The Diffusion of Values among Democracies and Autocracies Franziska Deutsch & Christian Welzel Paper submitted for presentation at the 24th World Congress of Political Science, July 23-28, 2016 in Poznań, Poland. Panel: "The Global Diffusion of Policy Innovations: Democracies and Autocracies Compared" Abstract Over the past couple of years, our knowledge about diffusion processes has increased greatly. So far, the focus has been on public policies, political institutions, and civil society repertories. Our paper breaks new ground by exploring value diffusion at the societal level. Using the World Values Survey data, we are interested in the diffusion of emancipative values in different regime types, building a bridge between research on the micro foundations and macro political expressions of democratization and political regimes. In particular, we wish to understand whether the psychological bedrock of a pro-democratic mass culture emancipative values is diffusing between countries and, if yes, to what extent this diffusion is limited to democracies or also makes inroads into autocratically governed societies. Franziska Deutsch Bremen International Graduate School of Social Sciences (BIGSSS) Jacobs University Bremen Email: f.deutsch@jacobs-university.de Christian Welzel Center for the Study of Democracy (CSD) Leuphana University of Lüneburg Email: cwelzel@gmail.com 1
Introduction In the last years, our knowledge of the diffusion processes has increased greatly. Originating in the literature on innovations, this research has informed us about how new policies, institutions or regime contention spread and are adopted across different levels and contexts (for early research, see Walker, 1969; Gray, 1973; Berry and Berry, 1990; more recently: Simmons and Elkins, 2004; Simmons, Dobbin and Garrett, 2008a; Della Porta and Tarrow 2012; Walsh Russo, 2014). While the political science literature has paid attention to the question of how democracy diffuses geographically and across time (Huntington, 1991; O Laughlin et al., 1998; Gleditsch and Ward, 2006, 2008; Weijnert, 2014; Goodliffe and Hawkins, 2015), little is known about diffusion processes with regard to the psychological foundations of democracy: people s democratic beliefs, attitudes and value orientations. Our paper aims at filling this gap by exploring the diffusion of values at the societal level and their impact on the establishment and practice of political institutions. Using data from various waves of the World Values Survey, we are interested in the diffusion of values in different regime types, building a bridge between research on the micro foundations and macro political expressions of democratization and political regimes. In particular, we wish to understand whether the psychological bedrock of a pro-democratic mass culture emancipative values is diffusing between countries and, if yes, to what extent this diffusion is limited to democracies or also makes inroads into autocratically governed societies. Before we address these questions empirically, we will outline what emancipative values are, how they are linked to political regimes, and how they diffuse. Emancipative values and value diffusion In their most general definition, values are conceptions of the desirable (Kluckhohn, 1951, p. 395) and can be held by individuals and groups. According to Shalom Schwartz, social psychologist and widely cited values researcher, values transcend specific situations, ( ) guide selection or evaluation of behavior and events, and ( ) are ordered by relative importance (Schwartz, 1992, p. 4). At the societal level, the predominant values in a society are a key characteristic of culture (Schwartz, 2008, p. 4) and influence what kind of beliefs, goals and behaviors are considered acceptable. If values are understood as widely accepted 2
conceptions of what is good and desirable, then the prevailing values in a society formulate, implicitly and explicitly, expectations to all its members. The political implications of cultural values shared in a society ( cultural identity ) are examined in the political culture framework. Already 50 years ago, the idea of cultural congruence was introduced by Almond and Verba (1965) and Eckstein (1966, 1997). Several studies have shown an empirical link between the prevalence of certain values in a society (by the early authors referred to as authority beliefs ) and the quality and durability of political institutions (Inglehart and Baker, 2000; Inglehart and Welzel, 2005; Welzel, 2013). This does not only apply to democracies, as Almond and Verba (1965, p. 34) noted early: Each kind of polity ( ) has one form of culture that is congruent with its own structure. Values, together with beliefs, attitudes and norms, form the psychological underpinnings of political regimes. The exact nature of those mass beliefs that are most conducive to a democratic regime and its endurance has been subject to debate. While Almond and Verba (1965) had argued in favor of specifically political orientations towards the political system and the role of the self within the system, such as civic competence and civic allegiance, more recently Welzel (2013) and Inglehart and Welzel (2005) could show that emancipative values (or: self-expression values) and assertive orientations (Welzel and Dalton, 2014) are most conducive to democracy. Emancipative values are the psychological manifestation of a broader human empowerment process. They emphasize human self-expression, freedom of choice and equality of opportunities and thereby encourage people to be masters of their own lives and active participants in their society's development (Welzel, 2013, p. 47). Emancipative values are closely linked to a stronger sense of agency, a stronger affinity to participate in elitechallenging political activism, and therefore ultimately contribute to a better quality of political institutions (Inglehart and Welzel, 2005; Welzel and Inglehart, 2009; Welzel and Deutsch, 2012; Inglehart and Welzel, 2014). To explain how values diffuse, we first need to have a closer look at the concept of diffusion. Rather vaguely, diffusion has been described as enduring, cross-boundary dependencies in the evolution of policies and institutions (Gleditsch and Ward, 2008, p. 264). Often, scholars use the term to refer to processes of spatial spreading and clustering of institutions, policies and other social phenomena. At times, diffusion addresses the impact of external factors such as the international or regional context or a network on a certain phenomenon. In this case, the diffusion approach stands in direct contrast to a developmental approach that focusses on internal factors or endogenous processes within a country, such as the 3
development of political, social or economic characteristics (Berry and Berry, 1990; Wejnert, 2014). Rochon (1998, p. 10) describes value diffusion as cultural change that may happen at two speeds: as rapid, explosive change, often accompanied by a moment of crisis, protest or unrest, and as slow change through generational replacement in a population. This mechanism of gradual cultural change has also been at the heart of Inglehart s thesis of postmaterialist value change (Inglehart, 1977): because younger generations in Western societies grow up in unprecedented conditions of prosperity and peace, their value priorities shift from materialist to postmaterialist values. As a consequence of this diffusion mechanism, over time these societies place higher priority on values such as freedom of speech, self-expression, emancipation and autonomy. While the approach of generational replacement aims at understanding the diffusion of values within a society, less is known about how values diffuse between societies. An established theoretical framework developed by Simmons, Dobbin and Garrett (2008b) does not seem to be a fruitful approach here: the four causal mechanisms of diffusion coercion, competition, learning and constructivist emulation all relate to a central agent s strategy in adopting a new policy. Clearly, value diffusion between different societies resembles a "swarm"-like herding process that lacks a central agent with a master plan. Roger s long-standing, almost classical definition of diffusion as the process by which an innovation is communicated through certain channels over time among the members of a social system (Rogers, 2003, p. 5) seems more helpful to understand value diffusion. Apart from the innovation itself, three elements are crucial for the study of diffusion: the channels of communication, defined as the means by which information is exchanged among the members of the social system; time, defined as the speed in which communication and adoption take place; and the social system made up by the sum of individuals and groups. Rohrschneider (1996, 1999), in an almost laboratory situation after the breakdown of communism in East Germany, studied the adoption of democratic values among the East German population and political elites in the early 1990ies. He juxtaposed two mechanisms of how democratic values can be learnt: institutional learning (the influence of the institutional setup on people s beliefs, the practice of democratic values the institutionalist perspective) and diffusion of values (exposure to and adoption of democratic values prior to regime change, i.e. through mass communication technologies or higher levels of education the 4
culturalist perspective). Rohrschneider attributed his result he found significant differences in democratic values between East and West to the first mechanism and argued that diffusion, if it works at all, can only happen under certain conditions, for instance when values require little democratic restraint or are compatible with socialist principles. In the next step, we turn to the empirical analyses. We are interested in how widespread emancipative values are in democracies and autocracies and if the latter are immune to the diffusion of emancipative values. If that is the case, we should not be able to find a rise in emancipative values over time in societies that are autocratically governed. Data and measures To measure emancipative values, we use data from six waves of the World Values Survey (WVS), with about 100 societies the largest collection of nationally representative surveys on human beliefs and values. The WVS began in the early 1980ies as an expansion of the European Values Study (EVS) and is built on a common core questionnaire that is implemented by an independent principal investigator in each country. The most recent wave finished in 2014, a seventh wave (2016-18) is currently planned. The measure of emancipative values has been continuously improved. i Most recently, emancipative values have been described as a combination of liberating and egalitarian orientations in four domains: choice, equality, voice and autonomy (Welzel, 2013, p. 67). Each domain includes three questionnaire items that have been continuously included in the World Values Survey. Choice measures how much people value having freedom in lifestyle and reproductive choices, asking respondents how acceptable they think (a) divorce, (b) abortion and (c) homosexuality are. Equality refers to gender equality, asking respondents how much they agree with the statements that (a) education is more important for a boy than a girl, (b) when jobs are scarce, men should have priority over women to get a job, and (c) men make better political leaders than women. People s voice includes three items that measure postmaterialist values (Inglehart, 1977), asking the respondents if they give first, second or no priority to (a) protecting freedom of speech, (b) giving people more say in important government decision, (c) giving people more say about how things are done at their jobs and in their communities. Autonomy is measured by asking the respondents whether they consider (a) independence and (b) imagination as desirable qualities when raising a child but not (c) obedience. 5
To combine the twenty items, they are first normalized between 0 and 1 (with fractions of 1 for intermediate positions) and then averaged into the four domain sub-indices. In a second step, the four sub-indices are averaged into the overall measure of emancipative values, providing a multi-point index between 0 and 1. As we are interested in the temporal diffusion of emancipative values and not their absolute values, we do not look at their overall distribution at a particular point in time, but focus only on changes in these values. In other words, it is possible that a society can strongly endorse emancipative values while still receiving a change score of 0 if the values did not increase over time. The change score can range between -1 and 1. To measure regime type, we draw on two types of data: Freedom House and the classification of autocracies provided by Geddes, Wright and Frantz (2014a). Since 1972, the yearly Freedom in the World report monitors and informs about the countries compliance with two essential parts of freedom: political rights and civil liberties. ii Strictly speaking, Freedom House is not aiming at assessing the quality of democracy but of political rights and civil liberties (Karatnycky, 1999). However, these two scales cover central elements of democracy, such as prevailing political liberties and equality, or political and judicial control (Lauth, 2003, p. 4). The measurement of freedom is based on the two scales, ranging between 1 and 7, with 1 meeting the highest standards (=highest level of freedom) and 7 failing to meet the standards (=lowest level of freedom). An average value for the two scales is calculated, indicating whether a country can be classified as free (average of 1 to 2.5), partly free (average of 3 to 5.0) or not free (average of 5.5 to 7). In addition, Freedom House classifies countries into electoral democracies (for an overview and critique, see Bogaards, 2012): All free countries are electoral democracies, all countries rated as not free are not. Partly free countries are a mixed bag some of them qualify as electoral democracies, some do not. Geddes et al. (2014a) present a new dataset to measure continuity and breakdown of autocratic regimes. The authors classify countries between 1946 and 2010 according to regime type, with an emphasis on the classification of autocracies into four different autocracies: party-based, personalist, military and monarchical. The classifications are documented in a supplement (Geddes et al., 2014b) to the journal article. In the following, we use the information from the two sources Freedom House and Geddes et al. (2014b) first separately, then combined. We combine the sources in the following way: Countries that are rated free by Freedom House are liberal democracies; countries that are rated electoral democracy but not considered free by Freedom House are electoral democracies. 6
Countries that are classified as one of the four autocratic regime types by Geddes et al. (2014b) are autocracies, with the respective sub-classification. In two cases of disagreement between both data sources, we follow Geddes et al. (2014b). iii As the data for regime types provided by Geddes et al. (2014b) are only available until 2010 and the most recent measure of emancipative values for many countries is dated later, we update the regime classification for those countries where the regime type could be determined beyond doubt. In case of doubt we refrain from classifying the countries and omit them from the analysis with regard to regime types. iv The six categories of regime type (liberal and electoral democracies; partybased, personalist, military and monarchical autocracies) are mutually exclusive. Empirical results and discussion We wish to answer the question to what degree values diffuse over time. To examine this question, we are using longitudinal data from all waves of the World Values Survey. Figure 1 shows the longitudinal change in emancipative values for each society that was surveyed at least twice, allowing us to follow change as a function of time from the earliest to the latest available survey. ***Insert Figure 1 about here*** As Figure 1 illustrates, there is clear evidence of a trending cultural change in values (r = 0.65, p =.00): for the majority of societies, emancipative values have changed over time, and for most societies, they have increased. Moreover, emancipative values increase merely as a function of time passage, which is the best evidence for trending cultural change a pivotal aspect of diffusion. The trend is most pronounced among postindustrial Western societies, which also happen to be mature democracies. Examples include Norway, Sweden, Spain or Canada. Yet, the rise of emancipative values is by no means limited to postindustrial Western democracies. They are just the countries with the longest time series evidence. Nevertheless, there are a significant number of countries that are either immune to a diffusion mechanism (those societies that cluster around the zero score on the vertical axis) or even witnessed a decrease of emancipative values: among them are several post-soviet countries like Russia, Belarus, Kyrgyzstan, Armenia and Azerbaijan, as well as countries with a resurgence of religious fundamentalism, such as Nigeria and Turkey. Because all of these 7
countries either never really democratized or slid back to authoritarianism, the absence of an emancipatory cultural shift might not come as a surprise. In fact, the absence of such a shift might be the explanation of these countries resistance to democracy. How robust is the trending force in cultural change against plausible controls? When controlling for GDP/capita, the effect of time passage on value change remains fairly strong and highly significant (r = 0.50, p =.00). In other words: Independent of the level of emancipative values and socioeconomic development, emancipative values diffuse over time. Figure 1 already suggested that values may diffuse differently in democratic and autocratic countries for many publics in Western democracies, emancipative values rise with the passage of time, while in many autocracies the results are more mixed. In Figure 2 we examine this possible influence of regime type more systematically and present change in emancipative values according to different typologies: On the left, Figure 2 displays value change according to the Freedom House classification in free, partly free and not free societies. On the right, results are shown for a dichotomous regime classification: democracies (liberal and electoral democracy combined) and autocracies (party based, military, personalist and monarchical autocracy combined). The results are visualized in boxplots, which allow us to compare groups in their means and distributions. Boxplots present data according to quartiles: The bottom and top of the box show the first and third quartiles of the data, the line in the middle is the median, the second quartile. In other words: 75 percent of the data for each group are located within each box. The vertical lines above and below the box show the maximum and minimum values of the distribution. ***Insert Figure 2 about here*** The left diagram of Figure 2 indicates that the increase in emancipative values over time is most pronounced in free societies. Almost all of them experienced a rise in these values, even though to very different degrees, as can be seen in the size of the box. Partly free societies are more homogenous, even more so are non-free societies (the box is smaller), but also here for more than half of the countries we see a small rise in emancipative values. The reason why partly free societies are more similar to non-free instead of free societies can be found in the right diagram of Figure 2: partly free societies are indeed a mixed bag. When classified according to criteria of electoral democracy, the difference in value change between democracies and autocracies becomes even more pronounced. In other words: autocracies are indeed more resistant to the diffusion of emancipative values than democracies. At the same 8
time, autocratic regimes are much more similar to each other compared to democracies where three quarters of all countries cluster rather closely together but the difference between the maximum (not only limited to the outliers Sweden and Norway) and minimum value is large. It should be noted though that Figure 2 does not say anything about the level of emancipative values, only about change. Finally, Figure 3 looks more closely at change in emancipative values in different regime types. Following Geddes et al. (2014a) and Tosun and Croissant (XXX, introduction to the special section), value change is shown for four types of autocracies: party-based, personalist, military and personalist autocratic regimes. The theoretical distinction between permissive and restrictive opportunity structures in democracies (Tosun and Croissant XXX, p. 14) is captured by our distinction between liberal and electoral democracies. ***Insert Figure 3 about here*** The results are interesting in many ways: While liberal democracies to a large extent experienced a strong rise in emancipative values over time, this cannot be said about electoral democracies (which for our purposes exclude liberal democracies), even though for the majority of them the change is still positive. The results for the four types of autocracies explain the mixed finding for autocratic regimes in Figure 2: some autocratic regimes are indeed more resistant to value diffusion than others: personalist autocracies show by majority no change at all or to the negative. They are most immune to emancipative values, followed by party-based autocracies. The results for the other two autocratic regime types are intriguing too: military regimes, and even more so autocratic monarchies, not only experienced an increase in emancipative values; they actually experienced a bigger increase than most electoral democracies. Conclusion The goal of this study was to examine the diffusion of values in different regime types. Using data from six waves of the World Values Survey, we were interested in whether emancipative values diffuse at all, whether value diffusion is limited to democracies, or whether or not autocracies are immune to the diffusion of those values. Our empirical results show that emancipative values are indeed on the rise, which we call temporal diffusion (i.e., trending cultural change). Moreover, the increase in emancipative values is more pronounced but not limited to societies that are democratically governed. However, it matters how well a 9
democracy is governed: mass publics in liberal democracies experienced a much stronger increase in emancipative values than publics in electoral democracies. For autocracies, a differentiation between different autocratic regime types is necessary not all autocracies are alike. While emancipative values are on the rise in monarchies and military regimes, partybased autocracies and in particular personalist dictatorships are immune to emancipative value diffusion. All of this seems to indicate that regime type is a causal factor in cultural change. But this conclusion would be premature: in a regression analysis that explains change in emancipative values by the time passage over which it is measured and modernization as well as regime type at the beginning of the change, regime type turns out to be insignificant. In light of this evidence, the existing linkage between regime type and value diffusion does not suggest that the former determines the latter. Quite the contrary, it suggests that, if non-democratic regimes persist, they do so because the processes that give rise to emancipative values do not kick in. 10
References Almond, G. A. and Verba, S. (1965) The Civic Culture. Political Attitudes in Five Western Democracies. Princeton: Princeton University Press. Berry, F. S. and Berry, W. D. (1990) State Lottery Adoptions as Policy Innovations: An Event History Analysis, American Political Science Review, 84 (2), pp395-415. DOI: 10.2307/1963526. Bogaards, M. (2012) Where to Draw the Line? From Degree to Dichotomy in Measures of Democracy, Democratization, 19(4), pp690-712. DOI: 10.1080/13510347.2011.613563. Della Porta, D., & Tarrow, S. (2012) Interactive Diffusion: The Coevolution of Police and Protest Behavior with an Application to Transnational Contention, Comparative Political Studies, 45 (1), pp119-152. DOI: 10.1177/0010414011425665. Eckstein, H. (1966) A Theory of Stable Democracy. Princeton: Princeton University Press. Eckstein, H. (1997) Congruence Theory Explained. Center for the Study of Democracy, University of California/Irvine. CSD Working Paper. http://repositories.cdlib.org/csd/97-07 Geddes, B., Wright, J. and Frantz, E. (2014a) Autocratic Breakdown and Regime Transitions: A New Data Set, Perspectives on Politics, 12 (2), pp313-331. DOI: 10.1017/S1537592714000851. Geddes, B., Wright, J. and Frantz, E. (2014b) Autocratic Regimes Code Book. Version 1.2. Supplement to Autocratic Breakdown and Regime Transitions: A New Data Set, Perspectives on Politics, 12 (2), pp313-331. Gleditsch, K. S. and Ward, M. D. (2006) Diffusion and the International Context of Democratization, International Organization, 60 (4), pp911-933. DOI: 10.1017/S0020818306060309. Gleditsch, K. S. and Ward, M. D. (2008) Diffusion and Spread of Democratic Institutions, in B. A. Simmons, F. Dobbin and G. Garrett (eds.), The Global Diffusion of Markets and Democracy. Cambridge et al.: Cambridge University Press, pp. 261-302. Goodliffe, J. and Hawkins, D. (2015) Dependence Networks and the Diffusion of Domestic Political Institutions, Journal of Conflict Resolution. first published on July 27, 2015. DOI: 10.1177/0022002715596772. 11
Gray, V. (1973) Innovation in the States: A Diffusion Study, American Political Science Review, 67, pp1174-1185. Huntington, S. P. (1991) The Third Wave: Democratization in the Late Twentieth Century. Norman: University of Oklahoma Press. Inglehart, R. (1977) The Silent Revolution. Changing Values and Political Styles among Western Publics. Princeton: Princeton University Press. Inglehart, R. and Baker, W. E. (2000) Modernization, Cultural Change, and the Persistence of Traditional Values, American Sociological Review, 65 (1), pp19-51. Inglehart, R. and Welzel, C. (2005) Modernization, Cultural Change, and Democracy. The Human Development Sequence. Cambridge: Cambridge University Press. Inglehart, R. and Welzel, C. (2014) Political Culture, in D. Caramani (ed.), Comparative Politics. Third edition. Oxford: Oxford University Press, pp. 284-301. Karatnycky, A. (1999) The Decline of Illiberal Democracy, Journal of Democracy, 10 (1), pp112-125. Kluckhohn, C. (1951) Values and Value Orientations in the Theory of Action. An Exploration in Definition and Classification, in T. Parsons and E. Shils (eds.), Toward a General Theory of Action. Cambridge: Harvard University Press, pp. 388-433. Lauth, H.-J. (2003) Problems and Limitations of Measuring Democracy. Paper Presented at a Conference Reassessing Democracy: New Approaches to Governance, Citizenship and Multiple Identities in Comparative Research in Bremen, Germany, June 20-22, 2003. O Laughlin, J., Ward, M. D., Lofdahl, C. L., Cohen, J. S., Brown, D. S., Reilly, D., Gleditsch, K. S. and Shin, M. (1998) The Diffusion of Democracy, 1946-1994, Annals of the Association of American Geographers, 88 (4), pp545-574. Rochon, T. R. (1998) Culture Moves. Ideas, Activism, and Changing Values. Princeton: Princeton University Press. Rogers, E. M. (2003) Diffusion of Innovations. 5 th edition. New York et al.: Free Press. Rohrschneider, R. (1996) Institutional Learning versus Value Diffusion: The Evolution of Democratic Values among Parliamentarians in Eastern and Western Germany, The Journal of Politics, 58 (2), pp422-446. 12
Rohrschneider, R. (1999) Learning Democracy. Democratic and Economic Values in Unified Germany. Oxford and New York: Oxford University Press. Schwartz, S. H. (1992) Universals in the Content and Structure of Values. Theoretical Advances and Empirical Tests in 20 Countries, in M. P. Zanna (eds.), Advances in Social Psychology. New York: Academic Press, pp. 1-65. Schwartz, S. H. (2008) Cultural Value Orientations: Nature and Implications of National Differences. Moscow: State University - Higher School of Economics Press. Simmons, B. A. and Elkins, Z. (2004) The Globalization of Liberalization: Policy Diffusion in the International Political Economy, American Political Science Review, 98 (1), pp171-189. DOI: 10.1017/S0003055404001078. Simmons, B. A., Dobbin, F. and Garrett, G. (2008a) The Global Diffusion of Markets and Democracy. Cambridge et al.: Cambridge University Press. Simmons, B. A., Dobbin, F. and Garrett, G. (2008b) Introduction: The Diffusion of Liberalization, in B. A. Simmons, F. Dobbin and G. Garrett (eds.), The Global Diffusion of Markets and Democracy. Cambridge et al.: Cambridge University Press, pp. 1-63. Tosun, J. and Croissant, A. (XXX): Policy Diffusion: A Regime-Sensitive Conceptual Framework, Introduction to the Special Section, Global Policy, XXX. Walker, J. L. (1969) The Diffusion of Innovation among American States, American Political Science Review, 63, pp880-99. Walsh Russo, C. (2014) Diffusion of Protest, Sociology Compass, 8 (1), pp31-42. DOI: 10.1111/soc4.12115. Wejnert, B. (2014) Diffusion of Democracy. The Past and Future of Global Democracy. Cambridge: Cambridge University Press. Welzel, C. (2013) Freedom Rising. Human Empowerment and the Quest for Emancipation. Cambridge: Cambridge University Press. Welzel, C. and Dalton, R. J. (2014) From Allegiant to Assertive Citizens, in R. J. Dalton and C. Welzel (eds.), The Civic Culture Transformed. From Allegiant to Assertive Citizens. New York: Cambridge University Press, pp. 282-306. 13
Welzel, C. and Deutsch, F. (2012) Emancipative Values and Nonviolent Protest: The Importance of Ecological Effects, British Journal of Political Science, 42 (2), pp465-479. DOI: 10.1017/S0007123411000421. Welzel, C. and Inglehart, R. (2009) Political Culture, Mass Beliefs, and Value Change, in C. H. Haerpfer, P. Bernhagen, R. F. Inglehart and C. Welzel (eds.), Democratization. Oxford and New York: Oxford University Press, pp. 126-144. 14
Figure 1. Temporal diffusion of emancipative values. Source: World Values Survey (WVS). N=81, r =.65, p =.00. The surveys were conducted 1981-84 (wave 1), 1990-94 (wave 2), 1995-98 (wave 3), 1999-2004 (wave 4), 2005-09 (wave 5), 2010-14 (wave 6). The figure only includes those societies where emancipative values have been surveyed at least twice and uses the longest time distance between the available surveys (distance in years between earliest and latest available survey). 15
Figure 2. Diffusion of values in different regime types (boxplots). Source: World Values Survey, see Figure 1. To allow for comparisons, the composition of the groups is kept identical at both data points at which emancipative values were measured: to be included in the analysis, a society had to be coded the same within each typology (i.e. free/free; democracy/democracy) at the measurement years of earliest and latest survey available. 16
Figure 3. Diffusion of values in different democratic and autocratic regimes (boxplot). Source: World Values Survey, see Figure 1. The regime type categories based on Geddes et al. (2014, b) and Freedom House are mutually exclusive. Monarchical, military, personalist and party-based regimes are autocracies. i For an older version, see Inglehart and Welzel (2005). ii Political rights are assessed by evaluating the electoral process (free and fair elections), presence or absence of political pluralism and participation opportunities, functioning of government and protection of minority rights in a country. Freedom of expression and belief, the right to form an association and organization, compliance with the rule of law as well as the guarantee of personal autonomy and individual rights are among the standards for evaluating civil liberties. iii These two cases of disagreement are (1) Russia, that has been classified as personalist autocracy since 1993 by Geddes et al. (2014b: 14) but as an electoral democracy between 1993 and 2004 by Freedom House; and (2) Zambia, that has been classified as party-based autocracy since 1996 by Geddes et al. (2014: 14), but as an electoral democracy since 2006 by Freedom House. iv For example Kyrgyzstan 2011, Nigeria 2011, or Pakistan 2012. 17