capita income, and economic growth are known to have powerful impacts on state politics and policy. On the whole, the moralistic and individualistic political cultures were much more strongly related to the seven factors than the traditionalistic political culture was.

It is tempting to interpret the factor scores as indications of particular orientations to political action. However, the hypothesis in this phase of the analysis rests on the conservative use of the factor scores supported by Rosenberg, Schnurr, and Oxman.30 The hypothesis is only that governors operating in different political cultures should use different groups of symbols, not that certain governors should use certain symbols.

Table 4 presents the standardized coefficients for the OLS regressions of the frequencies of each of the twenty-eight words and phrases on the explanatory variables.

(Table 4 about here)

Unlike the inductive phase, this phase of the analysis tests hypotheses that rely on assumptions about the meaning and significance of the retrieved words.

The results of these regressions were also basically supportive of Elazarís typology. After controlling for the effects of the political and economic variables, the moralistic and individualistic political culture dummies proved to be comparatively successful in accounting for the use of words and phrases that communicated the central goals or employed the characteristic rhetorical strategies of the moralistic and individualistic political cultures. Ten of the twelve coefficients for the moralistic dummy and seven of the nine coefficients for the individualistic dummy were in the expected direction. The coefficients of these dummies with the total frequencies for each set of words were the largest in their respective regression equations. The traditionalistic political culture dummy was considerably less successful. This may be due to the recent erosion of that political culture which Elazar and others have observed.

The speech with the highest frequency of terms coded as indicators of the individualistic political culture was the speech that Governor Almond, Republican of Rhode Island, gave. This speech contains thirty-one occurrences of the coded terms, twenty-nine of which were used to compute the frequency count.

The high frequency count was due primarily to the twelve occurrences of the word, "jobs." The word always indicated attentiveness to the economic well-being of the stateís people and the role of government in creating and maintaining an economic order that provides opportunity. These concerns are characteristic of the individualistic political culture. The following are the first two passages that contain occurrences of the word "jobs."

Last year, I pledged that during my term in office I would bring fundamental change to Rhode Island Ė change aimed at developing new strategies to increase jobsÖ

Creating jobs is not easy. It takes years to have available, ready, and fully permitted commercial sites.

Not surprisingly, this speech differed from most of the others in its high level of attentiveness to the issue of jobs.

The speech with the highest frequency count for words coded as indicators of the traditionalistic political culture was the speech that Governor Leavitt, Republican of Utah gave. This was unexpected, because Elazar classified Utah as a moralistic state. This speech had sixteen occurrences of coded terms. One of these, "Utah is already a national leaderÖ" was not counted because it did not express the idea that leaders should play an essential role in the political life of the state.

The speechís high score was largely due to its unusually high frequency of the coded word, "values." It occurred in passages such as

Utah will be a place of quiet quality, a mentor state, a place where people pass on to future generations the ageless values. Like the youngest brother who preserved the glowing embers until the flame could be rekindled, Utah can be among the places where the world turns to renew its sense of basic values.

On the whole, this speech placed more emphasis on the continuity of traditions and was somewhat more critical of bureaucracy than most of the other speeches. It also emphasized the role of leaders and their obligation to the welfare of the whole community more than most of the other speeches. These features of the speech indicate that its high score on the traditionalistic indicator is due to a strong presence of traditionalistic ideas. The fact that the states of the South did not score high on this indicator provides additional evidence that the traditionalistic political culture is no longer dominant there. At the same time, Utahís high score amplifies Elazarís warning that the moralistic and traditionalistic types should not be taken to be polar opposites. Moralists are interested in maintaining communal traditions when they believe that those traditions enhance the ideals and cohesion of the community.31

The speech with the highest frequency of items coded as moralistic was the speech that Governor Graves, Republican of Kansas gave. The speech contained eighteen occurrences moralistic items. One occurrence of the word "ethic" was not coded because it occurred in the phrase "work ethic," which may not be a typically moralistic concern.

The term most responsible for the high score for this speech was "public trust," which occurred four times. It occurred in passages such as the following:

I propose a public trust initiative that will put reasonable and sound ethics policies into place.

During this session I know there will be numerous proposals to enhance public trust of government.

Concern with political ethics, and the framing of those issues in the context of "public trust" are consistent with the moralistic political culture. On the whole, this speech is much more concerned than most with the moralistic theme of using government to help construct and maintain a more righteous social order.

CONCLUSION This study shows that the content of the governorsí speeches varies with the political cultures in which they operate. The implications of this finding are methodological as well as substantive.

In recent years, records of communications between political actors have become increasingly available and we have gained the capacity to conduct replicable and increasingly sophisticated analyses of ever larger quantities of text. These developments could significantly enhance our ability to describe and compare the agendas and strategies of various political actors in a variety of situations and contexts.

Substantively, this study adds further weight to the claim that the political culture of the Southern states is less distinctive today than it was thirty years ago. At the same time, however, the study also suggests that although public policy in the states is largely determined by national and international events and trends, the agendas of the governors are one avenue through which cultural forces that are specific to particular regions influence the policy process.


Endnotes

1. Gabriel A. Almond and Sidney Verba, The Civic Culture (Boston: Little, Brown, 1963).

2. The speeches used for this study are available from the author. Word frequencies were obtained using Flextext. Statistical analyses were conducted on NCSS.

3. The author is grateful to James C. Lech for his capable assistance in obtaining copies of the governors' speeches.

4. See, for example, Aaron B. Wildavsky, "Choosing Preferences by Constructing Institutions: A Cultural Theory of Preference Formation," American Political Science Review 81 (March 1987): 3-21.

5. See Gabriel A. Almond and Sidney Verba, The Civic Culture.

6. Ronald Inglehart, "The Renaissance of Political Culture," American Political Science Review 82 (December 1988): 1203-30.

7. See, for example, Ruth Lane, "Political Culture: Residual Category or General Theory?" Comparative Political Studies 25 (Fall 1992): 362-387 and Lucian W. Pye, "Political Culture Revisited, - Political Psychology 12 (September 1991): 487-508.

8. See V.0. Key, Southern Politics in State and Nation (New York: Alfred A. Knopf, 1949).

9. Doris A. Graber, "Political Languages," Handbook of Political Communication, eds. Daniel D. Nimmo and Keith R. Sanders (Beverly Hills: Sage, 1981), 196.

10. Roderick P. Hart, Verbal Style and the Presidency, (Orlando: Academia, 1984); J. Zvi Namenwirth and Robert Phillip Weber, Dynamics of Culture, (Winchester, Mass.: 1987); Jeffrey K. Tulis, The Rhetorical Presidency (Princeton: Princeton University Press)

11. See, for example, Jeffrey E. Cohen, "Presidential Rhetoric and the Public Agenda," American Journal of Political Science 39 (Winter 1995): 87-107; Doris A. Graber, "Political Languages," pp. 199, 216; David Green, Shaping Political Consciousness: the Language of Politics in America from McKinley to Reagan (Ithaca: N.Y.: Cornell, 1987); Harold Lasswell, Nathan Leites, and associates The Language of Politics (Cambridge, Mass, M.I.T. Press, 1966),p. 67.

12. Manuel Eisner, "Long-term Dynamics of Political Values in International Perspective: Comparing the Results of Content Analysis of Political Documents in the USA, GB, FRG and Switzerland," European Journal of Political Research 18 (Fall 1990): 605-621; Paul Nesbitt-Larking, "Methodological Notes on the Study of Political Culture," Political Psychology 13 (Winter 1992): 79-90.

13. Ruth Lane, "Political Culture: Residual Category of General Theory?": 362- 387; Robert L. Savage, "Looking for Political Subcultures: A Critique of the Rummage Sale Approach," Western Political Quarterly 34 (June 1980): 331-336; Frederick W. Wirt, "'Soft' Concepts and 'Hard Data: A Research Review of Elazar's Political Culture," Publius: The Journal of Federalism 21 (Spring 1991): 1-13.

14. Richard Joslyn, "Manifestations of Elazar's Political Subcultres: State Opinion and the Content of Political Campaign Advertising," Publius: The Journal of Federalism 10 (Spring 1980): 37-58.

15. Daniel J. Elazar, "Afterword: Steps in the Study of American Political Culture," Publius: The Journal of Federalism 10 (Spring 1980): 135.

16. The states represented in the data set with budget speeches were Kentucky and Pennsylvania. The states not represented were Arizona, where Governor Symington's budget speech was not an agenda-setting speech, and Arkansas, Montana, Nevada, North Carolina and Texas, where the governors gave neither state of the state nor budget speeches in 1996.

17. Thad L. Beyle, "The Governor as Chief Legislator," Being Governor: The View from the Office, ed. Thad L. Beyle (Durham, N.C.: Duke University Press, 1983): p. 133; James J. Gosling, "Patterns of Stability and Change in Gubernatorial Policy Agendas," State and Local Government Review (Winter 1991): 3-12; Margaret R. Ferguson and Donald H. Ostdiek, "The Scope and Success f Gubernatorial Legislative Agendas" (paper presented at the annualmeeting of the Midwest Political Science Association, Chicago, April 1997); Eric B. Herzik, "Governors and Issues: A Typology of Concerns," State Government, February 1983, pp. 58-64.

18. Stanley D. Rosenberg, Paula P. Schnurr, and thomas B. Oxman, "Content Analysis: A Comparison of Manual and Computerized Systems," Journal of Personality Assessment 54 (Spring 1990): 298-310.

19. Boris Kabanoff and John Holt, "Changes in the Espoused Values of Australian Organizations, 1986-1990," Journal of Organizational Behavior 17 (Spring 1996): 206.

20. Kevin W. Mossholder, Randall P. Setton, Stanley G. Harris, and Achilles Armenakis, "Measuring Emotion in Open-Ended Survey responses: An Application of Textual Data Analysis," Journal of Management 21 (Summer 1995): 335-355.

21. Lowell Dittmer, "Political Culture and Political Symbolism," World Politics 29 (July 1977): 566; see also Stephen P. Chilton, "Defining Political Culture, Western Political Quarterly 41(Fall 1988): 419-445.

22. Nine nouns were among the twelve most frequently occurring in two or more speeches, but were not included in the analysis because they did not have consistent positive or negative connotations. Those nouns the following: future, government, percent, system, welfare, dollars, funding, year, and years.

23. Scores on the seven factors for each speech are available on request. The author does not claim that the factors signify emphasis on any particular agenda items or the use of any particular rhetorical strategies, only that they indicate dimensions of variation in content.

24. Robert S. Erikson, Gerald C. Wright, and John P. Mclver, Statehouse Democracy: Public Opinion in the American States (New York: Cambridge, 1993), p. 16.

25. Ibid., 106-113.

26. Ronald Inglehart and Paul R. Abramson, "Econoirtic Security and Value Change," American Political Science Review 88 (June 1994): 1203-1230.

27. Thad L. Beyle, Governors and Hard Times, (Washington, D.C., 1992).

28. Data on personal income and change in personal income from United States Department of Commerce, Statistical Abstract of the United States (Washington, D.C.: U.S. Department of Commerce, 1996), p. 452.

29. Upon request, the author will furnish a tabulation of the occurrences of listed words that were excluded from the frequency counts.

30. Stanley D. Rosenberg, Paula P. Schnurr, and Thomas B. Oxman, "Content Analysis: A Comparison of Manual and Computerized Systems," Journal of Personality Assessment 54 (Spring 1990): 298-310.

31. Elazar, "Afterword: Steps in the Study of American Political Culture."


Table 1: Words and phrases communicating core messages of Elazar's political cultures

I. Moralistic: A. Goal of a just community 1. common purpose, common goal, common goals, common good 2. public good, public service, public trust 3. vision, visionary 4. work together, working together, worked together B. Rhetorical emphasis on the ethical or moral dimensions of conduct and motives 1. commitment, commitments 2. duty, duties 3. honest, honesty, dishonest, dishonesty 4. selfish, unselfish, selfless, selfishly, selflessly 5. principle, principles, principled 6. corrupt, corrupting, corruption 7. ethics, ethical, unethical 8. lobby, lobbies, lobbyist, lobbyists, lobbying

II Traditionalistic: A. Goal of preserving existing social order by resisting bureaucracy 1. bureaucrat, bureaucrats, bureaucratic, bureaucracy B. Rhetorical invocation of tradition and traditional sources of authority 1. God 2. heritage 3. tradition, traditions, traditional 4. uphold, upholding 5. values C. Rhetorical strategy of justifying reliance on hierarchical socio-political order 1. leader, leaders, leadership

IV. Individualistic A. Goal of responsiveness to the demands of citizen/clients 1. people demand, peopleís demands, people demanded, people demanding, people want, people wanted 2. public demands, publicís demands, public demanded, public demanding, public wants, public wanted 3. respond, responds, responding, responded, responsive, responsiveness B. Goal of economic opportunity and economic well-being of citizen/clients 1. business climate, business needs, business environment, business appeal 2. entrepreneur, entrepreneurs, entrepreneurial 3. job, jobs 4. new business, new businesses 5. small business, small businesses 6. economic growth, economic development, economic performance, economic well-being, economic health, economic resurgence, economic progress, economic climate, economic opportunity, economic momentum, economic vitality, economic advancement, economic improvement, economic success, economic prosperity.


Table 2: Loadings of Twenty-One Frequently Occurring Words on Rotated Factors

Factor 1

Factor 2

Factor 3

Factor 4

Factor 5

Factor 6

Factor 7

children

.8191

health

-.8948

schools

.7936

people

-.7854

jobs

-.8013

business

.7513

program

.7320

kids

.8056

care

-.6844

students

families

development

.7269

job

-.7850

economy

.7264

services

.6020

care

.5013

tax

.3444

parents

.4569

family

.3577

families

-.2513

families

-.3330

business

.2500

family

.3897

program

-.2556

department

-.4253

services

.3297

community

-.2345

development

.3191

economy

-.2491

tax

-.3756

community

-.2308

services

-.3184

economy

.3108

care

-.2104

department

.2839

department

.2248

education

-.2662

economy

-.2154

family

-.2377

tax

-.2570

parents

-.2040

services

.2469

jobs

-.2070

schools

.2494

family

.1761

development

-.1825

families

.2019

schools

-.2039

care

.1924

job

.1924

families

.2363

families

-.1506

people

-.1552

kids

-.1508

people

-.1957

students

-.1344

care

.1868

services

-.2243

students

-.1191

business

-.1541

parents

.1230

students

-.1818

schools

-.1342

health

.1848

development

-.2078

jobs

-.1026

job

.1536

students

.1216

kids

.1328

education

.1090

people

-.1659

job

.1922

parents

.0779

education

.1078

children

.1200

tax

.1149

job

.1069

tax

-.1413

economy

.1546

children

-.0677

economy

-.0999

health

.1013

business

-.1143

parents

.0892

children

-.1388

jobs

-.1376

services

-.0488

program

.0969

care

-.0890

department

-.1114

jobs

.0557

parents

-.1214

community

.1224

development

.0353

care

.0925

program

.0863

development

-.1052

children

-.0388

community

.1121

department

-.1215

department

.0255

jobs

.0772

department

.0861

education

-.1016

program

.0280

schools

.1090

people

-.1099

people

-.0077

health

-.0454

jobs

.0794

health

.0739

community

-.0266

family

-.0886

program

-.0624

business

-.0067

children

.0386

education

.0618

children

-.0656

health

.0226

development

-.0582

business

-.0297

schools

.0053

community

.0346

schools

.0235

program

.0261

tax

.0220

students

-.0537

health

-.0185

job

-.0039

kids

.0304

job

-.0190

services

-.0245

kids

-.0167

families

.0456

students

-.0002

kids

-.009

tax

.0070

business

.0032

economy

-.0214

people

-.0147

kids

.0216

eigenvalue

3.5051

eigenvalue

3.0214

eigenvalue

1.8967

eigenvalue

1.7744

eignvalue

1.4991

eigenvalue

1.2993

eigenvalue

1.2823


Table 3

OLS Regression of Factors on Political Culture Dummies and

Political and Economic Control Variables1

Highest-

Loading

Words:

Factor 1:

children,

kids

Factor 2:

care (-),

health (-)

Factor 3:

schools,

students

Factor 4:

people (-),

development

Factor 5:

jobs (-),

job (-)

Factor 6:

business,

economy

Factor 7:

program,

services

Political Cultures:

(combination with

largest beta wts)

MPC2:

.3009

IPC:

.2092

MPC: 

-.2145

IPC4:

-.2020

MPC:

.1027

TPC:

.0304

MPC:

.2917

IPC: 

.4765*

MPC:

.5280***

TPC:

.4155

MPC:

-.0406

IPC:

-.0314

TPC3:

.0588

IPC:

.2633

Party5 .0624 -.3955** .3356* .0844 .3154** .0025 -.1991
Liberal6 .0547 -.3030 -.0293 -.1369 -.3966** .2214 -.2156
Growth7 .0826 -.0985 -.1685 -.0019 -.1437 -.1362 -.1082
Income8 .2511 .3979 .0887 -.2088 .1881 -.0271 .1395
R2 .1012 .3019 .1054 .1183 .3700 .0534 .1627

1. The coefficients are beta weights.

2. States with predominantly moralistic political culture = 1, others = 0.

3. States with predominantly traditionalistic political culture = 1, others = 0.

4. States with predominantly individualistic political culture = 1, others = 0.

5. Democratic governors = 1, others = 0.

6. Public opinion liberalism, Erikson, Wright, McIver index (1993). No measures for Alaska and Hawai'i, therefore these regressions did not include those states.

7. Change in state personal income, 1994 to 1995.

8. Per capita income, 1995.


Table 49

OLS Regression of Frequencies of Selected Words and Phrases on Political Culture Dummies

and Economic and Political Controls

MPC TPC IPC Party Liberal Income Growth R2
Moralistic
commit... .31 .10 .23 .34 .01 -.06 .212
duty .20 .23 .07 -.06 .37 -.08 .075
honest .09 .16 -.33** -.23 -.36 .18 .364
...self... .16 .26 -.01 -.28 -.37 .01 .079
principle .04 .32 -.12 -.13 -.05 -.05 .192
common .07 -.01 .16 .02 .06 -.06 .031
public .10 .07 -.18 -.10 .04 .08 .057
vision .21 -.15 .15 .44** -.35 .07 .255
work together -.26 .05 .06 .08 -.21 .06 .125
lobby .11 .37 .02 -.28 .36 .21 .172
corrupt -.14 .05 -.13 -.06 -.24 .09 .125
ethic .29 .22 .23 -.22 .00 .23 .161
Total Moralistic .30 .29 .04 .14 -.11 .18 .148
Traditionalistic
bureaucra... -.04 -.04 -.18 -.16 .19 .11 .067
God -.00 -.02 -.33* .02 -.15 .05 .114
heritage -.09 .06 .07 .26 -.27 .04 .081
leader .22 .00 .13 -.05 -.14 .11 .112
tradition -.07 -.18 -.31* .13 -.15 .23 .100
uphold .08 -.09 -.02 .03 .01 .09 .031
values .39** -.00 -.18 -.13 .19 .31* .297
Total Traditionalistic .21 -.06 -.26 -.06 .-.04 .28* .186
Individualistic
public demands -.45** -.63*** .32* -.40* .62** .04 .308
people demand -.22 -.04 -.20 -.12 -.07 .25 .138
responsive, respond .20 .61*** -.25 -.14 -.45* -.00 .312
business climate .00 .28 .15 -.61*** .34 .02 .283
economic growth -.10 .11 .12 -.05 -.06 -.23 .102
entrepreneur... .15 .08 .04 .33 -.32 .05 .084
jobs -.04 .46** -.19 .33 -.26 .19 .251
small business -.01 -.08 -.24 .24 -.03 -.06 .101
new business .05 .12 -.32* .39* -.36 .02 .135
Total Individualistic -.03 .48** -.25 .37* -.33 .20 .250

* p < .10; ** p < .05; *** p < 01