1
Human Development as a Theory of Social Change:
∗
A Cross-Cultural Perspective
Chris Welzel
School of Humanities and Social Sciences
International University Bremen (IUB)
Ronald Inglehart
Institute for Social Research (ISR)
University of Michigan
Hans-Dieter Klingemann
Department “Institutions and Social Change”
Social Science Research Center Berlin (WZB)
Abstract
This article demonstrates that socioeconomic development, cultural change and democratization
constitute a coherent syndrome of social change— a syndrome not properly specified by classical
modernization theory. We refer to this syndrome as Human Development, arguing that its three
components have the common theme of broadening human choice. Socioeconomic development
broadens peoples’ choice by providing them with individual resources; cultural change gives rise to
greater emphasis on self-expression values that lead people to give higher priority to having
autonomous choice in society; and democratization provides people with increasingly effective
rights, giving legal guarantees for human choice in politics. Analysis of data from the World Values
Surveys demonstrates: (1) that the syndrome of individual resources, self-expression values and
effective rights is universal in its presence across nations, regions and cultural zones; (2) that this
Human Development syndrome is shaped by a causal effect from individual resources and selfexpression values on effective rights; and (3) that this effect operates through its impact on elite
integrity,
∗
as
the
factor
that
makes
given
rights
effective.
We are grateful to Barry Hughes, Hanspeter Kriesi, Seymour Martin Lipset and the anonymous
revewiers of EJPR for helpful comments on earlier drafts of this article.
2
Introduction
Students of social change have emphasized three major processes. The most fundamental one,
socioeconomic development, has been described most extensively (among many others see Lewis
1955; Rustow 1963; Bell 1973; Chirot 1986; Perkin 1996; Rowen 1996; Barro 1997; Sen 1997;
Estes 1998; Rodrik 1998; Hughes 1999). There is broad consensus that socioeconomic
development reflects a set of closely linked changes including productivity growth, improving health
and life expectancy, increasing material prosperity, expanding education and communication, and
increasing social complexity.
The second process, value change, is linked with socioeconomic development, that brings
rising life expectancies and rising levels of existential security. Moreover, expanding markets and
social mobilization increase human interactions and horizontal networks among societies, that tends
to transform authority relations into bargaining relations, emancipating people from rigidly hierarchical
ties that restrict human autonomy (Weber 1954; Banfield 1958; Eckstein 1988; Coleman 1988). As
this happens, peoples’ prevailing value orientations tend to be reshaped in ways that have been
described in various terms, such as the emergence of “civic cultural” values (Almond & Verba
1963), “individual modernity” (Inkeles & Smith 1974; Inkeles 1983), “postmaterialist values”
(Inglehart 1977; 1990), “liberal values” (Brint 1984; Flanagan 1987; Nevitte 1996),
“anthropocentric values” (Bürklin, Klein & Ruß 1996), “emancipatory values” (Clark 1998; Welzel
2002) or “self-expression values” (Inglehart & Baker 2000; Klages & Gensicke 1999;). Whatever
the terminology, most theories of value change converge in the notion that traditional-deferential
orientations, that subordinate the individual to the community, tend to give way to growing emphasis
on autonomous human choice and individual self-expression. Following Inglehart and Baker (2000),
we characterize this process as a shift from survival values to self-expression values.
A third major process involves a society’s political institutions. The most notable development
in this field has been a massive rise in societies’ democratic performance. This has happened in two
ways during the past three decades. Most obviously, many previously authoritarian regimes changed
into constitutional democracies by adopting basic democratic rules in the “Third Wave of
Democratization” (Huntington 1991; Sørensen 1993; Kurzman 1998; Nagle & Mahr 1999;
Dorenspleet 2000). But at the same time, a more subtle change has taken place in established
democracies. Since the late 1970s, most of them have implemented or extended direct democratic
3
institutions (Butler & Ranney 1994; Cronin 1998; Scarrow 1999) and they have experienced rising
levels of direct civic participation (Barnes & Kaase et al. 1979; Budge 1996; Dalton 1996). Some
scholars see these changes as an acceleration of a more enduring historical trend towards the
“growth of democracy” (Gurr, Jaggers & Moore 1990; Diamond 1993; Modelski & Perry 1993;
Jaggers & Gurr 1995).
Some writers have called questioned the existence of processes of socioeconomic
development, value change and democratization (see Randall & Theobald 1998:chapters 1-2). It has
been debated, for instance, whether these processes manifest irreversible linear trends or follow
cyclical patterns with major setbacks; whether they are uniformly global or culture-specific in a way
that prescribes an inherently Western model; and even whether they are desirable or not.
But one point seems clear: if socioeconomic development, cultural change and democratization
do occur, they tend to go together. Impoverished societies, suffering from scarce resources, tend to
be characterized by survival values. And these societies usually have autocratic political regimes,
being “formal democracies” at best and rarely “effective democracies.” At the other end of the
continuum, literally all of the OECD-societies have high levels of economic development and, as has
been demonstrated (Inglehart, 1997;
Inglehart and Baker, 2000), their value systems are
characterized by relatively strong emphasis on self-expression. Moreover, in virtually every case,
their political systems function as effective democracies. Overall, high (or low) levels of individual
resources, self-expression values and effective democracy tend to go together, as Figure 1
demonstrates (measurement of the variables in this figure is explained in section 2.1).
4
90
80
(1999-2000)
70
60
Spain
50
40
30
20
10
60
50
40
30
20
10
(e
al
du s
ivi rc e s)
In d so u 1990
Re a rly
Figure 1.
Protestant Western Zone
Germany (E.)
France
Catholic Western Europe
Belgium
Japan
Portugal
Sl o Chile
Italy
ve
ni
a
Taiwan
Uruguay
Gr
ee
ce
Estonia
Czech R.
S-Africa
Hungary
Poland
Argentina
Latvia
Lithuania
Slovakia
El Salv.
Brazil
Peru
Mexico
Ghana
Columbia
India
Croatia
Uganda
va
ij.
l d o w e ba
Mo b a b zer
A
Zim
+
G.B.
Germany (W.)
Ireland
Catholic Eastern Europe
Latin America
a
S-Kore
Effec tive Dem oc racy
+
Netherld. Denmark
Finland
New Zealand
Iceland
Sweden
da
Norway
Luxemburg na
Australia
a
C
Switzerld.
100
China
-1,6
Egypt Nigeria
-1,0
-,4
,2
Developed Far East
Orthodox Eastern Europe
Islamic Zone
Sub-Saharan Africa
,8
Pakistan
1,4
alues
Se lf-Exp ression V
2,0
+
(ca. 1995)
The Syndrome of Human Development and Its Three Components
This insight is not new. The fact that richer countries are more likely to be democratic, was
argued by classical modernization theory (see Lerner 1958; Lipset 1959; Coleman 1968; Pye 1990;
Diamond 1992). What is new, is the empirical evidence of the role of value change that has become
available in recent years (Inglehart 1997; Inglehart & Baker 2000). Nevertheless, we do not yet
have an integrated theory of social change. Modernization theorists have argued that there are close
relations between socioeconomic development, value change and degrees of democracy, without
specifying any common theme underlying these three phenomena. Modernization was either used as
an umbrella term that was defined by enumerating its components without specifying what integrates
them (for example Lerner 1968:385); or modernization was specified in abstract terms, such as
“functional differentiation” (Mouzelis 1999), that provide no clear criteria to distinguish what is and
what is not an element of modernization.
Empirical studies reflect this lack of theoretical integration. Most analyses focus on only one of
the three relationships between socioeconomic development, cultural change and democracy,
without integrating the three factors coherently. Even the few studies that deal with all three
processes dissolve the whole complex into single pairs of relations, each of that is discussed
5
separately (Muller & Seligson 1994; Inglehart 1997; Sides 1999; Inglehart & Baker 2000). As a
result, the debate is fragmented.
Following Lipset (1959), various authors have claimed that socioeconomic development helps
to establish or sustain democracy (among others Cutright 1963; Bollen & Jackman 1985; Lipset,
Seong & Torres 1993; Burkhart & Lewis-Beck 1994; Barro 1997; Vanhanen 1997; Gasiorowski
& Power 1998), while others argue that democracy (or at least “effective democracy”) promotes
political stability, provides better economic policies and thus is conducive to socioeconomic
development (Ersson & Lane 1996; Rowen 1996; Leblang 1997; Yi Feng 1997; Frey & Al-Roumi
1999; Olson, Sarna & Swamy 2000). Some observers postulate that socioeconomic development
propels value change (Inkeles & Smith 1974; Inkeles 1983; Flanagan 1987; Inglehart & Baker
2000), but others hypothesize that rational “modern” values, such as interpersonal trust, accelerate
socioeconomic development (Putnam 1993; Fukuyama 1995; Knack & Keefer 1997; Landes
1998). And while some analysts suggest that democracy helps to produce pro-democratic values
(Rustow 1970; Muller & Seligson 1994; Jackman & Miller 1998), others emphasize the opposite
flow of causation: changing mass values put political elites under growing pressure to institutionalize
democratic rules and to make these rules effective (Inglehart 1997:chapter 5; Welzel & Inglehart
2001; Welzel 2002).
Summarizing these contradictions, Dahl (1998:35) concluded that “the exact nature of the
relationship among socioeconomic modernization, democratization, and the creation of a democratic
culture is almost as puzzling today as it was a quarter-century ago.”
It remains puzzling because no one, as far as we know, has started from the most fundamental
question: “What is the common denominator underlying socioeconomic development, changing
values and democracy?” This question is made all the more pressing by the fact that these three
processes go together to a striking extent, as we will demonstrate. Specifying the common theme that
underlies socioeconomic development, value change and democratization helps to better understand
the specific role played by each of the three subprocesses; and this in turn illuminates the fact that
these subprocesses are logically connected.
The concept that the core principle of modernization is the broadening of “human choice,” is
implicit in modernization theory (Lewis 1955:9-19). But the extent to that this concept can integrate
major changes in socioeconomic structure, political culture and political institutions has not yet been
fully developed. The following section proposes that we use the concept of Human Development as
an integrating framework. Anand (1993), Sen (2000) and Anand and Sen (2000) introduced the
6
term Human Development, arguing that “human choice,” or the capability of human beings to choose
the lives they want, should be the ultimate measure of social progress. We follow Anand and Sen’s
humanistic approach, developing the concept of Human Development more comprehensively in a
way that includes political culture. Using this framework, we analyze data from the World Values
Surveys together with socioeconomic data from Vanhanen (1997), civil and political rights ratings
from Freedom House and measures of elite corruption from Transparency International. Subsequent
sections demonstrate (1) that the syndrome of Human Development applies across nations, regions
and cultural zones; (2) that this syndrome is shaped by a process in that socioeconomic development
and emerging self-expression values lead to rising levels of effective democracy; and (3) that the
effect of self-expression values on effective democracy operates through their impact on elite
integrity. Indeed, elite integrity (i.e., the reverse of corruption, as measured by Transparency
International) is the key factor that makes the difference between mere formal democracy and
effective democracy.
1.
1.1
Theory
The Three Components of Human Development
The concept of human development enables one to understand socioeconomic development,
changing values and democracy as distinct but related facets of the same principle. Economic
development, changing values and democracy work together to expand human choice.
Economic development brings urbanization, social mobilization and occupational
differentiation. This increases social complexity and multiplies social transactions between human
beings (Bendix 1964; Durkheim 1988; Simmel 1984; Blau 1994). These effects tend to emancipate
people from clientelistic ties and replace vertical authority relations with horizontal bargaining
relations. Individuals gain autonomy and resources become increasingly individualized (Dworkin
1988). Moreover, socioeconomic development provides people with greater physical and intellectual
resources by increasing incomes, skills and information levels and by establishing the welfare state
(Boix 2000). Socioeconomic development reduces restrictions on human autonomy, providing the
objective means that enable people to pursue self-determination. This view is as old as Aristotle and
has been argued from Adam Smith and Karl Marx to Esping-Andersen (1990) and Sen (2000). In
short, socioeconomic development provides individual resources that constitute the meanscomponent of human choice.
7
Cultural change is the second key subprocess relevant to human choice. Growing individual
resources enable people to move beyond a narrow focus on obtaining the means to survival, leading
people to place increasingly high priority on self-expression and human choice. Indeed, growing
emphasis on self-expression is the central motivating force that leads people to demand broader
choice. For choice is not only a matter of one’s means but also of one’s mind and motivation
(Rokeach 1960).
Democracy represents the institutional component of human choice, providing a legal structure
that guarantees fundamental individual rights in a society’s private and public life. Democracy
provides effective rights to human choice and thus represents its rules-component. This notion can
be traced to Mill and Dewey, who saw legal guarantees for “individual self-development”
(Macpherson 1977:44-76) as the core value of democracy.
Individual resources, self-expression values and effective rights are the three components of
Human Development and represent its means-, motives- and rules-components. These components
become increasingly widespread through the processes of socioeconomic development, cultural
change and democratization, respectively. Table 1 summarizes this conception of Human
Development.
Table 1.
The Concept of Human Development
Components of Human Development:
Economic
Component
Cultural
Component
Political
Component
Individual
Resources
Self-Expression
Values
Effective
Institutionalized
Rights
Underlying Processes
Socioeconomic
Development
Cultural Change
Democratization
Spheres where
components manifest
themselves
Means
Motives
Rules
Components
Prevailing Causal
Direction
Underlying theme
Means-Motives Linkage
Motives-Rules Linkage
Growth of Institutionalized Human Choice
8
The three components of Human Development all reflect the underlying theme of growing
human choice. Progress in any of the three aspects widens human leverage in societies, giving people
greater means, stronger motivations and wider institutional guarantees, enabling them to make full use
of their personal potential and to develop individual creativity. Human Development of societies
brings growing human choice on a mass-level.
Human Development is not a teleological concept. It does not imply that its three
subprocesses necessarily proceed in a linear upward direction. Societies can move in either
direction, progressing or regressing. But this concept does imply that changes in means, motives and
rules tend to coincide among societies, either narrowing or widening the range of human choice.
The concept of Human Development goes beyond standard modernization theory in being
both more comprehensive and more specific. Usually, theories cannot maximize comprehensiveness
and their degree of specificity at the same time, but the concept of Human Development does. On
one hand, it is comprehensive in that it integrates major changes in socioeconomic structure, political
culture, and regime institutions into a common theoretic framework. On the other hand, this concept
is specific because it concentrates on one well-defined theme: the growth (or decline) of human
choice.
1.2
The Two Linkages of Human Development
We suggest that the Human Development syndrome is shaped by two linkages: (1) a meansmotives linkage that connects self-expression values with individual resources; and (2) a motivesrules linkage that ties effective rights to self-expression values. Let us briefly outline how these
linkages function.
The Means -Motives Linkage: People’s value orientations reflect the restrictions that their
social conditions put on human autonomy. Usually, people tend to adapt their aspirations to the
restrictions posed by the environment (Schwartz 1992; Diener et al. 1995; Cummins 2000;
Eckersley 2000; Schmuck, Kasser & Ryan 2000). This mechanism, known in social psychology as
“aspiration adjustment” (Costa, McCrae & Zonderman 1987), emerged through human evolution,
because it was conducive to survival (Birch & Cobb 1981; Doyal & Gough 1991; Tooby &
Cosmides 1992). Aspiration adjustment leads people to aspire to attain the most pressing things first
and to avoid wasting energy on unattainable goals (Maslow 1974). Although the search for
sustenance may come first, all humans also possess intellectual abilities and the potential to develop
9
personal creativity. Indeed, the striving for self-expression seems deeply embedded in human
motivations (Schwartz 1992; Schmuck, Kasser & Ryan 2000), since self-expression brings
satisfaction as has been argued by writers ranging from Karl Marx to Maslow (1974) to Inkeles
(1983) to Flanagan (1987). Evidence from the World Values Surveys provides strong empirical
support for the view that having relatively broad choice for self-expression, increases human life
satisfaction. In each of 148 national representative surveys, conducted in diverse societies ranging
from Uganda to China, Iran, Brazil, Sweden and Poland, there is a strong correlation between
people’s perception of how much choice they have in shaping their live, and their level of life
satisfaction.1
Nevertheless, people adapt their emphasis on self-expression to the restrictions posed by their
environment. Under conditions of economic deprivation, people lower the priority given to selfexpression, although this downward adjustment of aspirations has psychological costs, diminishing
life satisfaction. A downward adjustment of aspirations may necessary to survive under restrictive
conditions, such as those in low-income countries, where scarcity drives people into a struggle for
survival. This may bring a Hobbesian “homo hominem lupo” situation in which outsiders are
distrusted as hostile competitors for scarce resources. Distrust towards outsiders can force people
into a rigid in-group discipline that restricts human autonomy, such as Banfield (1958) found in parts
of Southern Italy. Putnam (1992), reaches similar conclusions in his analysis of the differences
between Italians of the affluent North and the poor South, finding that Southerners tend to distrust
their fellow citizens and support rigid community discipline. They emphasize social control, public
order, hierarchy, moral rigidity and strong authority—survival values that tend to prevail under
conditions of insecurity.
Inglehart (1997) has demonstrated on a global scale that the publics of poor societies are
much more likely to emphasize survival values than the publics of societies with high levels of rich
countries. High levels of resources provide greater human autonomy, reducing the need to focus on
1
Life satisfaction in the World Values Surveys is measured on a 10-point rating scale (as noted in
footnote 5 of Table 2). Choice perception, too, is measured on a 10-point rating scale based on
the following question (V82): “Some people feel they have completely free choice and control
over their lives, while other people feel that what they do has no real effect on what happens to
them. Please use this scale where ‘1’ means ‘none at all’ and ‘10’ means ‘a great deal’ to
indicate how much freedom of choice and control you feel you have over the way your life turns
out.” In each of 148 of a total of 178 surveys where these variables can be created, there is a
highly significant positive correlation. The average within-survey correlation points to .35. Pooled
across all surveys the individual level correlation is .42. At the aggregate level of nations the
correlation is .85. The individual level correlation holds even controlling for financial satisfaction.
10
survival and making it possible to place more emphasis on self-expression. Accordingly, relatively
rich societies show higher levels of interpersonal trust, greater tolerance of diversity, higher life
satisfaction, more emphasis on political participation and greater respect for individual autonomy as
opposed to community authority. Flanagan (1987) has argued that this value change reflects a
functional mechanism of aspiration adjustment at the societal level.
The Motives-Rules Linkage: Growing emphasis on self-expression leads to increasing
demands for civil liberties, freedom of speech and political rights—which provide institutional
guarantees for the kinds of private and public activities that self-expression requires.
Growing mass emphasis on self-expression brings quite different consequences for autocracies
and democracies. It tends to undermine the legitimacy of autocracies that restrict human rights. This
makes authoritarian rule increasingly ineffective and costly, since the regime must bear the growing
costs of “aspiration suppression” (Kuran 1991). The exhaustion of a regime’s moral resources
increases the probability of an intra-elite division in which one faction of the elite may split off in an
attempt to regain legitimacy by liberalization and institutional reforms (Bova 1991; Przeworski 1992).
Under the right circumstances (see Foweraker & Landman 1997; Welzel 1999), this can mobilize
mass demands for democracy that eventually overthrows the authoritarian regime.2
The impact of growing emphasis on self-expression is quite different in democracies, where it
does not bring regime change, but an increasing effective type of democracy. In formal democracies,
basic civil and political rights may be guaranteed on paper, but legal codification does not necessarily
make these rights effective. Formal democracies can, in fact, can be ruled by corrupt elites who
make a mockery of formal rights. In low income societies dominated by survival values, the bulk of
the citizenry has neither the resources nor the motivation to exert effective pressures to realize their
constitutional rights. In economically developed societies, where self-expression values are
widespread, people tend to be both more able and more willing to exert effective public pressures to
2
Indeed, using data from the third and fourth waves of the World Values Surveys, there is a .29
individual level correlation between respondents’ emphasis on self-expression and their
preference for democracy (using Klingemann’s index of democratic regime support, see
Klingemann 1999). At the aggregate national level, the correlation is .66. Interestingly, however,
mass support for democracy is a much weaker predictor of both constitutional and effective
democracy (explanation of these variables in section 1.3) than is mass-emphasis on selfexpression. This indicates that overt support for democracy entails a good deal of fashionable lip
service, while emphasis on self-expression values reflects values that are deeply rooted every
day life. Eckstein (1966) has argued that such values are particularly crucial to the functioning
of democracy.
11
keep the elites accountable. As Verba, Nie and Kim (1978:73) have pointed out, “in all nations,
citizens appear to convert socioeconomic resources into political involvement.”
In democracies, elites are chosen by the general public. Thus, changing values among the
citizens will affect the elites as well. Increasing emphasis on self-expression values among the masses
should tend to bring greater emphasis on self-expression among the elites. Evidence from the World
Values Surveys supports this assumption, demonstrating that, in virtually every society among the 65
countries studied, people with university degrees place greater emphasis on individual self-expression
than does the average citizen (Welzel 2002). But the cross-national differences in emphasis on selfexpression are just as large among people with university degrees as they are among the average
citizens. Since democratically elected elites are overwhelmingly recruited from people with university
education, this finding indicates that a society’s level of emphasis on self-expression is reflected
among its elites.
Growing emphasis on individual self-expression among elites implies that these elites
themselves will tend to condemn repressive and corrupt government. Thus, a societal shift towards
self-expression values tends to reduce corrupt elite behavior and increase the effectiveness of
political rights for two different reasons. One is because in more affluent, self-expression societies,
the masses have more resources and stronger motivations to bring the elites under democratic
control. The other reason is that in these societies, the value systems of the elites themselves give
increasing legitimacy to citizen rights. From a rational choice perspective, there is no reason to expect
that elites will refrain from maximizing their incomes through corruption, unless their own values or
popular pressure eliminate this option.
Now consider now the reverse relation: does the mere presence of civil and political rights
create self-expression values among the citizens? The answer is No-- not unless the resources that
support these values are present. For example, consider India: though it has offered its citizens a full
array of civil and political rights for more than 50 years of constitutional democracy, the Indian public
did not develop a correspondingly strong emphasis on self-expression, as is demonstrated by India’s
location in Figure 1. On the other hand, although Czechoslovakia provided its citizens much
narrower formal rights under four decades of communist rule, both the Czechs and, to a lesser
degree the Slovaks, developed much stronger emphasis on self-expression values than the Indian
public as a whole did—reflecting their much higher levels of income and education (see Figure 1). A
strong emphasis on elf-expression values seems to reflect high levels of economic development, not
the presence of democracy. Constitutional democracy can exist without self-expression values, as
12
the Indian example shows. But if so, constitutional democracy is likely to be ineffective, as the
massive corruption of Indian government illustrates. To be effective, formal rights need the support of
a corresponding value system, which does not come automatically. As we will demonstrate, effective
democracy is a consequence rather than the cause of mass emphasis on self-expression.
1.3
Constitutional Democracy and Effective Democracy
The Indian case illustrates how important it is to differentiate between formal democracy and
effective democracy. India is without doubt a formal democracy, since its constitution guarantees
basic civil and political rights. But as Heller (2001) has pointed out, most Indians do not have
sufficient resources to exert their rights effectively. Moreover, their values do not motivate them to
give high priority to exercising their formal rights. In terms of effective democracy (in contrast to
formal democracy), Indian ranks closer to China than to Japan (see Figure 1), although India has a
democratic constitution.
Democracy is central to Human Development because it grants civil and political rights to the
citizens. The formal rights enshrined in laws and constitutions are a necessary element of effective
democracy; without formal democracy there can be no effective democracy. But the presence of
formal democracy does not necessarily bring effective democracy. Whether or not these rights are
effective depends on the extent to which elites respect them in their actual behavior. Law-abiding
elite behavior or what we call “elite integrity” is a central aspect of the “rule of law,” that, as Linz and
Stepan (1996), O’Donnell (1999) and Diamond (2000), have claimed, distinguishes effective
democracy from mere formal democracy or constitutional democracy. Hence, our operational
measure of “effective democracy” is the product of constitutional democracy and elite integrity, using
elite integrity as a weighting factor (see section 2.1).
1.4
Do National Aggregates of Mass Values Make Sense?
The first major linkage in the causal chain of Human Development is the means-motives linkage, from
resources to self-expression values. This linkage originates at the individual level and then impacts on
the societal level through mass-accumulation. Thus, societies with relatively high levels of resources,
give relatively high emphasis to self-expression values; when self-expression values have become
widespread among the public, these values become conducive to effective democracy at the societal
level.
13
There is considerable empirical evidence that this means-motives linkage originates at the
individual level: consistently, we find that individuals with high levels of resources place significantly
stronger emphasis on self-expression values than does the rest of the public. In fully 112 surveys
available from the World Values Surveys we find a significantly positive correlation between
people’s income and education, and their emphasis on self-expression values—the average
individual-level correlation being .29 (standard deviation: .08).3 As one would expect, we find a
considerably stronger correlations at the societal level, where the average correlation is .91 (see
section 2.3).
Although the correlation is much stronger at the societal level than at the individual level, the
growth of individual-level resources plays a crucial role in the shift toward stronger mass-emphasis
on self-expression. A “central tendency effect” at the national level systematically exaggerates the
relative strength of societal-level correlations, overshadowing the importance of the individual-level
process in contributing to the rise of self-expression values at the societal level. This occurs in the
following manner.
Common communications systems and shared national experiences lead to the diffusion of
relatively similar values within given nations, than between them. This tendency is reinforced by the
fact that there are much larger economic differences between the nations of the world today, than
with the regions of given countries. Thus, the difference in mean income between the richest state in
the U.S. and the poorest state, is about a 2:1 ratio. By contrast, there is a ratio of about 400:1
between the GDP/capita of the richest country and the poorest country in the world. Consequently,
there is much larger variation in value orientations between citizens of different countries, than among
citizens of the same country. Although emphasis on self-expression varies considerably between
individuals within nations, within each country a majority falls near their nation’s mean emphasis on
self-expression: most people’s emphasis on self-expression is relatively close to that of the average
citizen. But the national averages vary enormously from nation to nation (see Figure 2, below).
Hence, the nations’ average emphasis on self-expression captures fully 40 per cent of the total
individual level variance among all people surveyed in the four waves of the World Values Surveys.
3
Income and education are measured as an interaction term between individuals’ years of
schooling and their financial income, measured in deciles of national currencies. The correlation
between this interaction term and individuals’ emphasis on self-expression (see section 2.1 for
measurement of this variable) is significantly positive in 112 of a total of 120 surveys (94%)
where these variables could be created. Using years of schooling and financial income as sole
correlates, the same applies 95% and 94% of the surveys, respectively.
14
This is a remarkably large proportion, with a mere 73 national units accounting for nearly half of the
variance among 158,802 individuals (this is more than 2,000 times the random likelihood).
The importance of this central tendency effect at the societal level becomes evident when we
consider another regularity in social sciences: variations in an independent variable x do not translate
in a deterministic manner into corresponding variations of a dependent variable y. All known
relationships in social reality are probabilistic, having a “range of tolerance” within which variation in
x has no systematic effect on y. Only variations in x that exceed this range of tolerance are reflected
in corresponding variations of y. With this in mind, consider the relation between individual-level
resources and self-expression values, assuming (as we know to be true) that both variables have
relatively concentrated distributions within nations. This implies that only a minority of individuals will
show substantial deviations from the national average on both variables. Most individuals will fall
within the range of tolerance range within which variations in resources are not necessarily reflected
in corresponding variations in self-expression values. This results in relatively small individual-level
correlation within nations. But across nations, there are large numbers of individuals whose resources
levels deviate so greatly from the international mean, that they exceed the range of tolerance beyond
which substantial variations in self-expression values occur. In this case, the national-level correlation
between resources and self-expression values would be much stronger than the individual-level
correlation, even if this correlation were entirely due to individual-level processes. It is crucial to
bear this in mind. Central tendency effects tend to obscure the importance of individual-level
processes, giving rise to national-level correlations that dwarf the individual-level correlations, even
when the underlying causal process is taking place entirely at the individual level.
Central tendency effects at the national level are common. They are produced by social
processes that affect national populations relatively uniformly, but vary greatly between nations. The
growth of individual-level incomes is typical. Germany is one of many countries in which most
people’s income lies relatively close to the national average; this was true 40 years ago and it still
holds true today. Nevertheless, during the past 40 years the Germans’ average real income
quadrupled, with little change in the relative distribution of incomes (Zapf & ?). Income growth is a
process that affects national populations relatively uniformly, although it varies tremendously between
nations. As Landes (1998:xx) has demonstrated, 200 years ago the income ratio between the richest
and the poorest nation in the world was approximately 5:1. But, because of uneven economic growth
during the past 200 years, this ratio has risen enormously, reaching 400:1 today. Consequently, any
15
effect linked with differences in income, such as the rise of self-expression values, will be far more
pronounced between nations than within them.
If there are strong individual-level correlations across nations, then there must also be strong
correlations at the aggregate national level. Usually, these aggregate correlations are even stronger
than the cross-national individual level correlations, since aggregation reduces the measurement error
at the individual level. Survey data in particular contain a large component of measurement error at
the individual level: many respondents give erratic answers that reflect “non-attitudes,” producing a
good deal of random variation in survey data (Converse 1970). As Yule and Kendall (1950) and
Blalock (1961) have pointed out, the variation in a variable consists of a systematic and a random
element. Thus, the correlation between two variables x and y also consists of a systematic term and a
random term that diminishes the systematic correlation (“attenuation effect”). But when x and y are
averaged across nations, the random elements offset each other: random negative and positive
deviations from the mean cancel each other out (Page & Shapiro 1993:40). Following the law of
large numbers, this reduction of error becomes stronger as the number of individuals being
aggregated rises. Consequently, the random term becomes smaller, and the systematic correlation
larger, at higher levels of aggregation. When this is the case, aggregation does not exaggerate what is
taking place at the individual level-- it reveals the “real” correlation.
Aggregate national-level correlations reflect the existence of corresponding individual-level
correlations across nations. To conclude that these correlations exist is not an “ecological fallacy”
(Alker 1969). An aggregate correlation between variables at the nation level cannot exist unless there
is a corresponding correlation among individuals across nations. The ecological fallacy only applies
when one assumes that they point to the existence of individual-level correlations within nations
(Robinson 1950). These individual-level correlations can be weak or insignificant, even when they
are strong across nations. But this does not mean that there is no real individual-level correlation in
the pooled cross-national sample, since the central tendency effect at the national level, compounded
by the effects of measurement error, can obscure the importance of the underlying individual-level
process. When this is true, aggregating survey data to national-level averages, far from being
problematic, helps solve the problem.
Aggregation captures genuine characteristics of the nation. The central tendency of a nation’s
level of self-expression values reflects a completely contextual feature that is exogenous to every
individual in that nation, even though the nation’s mean emphasis on self-expression is calculated
from individual-level data. Aggregate characteristics are exogenous to individuals, reflecting
16
contextual constants at the societal level. This means that it is appropriate to use nationally
aggregated self-expression values in analyzing their impact on democracy—for the nation-level is
where these values become linked with democracy.
The linkage between mass-level emphasis on self-expression and effective democracy reflects
a relation between two different kinds of societal characteristics. Mass-level self-expression values
represent an aggregate characteristic of societies that accumulates from the individual level. By
contrast, effective democracy is a system characteristic that has no variation at the individual level.
This difference between aggregate and system characteristics makes the question of their relationship
all the more interesting, because it reflects an important individual-system relation. This individualsystem relation focuses on the linkage between the citizens and their regime— a crucial relationship
from the perspective of democratic theory. When an individual-system linkage exists, it must
necessarily be reflected in the relationship between an individual-level variable, such as selfexpression values, and a system variable, such as effective democracy. Human Development theory
focuses on such linkages.
2. Analyses
2.1
Data Sources and Measurement
Self-expression values are the bridging element in the means-motives-rules chain of linkages that
constitutes Human Development. To measure these values at the societal level we use the broadest
available data base, the World Values Surveys, which now cover 73 countries representing 80 per
cent of the world’s population.4 We measure self-expression values using factor scores summarizing
several attitudes that Inglehart and Baker (2000) developed as indicators of self-expression values.
We replicate their measure, but we use (for the first time) all four waves of the World Values
Surveys conducted between 1981 and 2001. The results of the factor analyses we used to create
this measure are reported in Table 2.
Table 2.
4
The Dimension of Self-Expression Values
Ronald Inglehart is the principal coordinator of the World Values Surveys. Data from the first
through the third wave of the World Values Surveys can be obtained from the Interuniversity
Consortium for Political Research (ICPSR) as study-number 6160. Data from the fourth wave
are not yet in the public domain. More detailed information on questionnaires, methods and field
work can be obtained from the World Values Study Group’s homepage:
“http://wvs.isr.umich.edu.”
17
Levels of Analysis:
Variables:
Individual level within
nations (mean loadings)
Individual level across
nations (pooled loadings)
Aggregate cross-national
level (pooled loadings)
.47
.68
.82
.45
.65
.87
.54
.59
.82
.34
.47
.64
.13
.44
.76
–.29
–.37
–.41
23%
29%
54%
158,803
individuals
137
national surveys
Strong self-expression values
imply:
-
Tolerance of human diversity1
2
Inclination to civic protest
Liberty aspirations
Trust in people
3
4
-
High life satisfaction
-
6
Weak religiousness
5
Weak self-expression values imply
the opposite.
Explained variance
Number of cases
137
national
surveys
Notes: Entries are factor loadings. Explorative principal components analysis (extraction of factors with
Eigenvalues above 1 adviced), no rotation. Source: World Values Surveys I-IV.
1
“Not mentioned” for “disliked neighbors” coded “1” and dichotomized against 0; scores added for
neighbors
with AIDS (V58) and homosexual neighbors (V60).
Aggregate data are national averages on this 0-2 scale.
2
“Already done” for “signing petitions (V118) coded “1” and dichotomized against “0.”
Aggregate data are national percentages already done.
3
Respondents’ first and second priorities for “giving people more say in important government decisions”
and “protecting freedom of speech” (V106-107) added to a four-point index, assigning 3 points for both
items on first and second rank, 2 points for one of these items on first rank, 1 point for one of these items
on second rank and 0 for none of these items on first or second rank.
Aggregate data are national averages on this 0-3 scale.
4
Respondents believing “most people can be trusted” (V27) dichotomized as “1” against “0.”
Aggregate data are national percentages of people trusting.
5
10-point rating scale for life satisfaction from WVS (V65).
Aggregate data are national averages on this 1-10 scale.
6
“How important is God in your life?” (V190). 10-point scale (1: not at all, 10: very important).
Aggregate data are national averages on this 1-10 scale.
Self-expression values represent an attitude that, following Rawls (1993), can be described as
a “rational” or reciprocal emphasis on individual autonomy: people who give high priority to their
own autonomy want this autonomy to be respected by others, because others are unlikely to respect
one’s own autonomy if one does not respect theirs. Hence, it is rational for people who give high
priority to attaining autonomy for themselves, to respect others’ autonomy as well. In line with this
18
rational sense of reciprocity, self-expression values tap emphasis on one’s own self-expression and
acceptance of others’ self-expression as well. Thus, self-expression values entail “individual liberty”
attitudes, namely readiness to carry out civic protest (such as signing petitions)5 and liberty
aspirations6, as well as “social tolerance” attitudes, such as interpersonal trust and tolerance of social
diversity (see the footnotes in Table 2 for the operationalization of these variables). Self-expression
is a basic human goal, and its attainment enhances subjective well-being. Accordingly, high levels of
self-expression values are linked with relatively high levels of life satisfaction. Finally, as a survival
value, strong religiousness is negatively linked with the dimension of self-expression values.
As we would expect, the factor loadings increase systematically from the individual level within
nations to the pooled individual level to the aggregate level across nations. The reasons were
explained in the previous section: (1) the pronounced central tendencies within nations tend to bound
individuals’ value orientations within such a small range that the linkages between these orientations
does not become fully manifest until the much broader range of cross-national variation is taken into
account; (2) individual-level measurement error tends to be eliminated through aggregation, which is
another reason why the syndrome is most clearly manifest at the aggregate level. This is particularly
true of life satisfaction. High levels of life satisfaction are only weakly linked with the self-expression
values dimension, if one ignores individual level variation across nations. But across the pooled
sample of individuals from different nations, and (even more) at the aggregate national level, life
satisfaction is clearly linked with the self-expression values dimension: countries whose citizens show
greater tolerance of human diversity, who have a relatively strong inclination to engage in civic
protest, who have more pronounced liberty aspirations, who trust their fellow citizens more—these
people also show relatively high levels of life satisfaction and happiness. Happiness, in short,
flourishes in societies where the citizens place relatively great emphasis on self-expression.
In the following analyses, we use self-expression values from the second World Values Survey
(conducted in about 1990) as an independent variable to explain subsequent levels of effective
democracy in 1999-2000. And we use self-expression values from the third World Values Survey
5
6
As noted by Barnes and Kaase et al. (1979), signing petitions is a low cost form of civic protest.
Hence, a society with many people who sign petitions has a rich opportunity structure for low
cost protest. This in turn implies that there must be many people who invest the higher costs that
are necessary to create low cost opportunities for all.
Although these items are taken from the postmaterialism-scale (see fn. 3 in Table 2), we
distinguish them as “liberty aspirations” from other components of postmaterialism, such as
preferences for a “less impersonal society,” “beautiful cities,” and “a society in which ideas
count more than money.” This is argued in more detail by Welzel and Inglehart (2001).
19
(conducted in about 1995) as a dependent variable to be explained by prior variations in individuallevel income and education in 1990. This temporal ordering is used to ensure that independent
variables are measured prior to their presumed effects. However, not all of the 73 nations for which
at least one measure of self-expression values is available, participated in every wave of the World
Values Surveys. Thus, rather than eliminate a substantial number of countries from the analysis, we
chose to estimate missing data in one survey from existing data on self-expression values in another
survey. This enables us to analyze the entire set of countries covered by the World Values Surveys,
and the disadvantage of using estimated data in some cases is not severe, in this case, because value
change proceeds rather slowly. Consequently, cross-national levels of self-expression values at one
point in time give a pretty accurate estimate of their levels at a slightly earlier or later point in time. To
be specific, the prediction error in regressing self-expression values in 1995 on self-expression values
1990 (or vice versa) is less than 10 per cent. This error range is only half as large as the error range
that is typical for the means-motives and the motives-rules linkages of Human Development. Hence,
we modeling these linkages by estimating missing survey data from existing survey data. The
appendix gives a detailed description how we estimated missing self-expression values in 1990 and
1995 from existing self-expression values in another wave of the World Values Survey.
The means-component of Human Development, individual resources, is measured using
Vanhanen’s (1997) “index of power resources.” This index combines measures of the nations’
physical and intellectual resources, and a measure of social complexity. We use Vanhanen’s most
recent version of this index that captures the early 1990s.7
This measure is preferable to single indicators such as per capita GDP. Sen (1997) has argued
that per capita GDP is an incomplete measure of a society’s human resources that only taps financial
7
Vanhanen creates three subindices. The subindex of “physical resources” is generated from the
share of family farms in the agricultural sector (weighted for the agricultural sector’s share in
GDP) and the deconcentration of non-agricultural resources (measured by 100 minus the share
in GDP generated by the state, foreign enterprises and large national trusts). The subindex of
“intellectual resources” is measured by the number of students per 100,000 inhabitants and the
literacy rate. The subindex of “occupational diversification” (“social complexity” in our
terminology) is produced from the proportion of the urban population and the percentage of the
non-agricultural work force. All component variables are standardized before they are combined
to the subindices. The three subindices are each combined additively from their component
variables, assuming that each subindex represents an own dimension. The same assumption then
leads to a multiplicative combination of the three subindices to create the overall index of
individual resources. This index is standardized to 100 as the maximum. For a detailed
description of scale construction see Vanhanen (1997:42-63) and the appendices of his book for
extensive documentation of data sources.
20
income, excluding other important resources such as education. Some of the OPEC countries show
high levels of per capita GDP, but relatively low levels of education. Using a combined measure of
individual resources, these countries show intermediate scores—which seems to give a more
accurate picture of the underlying reality. Moreover, GDP does not measure the distribution of
income, even though from the viewpoint of democratic theory (Dahl 1973:chapter 4; Muller 1997), a
relatively equal distribution of resources is an important precondition for the functioning of
democracy. Most Latin American countries, for instance, show a more uneven distribution of income
than Asian countries at a similar level of per capita GDP. Finally, social complexity is important for
the individualization of resources, bit it also is not tapped by GDP.
For these reasons, the Human Development Index has been used to provide a more complete
measure of individual resources (Human Development Report 2000). This index has its own
limitations: one of its three components, life expectancy, shows little variation among societies below
and above a range between 1,000 and 5,000 US-$ per capita (Hughes 1999:98). By contrast, the
Vanhanen-index captures all aspects of individual resources that are relevant from our theory of
Human Development. It measures both physical and intellectual resources. And it measures not only
levels but also the distribution of these resources. Finally, it includes a measure of social complexity.
Although this index is partly based on estimates, its empirical validity is strong: it is more closely
linked with effective democracy (r=.86, N=99) and self-expression values (r=.90, N=73) than are
per capita GDP (.84 for both variables) and the Human Development Index (.69 and .75
respectively). The Vanhanen-index covers some important aspect of reality that are not captured by
other indices. Both theoretical and empirical considerations point to using the Vanhanen-index as the
most valid measure of individual resources.
The rules-component of Human Development-- effective democracy-- is measured by the
combined Freedom House scores for civil and political rights and estimates of elite corruption from
Transparency International.8 The Freedom House scores range from 1 to 7 on each of the two
scales, with 1 indicating the highest and 7 the lowest level of freedom.9 We reversed this scale so that
higher figures indicate a broader scope of freedom rights. The scores from Freedom House are
8
9
See Elkins (2000) who provides convincing theoretical reasons, plus empirical evidence, that
Przeworski and Limongi’s (1997) pleading for a dichotomous classification of democracies vs.
non-democracies is flawed and that continuous measures of democracy are preferable.
The Freedom House scores can be obtained from the homepage of Freedom House:
“http://www.freedomhouse.org.” For a description of the estimation process and scale
construction, see Freedom in the World (1996:530-535). On the validity of these indices
compared to other democracy scales see Bollen and Paxton (2000).
21
expert judgments that estimate the scope of given rights in a society. We interpret these estimates as
a measure of formal democracy that is a necessary but insufficient element of effective democracy.
We use the most recent scores from 1999-2000 in order to make sure that our measure of formal
democracy is temporally subsequent to self-expression values as its predictor.
The Freedom House scores are imperfect measures of citizen rights. They neglect the extent to
which given rights are respected by actual elite behavior. This problem can be solved using the
corruption perception indices from Transparency International (see Rose 2001 for a similar
argument).10 These scores are also based on expert ratings—in this case, ratings of how corrupt the
political, bureaucratic and economic elites of a given country are. One indication of the validity of
these estimates is that they strongly correlate with citizens’ perception of elite corruption in
representative surveys (Rose 2001). The Transparency International scores range from 1 to 100,
with 100 indicating the greatest amount of corruption. Reversing these scores, one obtains a measure
of “elite integrity.” As argued in the previous section, we operationalize effective democracy as the
interaction between formal democracy and elite integrity. In this conception elite integrity is the
weighting factor that makes formal rights effective. Since elite integrity is a weighting factor and not a
compensating factor, we calculate the product of the reversed and combined Freedom House scores
(standardized to 10 as the maximum) and elite integrity (also standardized to 10). This produces an
index of effective democracy from 0 to 100 per cent. Since we use the most recent Transparency
scores from 1999, we obtain a measure of effective democracy in 1999-2000. As can be shown by
a two-dimensional plot (not documented here), this measure is much more restrictive in assigning a
high level of democracy than are the combined scores from Freedom House (i.e., our measure of
formal democracy).
In summary, we have constructed measures of each society’s level of (1) individual resources,
(2) self-expression values and (3) effective democracy. These measures have been obtained from
completely different sources. Given that these measures may contain considerable measurement
error, the fact that we find strong systematic relationship between them, points to the conclusion that
each of them captures a robust aspect of reality.
10
Data and methodological report can be obtained from Transparency International’s homepage:
“http://www.transparency.org.”
22
2.2
Operationalizing Cultural Zones
In order to test the general applicability of Human Development theory, we must ask: “Are the
linkages between the three components of Human Development universal in their applicability across
cultural zones?” If the linkages of Human Development only applied within specific cultural zones,
Human Development could not be viewed as a general theory.
Weber (1958), Eisenstadt (1986), Castles (1996), Huntington (1996) Inglehart (1997) and
others have found that nations cluster into larger units that might be called “country families,” “cultural
zones,” or “civilizations.” Nations belonging to the same cultural zone tend to share similar
worldviews, institutional traditions and patterns of economic subsistence. Thus, cultural zones can be
considered as supra-national units of diffusion that reflect similar patterns of societal development.
Three variables have been considered as determinants of cultural zones: historical traditions,
especially (1) common religious roots and (2) common imperial legacies, and (3) region or vicinity, a
factor that facilitates diffusion between nations (Kopstein & Reilly 2000).
Each of the gray shadowed areas in Table 3 outlines a group of nations having a distinct
cultural zone. The first criterion underlying this classification is religious tradition, which shows 18
countries with a historically Protestant tradition or with Protestants as the largest religious group; 27
Catholic countries; 10 Christian Orthodox countries; 10 Islamic countries; and a residual category of
5 countries having an “Asian” religion, such as Buddhism, Hinduism or Confucianism (a quasireligion).
These five religious groups were subdivided according to region or imperial legacy, provided
there were enough cases to permit such a division. Thus, the Catholic countries were divided into the
zones of “Catholic Western Europe,” “Catholic Eastern Europe” and “Latin America.” The division
between Western and Eastern Europe reflects whether a country belonged to the Soviet communist
empire or not. Latin America reflects not only a distinctive region but also an Iberian imperial legacy.
For some countries, additional decisions had to be made. Among the Asian countries, we saw
no criterion to group China and India together with other nations. According to Huntington (1996),
both of these countries, each with a population of more than one billion, represents a “civilization” of
its own. On the other hand, Japan, South Korea and Taiwan share a Confucian tradition and have in
common the fact that they are economically advanced. So we summarized them as “Developed Far
East.” Moreover, Estonia and Latvia, though having a Protestant tradition, were grouped with the
Catholic Eastern European countries, with which they share the legacy of Soviet controlled
23
communism and the tradition of “Western Christianity,” as opposed to Orthodox Eastern
Christendom (Huntington 1996:159). Finally, the Sub-Saharan countries have not been divided on
the basis of religions. Although there are Christian and Islamic influences in Sub-Saharan Africa,
there remain specific pre-colonial Black African imprints, based on this region’s animist religious
roots and its distinctive ethnic make-up. This justifies classifying the Sub-Saharan countries as a
specific cultural zone (Huntington 1996).
Table 3.
The Location of the WVS-Nations within Cultural Zones and Regions
RELIGION:
Western Christian
REGION:
Western
Europe
Scandinavia
Protestant
Catholic
Germany
Austria
Switzerland
Denmark
France
Finland
Iceland
Norway
Sweden
Italy
Malta
Portugal
Spain
Netherlands
Belgium
Luxemburg
Ex-British
Overseas
Great Britain
Australia
New Zealand
Canada
U.S.A.
Eastern Europe
Estonia
Ireland
Catholic
Central
Europe
Catholic ExYugoslavia
Middle East
South Asia
CATHOLIC
EASTERN
EUROPE
Orthodox
Lithuania
Mediterranean
CATHOLIC
WESTERN EUROPE
Benelux
British Islands
Armenia
ORTHODOX
EASTERN EUROPE
Trans-Caucasus
Azerbaijan
Georgia
Czech Republic
Hungary
Poland
Slovakia
Belarus
Moldova
Russia
Ukraine
Croatia
Macedonia
Slovenia
Yugoslavia
Bulgaria
‚Oriental’ ExYugoslavia
‘Asian’
Greece
Ex-Soviet
Slavic
ISLAMIC
ZONE
Bosnia-Herzegovina
Albania
Romania
Other Balcans
Philippines
Iran
Turkey
Middle East
Egypt
Jordan
Arab
countries
Bangladesh
South Asian
Islam
Far East
Islamic
German Tongue
PROTESTANT
WESTERN ZONE
Latvia
Baltics
‘Oriental’
INDIA
Pakistan
CHINA
Japan
South Korea
Taiwan
Sub-Saharan
Ghana
South Africa
Uganda
Zimbabwe
Nigeria
DEVELOPED
FAR EAST
24
Latin America
SUBSAHARAN
AFRICA
Argentina
Brazil
Chile
Peru
Uruguay
Venezuela
Dominican Republic
El Salvador
Mexico
South America
LATIN
AMERICA
Central
America/Caribean
Our classification is crude. Yet, it explains 85 percent of the variance in individual-level
resources, 84 percent of the variance in self-expression values and 83 percent of the variance in
effective democracy across 73 nations. Even this crude classification designates relatively
homogeneous zones.
In addition to this differentiation between “cultural zones,” we also constructed a more finetuned classification based on 24 smaller “regions,” such as Scandinavia, the Baltics, Transcaucasia,
Mediterranean Europe, Central America and so forth. These regions are indicated by the smaller
boxes in Table 3. The regional classification captures 93 percent of the cross-national variance in
individual resources, 92 percent of the variance in self-expression values and 91 percent of the
variance in effective democracy. Though the classification into regions contains more than twice as
many categories than the classification into cultural zones, it explains only 8 to 9 per cent more of the
cross-national variance, which confirms the adequacy of the cultural zones in Table 3.
2.3
The Human Development Linkages in Cross-Cultural Perspective
Nations, regions and cultural zones are part of a multi-level context, with nations nested inside
regions and regions inside cultural zones. The present question is Do the linkages of the Human
Development process translate from the national to the regional and from the regional to the cultural
zone level with or without “frictions?” (Goldstein et al. 1995) “Frictions” would manifest themselves if
the Human Development linkages show substantially varying intercepts and slopes at different levels
of aggregation. Does the process of Human Development simply aggregate from lower-level to
higher-level units? Or are there “frictions,” indicating that some additional factor intervenes as one
moves from one level to another? If that were the case, Human Development could not be viewed as
a general theory, but depend on the level at which it is observed.
As Table 4 demonstrates, the Human Development linkages do not substantially vary in either
their intercepts or slopes at different levels of aggregation. Whether at the national, regional or
25
cultural zone level, the intercepts and slopes remain virtually constant. The correlations, by contrast,
systematically increase with higher levels of aggregation, for the reasons explained above. This,
however, simply indicates that there is more random variation at lower levels of aggregation. Since
aggregation averages random variations out, correlations increase with higher levels of aggregation.
But neither the intercepts nor the slopes of the relationships vary at different levels of aggregation.
This finding indicates that Human Development is a general theory, operating in the same fashion at
lower higher levels of aggregation.
Another way to express this finding is to specify an integrated two-level model in which we
estimate an overall intercept and slope that are constant across cultural zones, together with the
intercept- and slope-variances for cultural zones (Goldstein et al. 1995). In this way we formulate the
relation between individual resources and self-expression values and that between self-expression
values and effective democracy. The levels of variation are indicated with suffix “j” for the national
level and suffix “k” for the cultural zone level. The “random slopes and intercepts model” is written as
follows:
SELFEXVALS 1995jk
= ß0k + ß1k * INDIVRESOUR1990jk + ejk
EFFECDEMOC99-00jk
= ß0k + ß1k * SELFEXVALS 1990jk
+ ejk
We can express the composition of intercept and slope as follows:
Intercept:
ß0k = ß0 + u0k
Slope:
ß1k = ß1 + u1k
26
Intercept and slope are each composed of a fixed part that is constant across cultural zones
(ß0, ß 1) and a variable part that differs across cultural zones (u0k, u1k). In addition, there is an error
term for the nations’ remaining variation (ejk). that can neither be attributed to the overall effects nor
to their variation for cultural zones.
Table 5.
The Linkages of Human Development in Integrated Multi-Level Models
(Standard Errors in Parentheses)
Components:
Means -Motives Linkage:
Motives-Rules Linkage:
SELFEXVALS1995jk
= ß 0k + ß 1k * INDIVRESOUR1990jk + ejk
EFFECDEMOC1999-00JK
= ß 0k + ß 1k * SELFEXVALS1990jk + ejk
Intercept, fixed
component
ß0
–1.539***
Intercept-variance
for Cultural Zones
v 0k
Slope, fixed
component
ß1
.060***
(.003)
29.279***
Slope-variance for
Cultural Zones
v 1k
.001
(.034)
67.233***
Random Variance
ejk
14.168
(.205)
Variance explained
by fixed effects
(1.483)
67.319
Number of cases
Meaning of suffixes:
2.198
(.091)
Intercept
Slope
Nation-Level
Cultural Zone Level
(1.908)
343.481*** (18.533)
15382.658
(2.044)
(8.199)
(240.354)
49325.885
71 nations in 8 cultural zones
0
1
j
k
39.416***
SELFEXVALS1995:
INDIVRESOUR1990:
EFFECDEMOC1999-00:
SELFEXVALS1990:
66 nations in 8 cultural zones
Self-Expression Values 1995
Individual Resources 1990
Effective Democracy 1999-00
Self-Expression Values 1990
As Table 5 shows, the overall effects of both the Means-Motives linkage and the MotivesRules linkage are highly significant, greatly exceeding their standard errors. We find significant effects
of individual resources on self-expression values; and of self-expression values on effective
democracy; that are independent of cultural zones. Compared to these overall effects, the intercept
and slope variances for cultural zones are negligible, failing to account for significant proportions of
the variance that remains unexplained by the overall effects. This can be seen calculating the share
that the intercept and slope variances have in the unexplained variance: only the intercept variance in
the model explaining self-expression values captures more than 2 per cent of the random variance,
but measured against its standard error, this remains insignificant.
27
Another way to model these findings, is to measure the effect of individual resources on selfexpression values, controlling for the cultural zone level of self-expression values. To do this, we
assign to each nation its cultural zone average in self-expression values and introduce this variable as
an additional predictor. We now are testing the extent to which a nation’s own emphasis on selfexpression is a function of the average emphasis found in its cultural zone, modeling each nation’s
emphasis on self-expression as a function of diffusion within cultural zones. In order to avoid a
tautological measure of cultural zone diffusion, we assign each nation the mean cultural zone level
calculated by excluding a given nation’s own value. Hence, we specify for each nation an
exogenous cultural zone which is crucial for the concept of diffusion. Similarly, we estimate the effect
of self-expression values on effective democracy controlling for the cultural zone’s level of effective
democracy.
The results of these regressions are shown in Table 6. As is evident, the impact of individual
resources on self-expression values, and the impact of self-expression values on effective
democracy, remains highly significant across nations, even controlling for diffusion within cultural
zones. This confirms our previous finding: the linkages of the Human Development process are
independent of the cultural zones within which they operate: the process applies universally It is
important to bear in mind that the cultural zones also show significant impact, accounting for a large
proportion of the variance in both linkages of Human Development. In fact, diffusion within cultural
zones captures 30 per cent of the effect of individual resources on self-expression values, and 40 per
cent of the effect of self-expression values on effective democracy. Why this is so, is demonstrated in
Figures 2 and 3. These figures show that cultural zones and Human Development are not competing
factors. Instead they are complementary factors.
Table 6.
The Linkages of Human Development Controlled
for Diffusion Within Cultural Zones
Dependent Variable:
Self-Expression Values 1995
Predictors:
Model 1.1
Individual Resources
1990
.056***
(.003)
Self-Expression Values
1990
————
Cultural Zone Level of
Dependent Variable
Constant
Model 1.2
————
.970***
(.060)
–1.493***
–.006
Dependent Variable:
Effective Democracy 1999-2000
Model 1.3
Model 2.1
Model 2.2
Model 2.3
.036***
(.007)
————
————
————
————
29.397***
(2.066)
.401**
(.120)
–.969***
38.941***
12.768**
(3.933)
.969***
(.061)
.593***
(.126)
1.159
16.470**
28
(.090)
(.060)
(.190)
(1.912)
(2.876)
(5.152)
Adjusted R2
.81
.78
.85
.75
.77
.82
N
73
71
71
68
72
66
Entries are unstandardized regression coefficients (standard errors in parentheses).
Significance Levels: *** p<.001 ** p<.010 *** p<.100
Figure 2 examines the linkage between resources and self-expression values (the “MansMotives” linkage). Figure 3 examines the linkage between Self-expression values and Democracy
(the “Motives-Rules” linkage). The upper plots in both Figures show the Human Development
linkages across nations; the lower plots show how they operate across regions and cultural zones.
We can see that both linkages are clearly manifest across nations, although the confidence interval
within which these linkages appear is relatively broad. This points to relatively large “ranges of
tolerance” within which the linkages are invisible. “Ranges of tolerance” can be examined in the
vertical dimension (y) or in the horizontal dimension (x). The range of tolerance in x, for instance, is
the horizontal distance between the left and the right boundary of the confidence interval. This
distance is constant for any value of y, since the confidence interval has parallel boundaries. What
does this range of tolerance imply?
29
Figure 2. The Means-Motives Linkage of Human Development Across Nations,
Regions and Cultural Zones
Sweden New Zeald.
Netherld.
Australia
Germany (W.) Denmark
Canada
Norway
Switzerld.
Finland
U.S.A.
G.B.
Belgium
Germany (E.) Austria
France
Iceland
Luxemb.
Ireland
95%-Confidence
Interval
y = -1.49 + 0.06 * x
R sq. = 0.81
1,5
1,0
S-Korea
0,5
Uruguay
Argentina
Italy
Japan
Spain
Mexico
Greece
Colombia
Brazil
Chile
Dominic. R.
China
Slovakia
Malta
Slovenia
Portugal
Ghana
Latvia
Hungary Peru Venez. Taiwan
Bosnia
S-Africa
Yugosl.
Poland
Bulgaria India
Uganda
Philipp.
Nigeria Romania
Turkey
Belarus
Macedon.
Georgia
Bangladesh
Range of 'tolerance' within which
El Salv.
Pakistan
change in x has no effect on y
Jordan
Egypt
Moldova Iran
Croatia
0,0
-0,5
-1,0
-1,5
Czech R.
Azerbaijan
-2,0
-05
00
Zimbabwe
05
10
15
20
25
30
35
40
45
50
55
60
Level of Individual Resources (about 1990)
2,0
Strength of Self-Expression Values (1995)
Strength of Self-Expression Values (1995)
2,0
y = -1.49 + 0.06 * x
R sq. = 0.87
1,5
Australia/NZ
Protestant Western Zone
Scandinavia
Catholic Western Europe
British Islands
German Tongue
1,0
Catholic Eastern Europe
Benelux
North America
0,5
Catholic Ex-Yugosl.
0,0
-0,5
Developed Far
East
Central Europe
China
SubSaharan
Africa
India
Balcans
-1,0
Mediterranean
South America
Latin America
Central America
Baltics
Oriental Ex-Yugosl.
South Asian Islam
Orthodox Eastern Europe
Middle East
-1,5
Arab
Trans-Caucasus
-2,0
-05
Slavic Ex-Soviet
05
15
Islamic Zone
25
35
45
Level of Individual Resources (about 1990)
55
30
Figure 3. The Motives-Rules Linkage of Human Development Across Nations,
Regions and Cultural Zones
Degree of Effective Democracy (1999-2000)
105
95
Denmark
Canada
Switzerld.
Iceland
Norway
Luxemb.
95%-Confidence
Interval
85
75
Spain
Estonia
S-Africa
45
Slovenia
France
Japan
Belgium
Chile
Taiwan
Hungary
G.B.
Germany (W.)
Portugal
55
New Zeald.
Netherld.
Australia
U.S.A.
Austria Ireland
65
Finland
Sweden
y = 38.94 + 29.40 * x
R sq. = 0.75
Czech R. Italy
Poland
Uruguay
Lithuania Latvia Greece
Slovakia
El Salv. Philipp.
25
S-Korea
Bulgaria
Romania
Brazil Argentina
Macedon. India
Ghana
Peru
15MoldovaJordan
Colombia
Venez.
Ukraine
Turkey
Range of 'tolerance' within which
Mexico
Zimbabwe
Uganda
Russia
change in x has no effect on y
Croatia
05
Albania
Belarus
Nigeria
Azerbaijan Egypt
Yugosl.
China
35
-05
-1,5 -1,3 -1,1 -0,9 -0,7 -0,5 -0,3 -0,1 0,1 0,3 0,5 0,7 0,9 1,1 1,3 1,5 1,7 1,9 2,1
Degree of Effective Democracy (1999-2000)
Strength of Self-Expression Values (1990)
105
Scandinavia
95
y = 37.80 + 32.44 * x
R sq. = 0.92
Benelux
85
Catholic Western Europe
75
65
Australia/NZ
Protestant Western Zone
North America
German Tongue
Catholic Eastern Europe
British Islands
Mediterranean
55
SubSaharan
Africa
45
35
Orthodox
E.-Europe
25
Trans-Caucasus
Central Europe
Catholic Ex-Yugosl.
Baltics
South America
Central America
Balcans
15
Arab
Middle East
05
India
China
Developed Far
East
Latin America
South Asian Islam
Oriental Ex-Yugosl. Slavic Ex-Soviet
-05
-1,5 -1,3 -1,1 -0,9 -0,7 -0,5 -0,3 -0,1 0,1 0,3 0,5 0,7 0,9 1,1 1,3 1,5 1,7 1,9 2,1
Islamic Zone
Strength of Self-Expression Values (1990)
31
Consider the upper plot of Figure 3: when one moves a horizontal distance from weaker to
stronger self-expression values that is smaller than the “range of tolerance,” it is unclear whether the
next society one encounters will have a higher level of effective democracy than the previous one.
Within this range, there is considerable random variation in levels of effective democracy. But when
the movement along the self-expression values dimension exceeds the range of tolerance, there is a
more than 95 per cent probability that the next society encountered will have a higher level of
effective democracy than the previous one. The same logic applies to the impact of individual
resources on self-expression values. Only when there is a substantial cross-national difference in
resource levels, can one have a high level of certainty that the next society you encounter will show
stronger emphasis on self-expression values than the previous one.
Now consider the lower plots in Figures 2 and 3. It is evident that the horizontal distances
covered by cultural zones are so small that they usually do not exceed the range of tolerance beyond
which the Human Development linkages manifest themselves. The range of variation among the
Scandinavian countries, for example, is too small to show substantial effects; even the broader
Protestant Western zone falls within the range of tolerance: the cultural zones cluster regions and
nations into relatively homogenous units. This finding makes the question of whether the Human
Development linkages manifest themselves within cultural zones more or less irrelevant. Since the by
far most of the variance is between rather than within cultural zones, the decisive question is whether
the Human Development linkages operate across these cultural zones.
The lower plots in Figures 2 and 3 leave little doubt that they do. The means-motives linkage
and the motives-rules linkage do work across regions and cultural zones. In other words, these
linkages are so pronounced across nations because they operate across the supra-national units that
integrate nations into homogenous groups. These effects are even more pronounced at the level of
supra-national units because at this level there is no more random variation among nations.
The Human Development syndrome is strikingly evident at the cross-cultural level. And it
accounts for even more of the variance than cultural zones: cultural zones capture about 85 per cent
of the cross-national variance in each of the three components of Human Development, but the
linkages between these components explain more than 90 per cent of the cross-cultural variation.
The Human Development linkages are not culture-specific but universal.
32
2.4
The Genesis of Human Development
Our theory argues that growing resources at the individual-level tend to shift a society’s value system
toward greater emphasis on self-expression. Growing mass-emphasis on self-expression then gives
rise to increasing demands for effective democracy at the societal level. The most debatable
assumption in this argument concerns the causal relation between mass values and democracy. A
number of writers have claimed that the causal relation between political culture and political
institutions can operate only in the opposite direction: democratic institutions can produce a prodemocratic culture but a pro-democratic culture has no impact on democracy (Rustow 1970; Muller
& Seligson 1994; Miller & Jackman 1998).
If this view is correct, mass-emphasis on self-expression must be a consequence of a nation’s
previous level of democracy, rather than the cause of its subsequent democratic performance. This
assumption contradicts our own theory, which emphasizes the opposite causal connection.
Fortunately, the two hypotheses can be tested with our data. In order to do so, we specified a
cross-national path model that starts from individual resources in 1990 and democratic traditions up
to 199511, continues with self-expression values as of 1995 and ends up with effective democracy in
1999-2000. If the Human Development model is correct, self-expression values should have a
stronger effect on subsequent democracy than prior democratic traditions have on self-expression
values. If the institutional determinism assumption is correct, the contrary should hold true.
11
The variable “democratic tradition” measures the number of years that a country has spent
under a democratic constitution. These years haven been counted from the beginning of a
nation’s independence (or from 1850 onward in case of countries that haven been independent
before 1850) till 1995. Countries that emerged from the dissolution of the Soviet Union and
Yugoslavia have been dealt like their former mother country as long as they belonged to it. A
year has been counted as one under democratic constitution, if a country obtained at least +7
points on the “Autocracy-Democracy” index from Jaggers and Gurr (1995). This index is based
on an analysis of constitutions considering of how many restrictions there are for executive
power and of how effective the influence of the electorate is in constituting government. Gurr
and Jaggers classify countries as “coherent democracies,” if they reach +7 or more points on
their –10 to +10 index. Data and methodological description can be obtained from the homepage
of the “Polity 98” project: “http://www.bsos.umd.edu/cidcm/polity.” We used these data here
because they reach farther back in time than the scores from Freedom House and are therefore
more adequate to measure democratic tradition in its length.
33
Figure 4.
Democracy as a Component of Human Development
H U M A N
D E V E L O P M E N T
in
socio-economic
sphere
in
political-cultural
sphere
in
legal-institutional
sphere
.31*
Individual
Resource
s (1990)
.81***
SelfExpressio
n Values
(1995)
.45**
.47**
.34*
Elite
Integrity
(1999)
.58***
Formal
Effective Democracy
Democra
cy(1999-2000)
(2000)
Years of
Democra
cy (till
1995)
TIME
Remarks:
Coefficients are standardized path coefficients. Bold arrows indicate strongest effect on respective dependent variable.
Interrupted arrows show insignificant effects. Fully recursive model, i.e. all effects from left-hand- to right-hand-side
specified. There is no overall fit for fully recursive models. Number of cases: N=68.
The path model in Figure 4 clearly supports the Human Development model. As it indicates,
while democratic traditions have no effect on self-expression values when we control for individual
resources, self-expression values do have a significant impact on subsequent levels of effective
democracy. This effect holds up when we control for individual resources and democratic traditions,
with the latter showing no significant effect on subsequent levels of democracy. The Human
Development linkages seem to have considerably more impact on a society’s level of effective
democracy than democratic traditions. This implies that rising levels of resources give rise to higher
levels of mass emphasis on self-expression values, which in turn tends to promote effective
democracy—even if there is a weak democratic legacy from the past.
34
Our path model gives further insight concerning how self-expression values impact on effective
democracy. To analyze this process, we decomposed effective democracy into its two components,
elite integrity and formal democracy. Controlling for all other effects in the model, elite integrity
proves to be the only factor that has a significant impact on formal democracy. This indicates that
elites are the prime force in shaping constitutional democracy. And indeed, the scope of civil and
political rights is determined by what the elites write into the laws and constitutions—a basic premise
of the elite approach (Schmitter & O’Donnell 1986; Gunther & Higley 1992; Casper & Taylor
1998). On another point, however, adherents of this approach seem to be wrong. Elite behavior is
not largely independent of mass preferences, as they assume; on the contrary, the extent to which
the public emphasizes self-expression has a major effect on elite integrity: in societies where the
public gives high priority to civil rights and participates actively in politics, the elites show high levels
of transparency. In other societies, they tend to be corrupt and authoritarian. The impact of mass
culture on effective democracy operates primarily through its impact on elite behavior. Unless the
public keeps tabs on them, elites will tend to do anything they can get away with. Mass emphasis on
self-expression tends to produce elites who extend or sustain constitutional democracy.
There is little reason to assume a reverse causation behind the relationship between selfexpression values and elite integrity. Elites may have leverage to influence public attitudes on
narrowly defined issues. But such basic and deeply rooted values as emphasis on self-expression,
tolerance of outgroups, interpersonal trust and participatory orientations, can not be simply created
by elite campaigns. Elites can appeal to such values but not create them. Moreover, even if they
could instill such values among the public, elites have an obvious self-interest in breeding a public that
is not attentive to, and highly critical of, their behavior—their lives are much easier if the public
doesn’t keep them under close control. From a rational choice perspective, elites have no reason to
avoid maximizing their incomes by engaging in corruption, unless there are severe restrictions on such
behavior. Growing mass emphasis on self-expression constitutes the strongest such restriction. It
translates into powerful mass pressure on corruption-seeking elites; and (since elites are recruited
from the public and tend to reflect prevailing norms), it gradually gives rise to elites whose own
beliefs reflect mass values, so that violating basic civil rights, political liberties and bureaucratic norms
increasingly becomes excluded from their calculations.
The fact that increasingly critical and activist citizens have emerged in recent decades, can
scarcely be attributed to elite manipulation. It reflects a cultural shift linked with high levels of
economic development, that causes mass publics to place growing emphasis on self-expression
35
values. The existence of merely formal laws against corruption and repression is not enough to avoid
them, if elite norms do not support them. One could cite almost countless cases in which the laws on
the books guaranteed responsive government and high levels of liberty, but were completely
ineffective. The most plausible reading of the evidence is that high levels of mass emphasis on selfexpression leads to elite integrity, and not the reverse.
Conclusion
Socioeconomic development, changing values, and democratization constitute a coherent syndrome
of social progress. Modernization theorists have so far failed to integrate these three components into
a coherent theory. The concept of Human Development, as introduced by Anand and Sen, has the
potential to provide an integrating theory, but it omitted the cultural component of the syndrome,
which helps explain why (and under what conditions) economic development gives rise to
democratic institutions. Human Development is an integrated syndrome, and the underlying theme of
its three components is the growth of human choice: socioeconomic development widens human
choice by enlarging people’s individual resources; cultural change increases mass emphasis on selfexpression values, which lead people to place greater emphasis on establishing an institutional basis
that secures human choice; and institutional change towards effective democracy extends human
choice by granting legal rights and keeping them effective.
Inglehart and Baker proposed a revised theory of modernization, showing that cultural zones
have a significant additional impact on self-expression values, beyond the effects of economic
development. We go one step further here, starting with individual resources and self-expression
values and adding effective democracy as a third component, arguing that these three components
constitute a comprehensive syndrome of Human Development. Furthermore, we have argued that
the role of cultural zones does not compete with the process of Human Development. Instead there
is a complementary interplay between the forces of Human Development and cultural zones, so that
the diffusion effects of cultural zones make Human Development more visible at the cross-cultural
level: Human Development is most evident at the cross-national level because it operates across the
cultural zones that integrate nations into homogenous units of diffusion.
The empirical evidence indicates that the syndrome of Human Development is shaped by a
causal sequence in which individual resources lead to self-expression values, which lead to effective
democracy. Democracy remains effective only in so far as it is supported by a mass culture that
emphasizes human self-expression. This culture needs a socioeconomic basis to take root in a
36
society. Effective democracy is much more an evolutionary phenomenon than something that can be
simply created through intelligent constitutional engineering. The emergence of effective democracy is
deeply embedded in mass-level changes, and it is closely linked to these changes by the
emancipatory logic of Human Development. Indeed, effective democracy is an inherent element of
Human Development.
The data used here are, of course, imperfect measures. The Vanhanen-index of individual
resources, the World Values Surveys that measured self-expression values, the political rights and
civil liberties scores from Freedom House, and the corruption estimates assembled by Transparency
International, all contain considerable measurement error. Despite these measurement errors,
however, (and despite the fact that these measures derive from completely different sources), we find
remarkably strong linkages between these variables—which suggests that they tap robust aspects of
social reality..
One might argue that because our data represent cross-sections, they allow for no dynamic
interpretation. We disagree with this critique. In some cases, cross-sectional data actually do point to
a dynamic interpretation. This is particularly true when one is dealing with evolutionary variables,
such as individual resources, self-expression values and effective democracy. Evolutionary variables
are variables whose short-term changes are so slow that they only account for a small proportion of
the levels that have been achieved. Annual changes in per capita GDP, for instance, constitute a
minor proportion of the GDP level that any given nation has accumulated so far. Short-term changes
in evolutionary variables are usually too small to exceed the range beyond which they have a
significant effect. To represent relevant changes beyond this range would require very long timeseries with a very large number of cases. By contrast, cross-sections represent differences that have
accumulated over many years, or even many decades. In this sense, cross-sections are a summation
of many small irrelevant changes, and aggregate them to the point where they go beyond the “range
of tolerance” and show important effects. In other words, cross-sections represent accumulated
change— and enable one to go beyond a merely static interpretation.
As we have shown, accumulated change in self-expression values has an effect on
accumulated change in effective democracy. This causal interpretation seems accurate, since selfexpression values themselves do not depend on the democratic tradition that has accumulated so far.
Hence, the linkage between democracy and mass-culture can not be interpreted as indicating that
effective democracy created self-expression values. It seems to reflect the final stage in a Human
37
Development sequence in which economic development leads to cultural changes that are conducive
to the emergence and survival of democratic institutions.
38
References
Almond, G.A. & Verba, S. (1963). The Civic Culture, Princeton: Princeton University Press.
Anand, S. (1993). Poverty and Human Development in Asia and the Pacific. In United Nations
Development Programme (ed.), Poverty Alleviation in Asia and the Pacific. New York:
UNDP Press, 1-39.
Anand, S. & Sen, A. (2000). Human Development and Economic Sustainability. World Development
28(12):2029-49.
Banfield, E.C. (1958). The Moral Basis of Backwardness, Glencoe: Free Press.
Barnes, S.H. & Kaase, M. et al. (1979). Political Action, Beverly Hills: Sage.
Barro, R.J. (1997): Determinants of Economic Growth. Cambridge: MIT Press.
Blalock, H.M. (1961). Causal Inferences in Nonexperimental Research, Chapel Hill.
Bell, D. (1973). The Coming of Postindustrial Society. New York: Penguin.
Bollen, K.A. & Jackman, R.W. (1985). Political Democracy and the Size Distribution of Income.
American Sociological Review 50 (August): 438-457.
Brint, S. (1984). ‘New Class’ and Cumulative Trend Explanations of the Liberal Political Attitudes of
Professionals. American Journal of Sociology 90 (1): 30-71.
Budge, I. (1996). The New Challenge of Direct Democracy. Cambridge: Polity Press.
Butler, D. & Austin, R. (1994). Referendums (2nd edition), Washington D.C.: American Enterprise
Institute for Public Research.
Burkhart, R.E. & Lewis-Beck, M.S. (1994). Comparative Democracy: The Economic Development
Thesis. American Political Science Review 88 (December): 903-910.
Chirot, D. (1986). Social Change in the Modern Era, Orlando: Harcourt Brace Jovanovich.
Coleman, J.S. (1968). Modernization: Political Aspects. In D.L. Sills (ed.), International
Encyclopedia of the Social Sciences (vol. 10). Washington D.C.: Free Press, 395-402.
Coleman, J.S. (1988). Social Capital and the Creation of Human Capital. American Journal of
Sociology 94 (Supplement):S95-S120.
Converse, P.E. (1970). Attitudes and Non-Attitudes. In E.R. Tufte (ed.), The Quantitative Analysis
of Social Problems. Reading, 168-189.
Costa, P.T., McCrae, R.R. & Zonderman, A.B. (1987). Environmental and Dispositional Influences on
Well-Being. British Journal of Psychology 78: 299-306.
Cronin, T.E. (1998). Direct Democracy (2nd edition), Cambridge: Harvard University Press.
Cummins, R.A. (1995). On the Trail of the Gold Standard for Subjective Well-Being. Social
Indicators Research 35: 179-200.
39
Cutright, P.C. (1963). National Political Development. American Sociological Review 28 (April): 253264.
Dahl, R.A. (1973). Polyarchy, New Haven: Yale University Press.
Dahl, R.A. (1998). Development and Democratic Culture. Journal of Democracy: 34-39.
Dalton, R.J. (1996): Citizen Politics, Chatham: Chatham House.
Diamond, L. (1992). Economic Development and Democracy Reconsidered. In G. Marks & L.
Diamond (eds.), Reexamining Democracy. London: Sage, 93-139.
Diamond, L. (1993). The Globalization of Democracy. In R.O. Slater, B.M. Schutz & S.R. Dorr (eds.),
Global Transformation and the Third World. Boulder: Lynne Rienner, 31-69.
Eckersley, R. (2000). The State and Fate of Nations: Implications of Subjective Measures of Personal
and Social Quality of Life. Social Indicators Research 52: 3-27.
Eckstein, H.S. (1988). A Culturalist Theory of Political Change. American Political Science Review
82(4):789-804.
Eisenstadt, S.N. (1986). The Axial Age Breakthroughs. In S. N. Eisenstadt (ed.), The Origins and
Diversity of Axial Age Civilizations. New York: State University of New York Press, 1-28.
Ersson, S. & Lane, J.-E. (1996). Democracy and Development. In A. Leftwich (ed.), Democracy and
Development. Cambridge: Polity Press, 45-73.
Estes, R.J. (1998). Trends in World Social Development. Journal of Developing Societies 14(1):1139.
Foweraker, J. & Landman, T. (1997). Citizenship Rights and Social Movements, Oxford: Oxford
University Press.
Freedom in the World. several volumes, Freedom House (ed.), Lanham: University Press of America.
Frey, R. S. & Al-Roumi, A. (1999). Political Democracy and the Physical Quality of Life. Social
Indicators Research 47: 73-97.
Fukuyama, F. (1995). Trust: Social Virtues and the Creation of Prosperity, New York: Free Press.
Gasiorowski, M.J. & Power, T.J. (1998). The Structural Determinants of Democratic Consolidation.
Comparative Political Studies 31 (December): 740-771.
Goldstein, H. et al. (1998). A User’s Guide to MLwin, Multilevel Models Project, University of
London.
Gurr, T.R., Jaggers, K. & Moore, W.H. (1990). The Transformation of the Western State. Studies in
Comparative International Development 25 (1):73-108.
Helliwell, J.F. (1993). Empirical Linkages Between Democracy and Economic Growth. British
Journal of Political Science 24 (April): 225-248.
Hughes, B.B. (1999). International Futures (3rd ed.). Boulder: Westview Press.
Huntington, S.P. (1991). The Third Wave, Norman: University of Oklahoma Press.
40
Huntington, S.P. (1996). The Clash of Civilizations and the Remaking of the World Order, New
York: Simon & Schuster.
Inglehart, R. (1977). The Silent Revolution, Princeton: Princeton University Press.
Inglehart, R. (1990). Culture Shift in Advanced Industrial Societies, Princeton: Princeton University
Press.
Inglehart, R. (1997). Modernization and Postmodernization, Princeton: Princeton University Press.
Inglehart, R. and Baker, W.E. (2000). Modernization, Cultural Change and the Persistence of
Traditional Values. American Sociological Review 65 (February): 19-51.
Inkeles, A. & Smith, D.H. (1974). Becoming Modern, Cambridge: Harvard University Press.
Inkeles, A. (1983). Exploring Individual Modernity, New York: Columbia University Press.
Jackman, R.W. and Miller, R.A. (1998). Social Capital and Politics. Annual Review of Political
Science 1:47-73.
Jaggers, K. & Gurr, T.R. (1995). Tracking Democracy’s Third Wave with the Polity III Data. Journal
of Peace Research 32 (4): 469-482.
King, G., Tomz, M. & Wittenberg, J. (2000). Making the Most of Statistical Analyses. American
Journal of Political Science 44(2):347-362.
Klingemann, H.-D. (1999). Mapping Political Support in the 1990s. In P. Norris (ed.), Critical
Citizens. New York: Oxford University Press, 31-56.
Knack, S. & Keefer, P. (1997). Does Social Capital Have an Economic Payoff? Quarterly Journal
of Economics: 1251-1288.
Kopstein, J.S. & Reilly, D.A. (2000). Geographic Diffusion and the Transformation of the
Postcommunist World. World Politics 53(October):1-37.
Landes, D.S. (1998). The Wealth and Poverty of Nations, New York: W.W. Norton.
Lasswell, H.D. (1958). Democratic Character. In H. Lasswell (ed.), The Political Writings. Glencoe:
Free Press, 465-525.
Leblang, D.A. (1997). Political Democracy and Economic Growth. British Journal of Political
Science 27 (3): 453-472.
Lerner, D. (1958). The Passing of Traditional Society, New York: Free Press.
Lerner, D. (1968). Modernization: Social Aspects. In D.L. Sills (ed.), The International Encyclopedia
of the Social Sciences (vol. 10), New York: Free Press, 386-395.
Lewis, W.A. (1955). The Theory of Economic Growth, Homewood: Richard D. Irvin.
Lipset, S.M. (1959). Some Social Requisites of Democracy. American Political Science Review 53
(March): 69-105.
Lipset, S.M., Kyoung-Ryung, S. & Torres, J.C. (1993). A Comparative Analysis of the Social
Requisites of Democracy. International Social Science Journal 45 (May): 155-175.
41
Macpherson, C.B. (1977). The Life and Times of Liberal Democracy, Oxford: Oxford University
Press.
Mouzelis, N. (1999). Modernity. British Journal of Sociology 50 (March): 141-159.
Muller, E.N. & Seligson, M.A. (1994). Civic Culture and Democracy. American Political Science
Review 88 (3): 635-52.
Nagle, J.D. & Mahr, A. (1999). Democracy and Democratization, London: Sage.
Nevitte, N. (1996). The Decline of Deference, Ontario: Broadview Press.
Olson, M.J., Sarna, N. & Swamy, A.W. (2000). Governance and Growth. Public Choice 102:341-364.
Page, B. & Shapiro, R.Y. (1993). The Rational Public and Democracy. In G.E. Marcus & R.L.
Hanson (eds.), Reconsidering the Democratic Public. Pennsylvania: Pennsylvania State
University Press, 35-64.
Perkin, H. (1996). The Third Revolution. London: Routledge.
Putnam, R.D. (1993). Making Democracy Work, Princeton: Princeton University Press.
Pye, L.W. (1990). Political Science and the Crisis of Authoritarianism. American Political Science
Review 84(March):3-19.
Randall, V. & Theobald, R. (1998). Political Change and Underdevelopment, Durham: Duke
University Press.
Robinson, W.S. (1950). Ecological Correlations and the Behavior of Individuals. American
Sociological Review 15: 351-357.
Rokeach, M. (1960). The Open and the Closed Mind, New York: Basic Books.
Rose, R. (2001). A Diverging Europe. Journal of Democracy 12(1):93-106.
Rostow, W.W. (1961). The Stages of Economic Growth, Cambridge: Cambridge University Press.
Rowen, H.S. (1996). World Wealth Expanding. In R. Landau, T. Taylor & G. Wright (eds.) The
Mosaic of Economic Growth. Stanford: Stanford University Press, 92-125.
Rustow, D.A. (1970). Transitions to Democracy. Comparative Politics 2 (April): 337-363.
Scarbrough, E. (1995). Materialist-Postmaterialist Value Orientations. In E. Scarbrough & J. van Deth
(eds.), The Impact of Values. Oxford: Oxford University Press, 123-159.
Scarrow, S. (1999). The Expansion of Direct Democracy. Paper presented at the Midwest Political
Science Association annual meeting, Chicago, April.
Sen, A. (2000). Development as Freedom. New York: Anchor.
Sides, J. (1999). It Takes Two: The Reciprocal Relationship Between Social Capital and Democracy.
Paper presented at the American Political Science Association annual meeting, Atlanta.
Sniderman, P. (1975). Personality and Democratic Politics, Berkeley: University of California Press.
Sørensen, G. (1993). Democracy and Democratization, Boulder: Westview Press.
Vanhanen, T. (1997). Prospects of Democracy, London: Routledge.
42
Verba, S., Nie, N.H. & Kim, J.-O. (1978). Participation and Political Equality, Cambridge:
Cambridge University Press.
Weber, M. (1958) [1905]. The Protestant Ethic and the Spirit of Capitalism. Boston: Allen and
Unwin.
Welzel, C. & Inglehart, R. (2001). Human Development and the ‘Explosion’ of Democracy. WZB
Discussion Paper, WZB: Berlin.
Feng, Y. (1997). Democracy, Political Stability and Economic Growth. British Journal of Political
Science 27 (3): 391-418.
Yule, G.U. & Kendall, M.G. (1950). An Introduction to the Theory of Statistics, London.
Zeller, R. & Carmines, E.G. (1980). Measurement in the Social Sciences, Cambridge: Cambridge
University Press.