Legitimacy and Participation: A Longitudinal Analysis of Public Opinion in Costa Rica, 2004-2008

Abstract: 

This study analyzes in a longitudinal context the dimensional conceptualization of legitimacy set forth by Booth and Seligson in The Legitimacy Puzzle in Latin America, in which they found that legitimacy was composed of six different dimensions: regime institutions, regime principles, local government, political actors, political community, and regime performance. This study first tests whether the evaluation level of each dimension of legitimacy remains constant. The second part of this study tests whether the dimensional structure of legitimacy remains stable over time using 2004 and 2008 Costa Rica Survey data, identifying issues that researchers who conduct longitudinal studies under this dimensional framework should address in order to yield empirically valid results, and challenging assumptions made in recent longitudinal studies.

Table of Contents: 

    The Dimensional Structure of Legitimacy

    In Easton’s early work on the subject (1975), he defined legitimacy as political support, citizen’s acceptance of the government’s right to rule, as described by Lipset’s seminal Political Man (1961). He divided this support into diffuse and specific categories, arguing in favor of a multidimensional conceptualization of legitimacy, rather than a one-dimensional conceptualization of legitimacy as a single measure of citizen support for a political system. Easton divided legitimacy between the more general components, related to support for the community and the regime, and more specific support for the authorities (1975).

    While not accepted by some scholars (Rogowski 1974), many built upon Easton’s conceptualization empirically, finding evidence for and further defining the components that make up diffuse and specific support (Muller and Jukam 1977; Weatherford 1992; Hetherington 1998). Norris proposed a five component conceptualization dividing Easton’s regime support category into support for regime principles, evaluation of regime performance, and support for regime institutions while keeping the other two dimensions of support for community and authorities (1999b). This conceptualization was discussed by scholars in a book edited by Norris (1999a).

    This discussion (Dalton 1999, Klingemann 1999, Newton 1999, and Norris 1999c) has largely shaped how scholars conceptualize legitimacy under a dimensional framework. Booth and Seligson found that these dimensions predicted by Norris and others are also found in Latin America, with some variance between countries, in an empirically rigorous study of eight Central American countries (2009). In their work on Costa Rica specifically, before this cross-national study, Booth and Seligson found an additional dimension of support for local government in their framework that had previously been largely absent from legitimacy research (2005).

    Legitimacy in a Longitudinal Context

    While much work has been done on the development of a multidimensional conceptualization of legitimacy, a very limited amount of research has been done on legitimacy over time. For example, Seligson (2002) links evaluation of one dimension of legitimacy — system support — to levels of voter turnout in Costa Rica. A later study by Booth and Seligson (2009) explored multiple dimensions of legitimacy with empirical rigor but did not track legitimacy over time. In this study, I explore some issues involved with multiple legitimacy dimensions over time in Costa Rica.

    I employ the dimensional framework found by Booth and Seligson, but add to their analysis by examining the structure and levels of legitimacy across time. I ask whether the legitimacy dimensions they identified for eight Latin American nations also exist in Costa Rica, and whether Costa Rican legitimacy dimensions persist there over time. Booth and Seligson have subsequent work (Seligson and Booth 2009; Seligson and Booth, Forthcoming) that assumes the structures they found in 2004 are stable (i.e. also exist in 2008), and that changes in legitimacy levels from 2004 to 2008 in Honduras have implications for that country’s political system stability.

    Such assumptions warrant careful empirical evaluation as to whether the cognitive structure of legitimacy is indeed stable. What is the empirical evidence of the persistence of legitimacy norms in Latin America? Should legitimacy’s dimensional structure prove to be volatile over time rather than stable, or to be sui generis for a particular time and country, the utility of identified legitimacy dimensions would diminish. The evaluation level of each dimension cannot be compared to that of other time periods if the factors that make up that dimension change over time.

    The adaption of a dimensional framework to a longitudinal context would make future research into the causes and consequences of legitimacy possible by allowing researchers to empirically analyze changes in norms over changing circumstances. Scholars could look at the level of legitimacy before an event that theory suggests would cause a change in legitimacy levels, observe whether a change takes place, and then note the consequences of that change as a long term empirical study. Indeed, Seligson and Booth do just this (2009) in a study of changing levels of three legitimacy dimensions in Honduras from 2004 to 2008.

    Hypotheses

    In order to measure legitimacy over time under this theory, it must be established that the dimensions found by Booth and Seligson are useful in a longitudinal context. By choosing a relatively stable period of time for this study, I am able to better understand the dimensional structure of legitimacy over time, but this limits the exploration of whether the evaluation level of each individual dimension of legitimacy varies appropriately with the social and political circumstances of Costa Rica during this time period. Limited analysis of this issue is possible in the context of this study; however, I can observe the lack of change in the evaluation levels as a product of the relative stability in the social and political context of Costa Rica during this time period. This leads to my first hypothesis:

    H1: The evaluation level of each dimension of legitimacy will remain relatively constant over time while the social and political context also remains constant.

    In order for there to be empirical comparability between different periods of time, the same dimensional structure would need to remain relatively constant over time, and the same dimensions should appear in both surveys. This is my second hypothesis:

    H2: The legitimacy dimensions observed in the 2004 survey data will persist in the 2008 survey data.

    Costa Rica is one of the most politically stable and economically developed countries in Central America. Based on 2003 data, the country scored the highest on the Human Development Index, had the second highest GDP per capita, and the highest percent change between 1990 and 2003. The country also scored highest among Central American countries on the Vanhanen Mean Democracy score, and had the highest score of Central American countries on the World Bank Effectiveness Index. Due to this stability, I expect that evaluation levels for each dimension will not show very much change, remaining relatively constant due to relatively stable political and economic circumstances in the country. Because Costa Rica has been an established democracy for over fifty years and has well developed political institutions, I expect the dimensional structure of legitimacy to also remain stable in Costa Rica.

    Method

    To test my hypotheses I conducted exploratory factor analysis on LAPOP survey data taken in Costa Rica in 2004 and 2008. This analytical technique, based on bivariate correlations among a large number of variables, seeks patterns or dimensions among the larger number of variables. Results are reported in the form of a factor structure of dimensions identified, plus a factor loading (essentially a correlation coefficient of each variable on each factor). For a table of the factor loadings for the 2004 and 2008 LAPOP Costa Rica surveys, see Appendix I. Survey questions are considered to be associated with a particular dimension (factor) if their loadings have an absolute value greater than 0.5 with that dimension.

    The factor analysis used an SPSS Oblimin Rotation method with Kaiser Normalization to account for and measure correlation among the dimensions found. This is important because rather than these being uncorrelated dimensions, I expected some correlation to be found among them. In The Legitimacy Puzzle, Booth and Seligson used this method as well as one in which the structure of legitimacy was modeled and then that model was tested via confirmatory factor analysis (2009). In my principal component analysis, I instead look for dimensionality among the survey factors without suggesting a model, allowing for insight into what considerations may be needed to adjust the model to a longitudinal study. After determining the factor composition of each dimension for both surveys, I calculate the descriptive evaluation level of each legitimacy dimension by averaging the means of the survey questions that compose each dimension. For a table of dimension evaluation level for both surveys, see Appendix II.

    Results: Evaluation Level

    Because Costa Rica was relatively stable politically and economically during this time period, it is difficult to tell how well each evaluation level would react to changing context, but the results I did find were encouraging because as circumstances remained stable, so did the evaluation level of each of the dimensions. The descriptive results of this showing the evaluation level of each dimension in 2004 compared to that of 2008 are shown in Figure 1 below.

    While analysis of a longer period of time with more data points and changing political and economic circumstances would be necessary to more conclusively address this question, this study allows a preliminary exploration of this issue. It seems that for the most part, the metric reflects important changes, or the absence thereof, in legitimacy norms. This finding comports with my hypothesis that the evaluation levels will show little change over time given a stable context.

    Results: Dimensional Structure

    Using data from 2002, Booth and Seligson performed a structural analysis of Legitimacy norms in Costa Rica (2005), providing a base-line for my analysis. They found seven dimensions in that study: existence of a political community, support for core regime principles, evaluation of regime performance, system support, support for regime institutions, support for local government, and support for political actors.

    A multi-dimensional structure was confirmed by Booth and Seligson using 2004 data (2009). I also found a multi-dimensional structure using 2008 data in this paper. The 2008 factor analysis is provided in Table IB of the appendix, and the 2004 analysis is provided in Table IA. While this is encouraging evidence of the multi-dimensional structure of legitimacy, some dimensions were more stable than others across this time period. This instability resulted from the tendency of some of the dimensions to merge.

    As discussed above, Costa Rica exhibited a great deal of institutional stability during this time period, and is one of the strongest Central American democracies. Because of this stability, I expected that the cognitive mapping of legitimacy as represented by factor analysis of survey data would remain relatively constant. The composition of many of the dimensions did remain constant between 2004 and 2008, but some showed change, partially disconfirming my hypothesis.

    The dimensions that stayed relatively stable were support for local government and support for regime principles. The support for the local government dimension was composed of the same two items regarding trust in local government and satisfaction with local government services in 2004 and 2008, at similar and relatively high factor loadings (.758 and .847 for 2004; .737 and .891 for 2008). The support for regime principles dimension exhibited similar stability and high factor loading values in both 2004 and 2008, being composed of the same three factors regarding democratic political participation in both datasets.

    In 2004, the political actors and regime performance dimensions blended together. Factors associated with both institutional and actor (focused on the president) legitimacy, identified by Booth and Seligson’s model (2009), combined into only one dimension, resulting in only five separate dimensions being identified. In contrast using the 2008 data, these factors showed a relatively significant correlation between one another of -.150, but were found to be separate dimensions under factor analysis rather than a single combined dimension as found in the 2004 data.

    Support for political institutions and the existence of political community dimensions were found to be separate dimensions in the 2004 data, but in the 2008 analysis the support for political institutions dimension partially merged with the political community dimension, three of the factors composing the support for political institutions dimension in 2004 associated with the political community dimension in 2008.

    Discussion and Conclusions

    My hypothesis that the evaluation level of each dimension would remain relatively constant between 2004 and 2008 due to the high level of political and social stability in Costa Rica during this time, comported with the data found in my descriptive analysis. My hypothesis that the dimensional structure of Costa Rica would remain stable, allowing for longitudinal comparability, was not always supported. I found that while some of the dimensions of legitimacy do remain stable during this time period, others did not due to a tendency of some dimensions to merge (associate with each other) in a specific dataset.

    My findings that the dimensional structure of Costa Rica did not remain stable, as I had initially expected, indicate that the theory of legitimacy may need further work and that more research should be done addressing the adaption of a multidimensional conceptualization of legitimacy to a longitudinal context. Scholars have assumed the stability of legitimacy’s structure over time in recent studies (Seligson and Booth 2009 and Forthcoming), using them to examine Honduras’s potential for political unrest. If these dimensions do not remain stable, as I find in this paper for Costa Rica, then the analysis in these studies of Honduras over time could be problematic.

    Longitudinal analysis using a dimensional structure of legitimacy needs more development before solid conclusions can be made. Perhaps dimensions could be aggregated to allow for dimensional analysis if dimensions blur over time. While a solution to this empirical hurdle is beyond the bounds of this paper, the assumption of structural stability should not be made in studies that compare legitimacy levels of specific dimensions among time periods In fact, the analysis of evaluation levels in the first portion of this paper are dependent on this assumption of structural stability, therefore solid conclusions cannot be made regarding whether the evaluation levels of individual dimensions vary appropriately with the social and political context of Costa Rica based on that analysis.

    References

    • Booth, John A. and Mitchell A. Seligson. “Political Legitimacy and Participation in Costa Rica: Evidence of Arena Shopping.” Political Research Quarterly 58 (2005): 537-550.
    • Booth, John A. and Mitchell A. Seligson. The Legitimacy Puzzle in Latin America: Political Support and Democracy in Eight Nations. New York: Cambridge University Press, 2009.
    • Dalton, Russell J. “Political Support in Advanced Industrial Democracies.” Critical Citizens: Global Support for Democratic Government. Ed. Pippa Norris. Oxford: Oxford University Press, 1999.
    • Easton, David. “A Re-Assessment of the Concept of Political Support.” British Journal of Political Science 5 (1975): 435-457.
    • Hetherington, Marc J. “The Political Relevance of Political Trust.” The American Political Science Review 92 (1998): 791-808.
    • Klingemann, Hans-Dieter. “Mapping Political Support in the 1990’s: A Global Analysis.” Critical Citizens: Global Support for Democratic Government. Ed. Pippa Norris. Oxford: Oxford University Press, 1999.
    • Muller, Edward N. and Thomas O. Jukam. “On the Meaning of Political Support.” The American Political Science Review 71 (1977): 1561-1595.
    • Newton, Kenneth. “Social and Political Trust in Established Democracies.” Critical Citizens: Global Support for Democratic Government. Ed. Pippa Norris. Oxford: Oxford University Press, 1999.
    • Norris, Pippa, Ed. Critical Citizens: Global Support for Democratic Government. Oxford: Oxford University Press, 1999a.
    • Norris, Pippa. “Introduction: The Growth of Critical Citizens?” Critical Citizens: Global Support for Democratic Government. Ed. Pippa Norris. Oxford: Oxford University Press, 1999b.
    • Norris, Pippa. “Institutional Explanations for Political Support.” Critical Citizens: Global Support for Democratic Government. Ed. Pippa Norris. Oxford: Oxford University Press, 1999c.
    • Rogowski, Ronald. Rational Legitimacy. Princeton: Princeton University Press, 1974.
    • Seligson, Mitchell A. “Trouble in Paradise? The Erosion of System Support in Costa Rica, 1978-1999” Latin American Research Review 37 (2002): 160-185.
    • Seligson, Mitchell A. and John A. Booth. “Predicting Coups? Democratic Vulnerabilities, The AmericasBarometer and The 2009 Honduran Crisis.” (2009): Americas Barometer Insights: 2009 Special Report on Honduras.
    • Seligson Mitchell A. and John A. Booth. “Criminal Violence, Economic Crisis and Democratic Stability in Central America” (Forthcoming): Journal of Democracy.
    • Weatherford, Stephen M. “Measuring Political Legitimacy.” The American Political Science Review 86 (1992):149-166.

    Figure 1: Evaluation Level of Each Dimension in 2004 Compared to 2008

    Appendix I: Factor Loadings for the 2004 and 2008 LAPOP Costa Rica Surveys

    Table IA: Factor Structure of Legitimacy Attitudes in Costa Rica, 2004.

    Variables Component
    1 2 3 4 5
    Political Institutions Regime (Democratic) Principles Regime Actors & Regime (Economic) Performance Local Government Political Community
    To what extent are you proud of being a Costa Rican? .199 .153 -.024 -.014 .684
    Despite our differences, we nationality have many things that unite us as a country. .007 .213 .170 .047 .698
    How much do you approve or disapprove: Of people participating in legal demonstrations? .075 .775 .111 .061 .036
    Of people participating in an organization or group to try to solve community problem> .072 .741 .090 .000 .214
    Of people working for campaigns for a political party or candidate? .233 .667 .101 .060 .116
    How would you describe the country’s economic situation? (very bad … very good) .145 -.158 .406 .322 -.223
    Current economy vs. that expected one year from now .189 -.006 .534 .272 -.179
    Hasta qué punto cree  que los tribunales de justicia de Costa Rica? .545 .191 .244 -.127 -.158
    To what extent do you respect the political institutions of Costa Rica? .524 .092 .221 .000 .269
    To what extent do you think that citizens’ basic rights are well protected by the political system of Costa Rica? .668 .056 .262 .197 -.067
    To what extent do you feel proud of living under the political system of Costa Rica? .677 -.045 .259 .241 .119
      .631 .101 .191 .183 .220
    Hasta qué punto tiene confianza en el Tribunal Supremo Elect .630 .087 .213 .157 .154
    Hasta qué punto tiene confianza en el Congreso Nacional .675 .186 .354 .162 -.086
    Hasta qué punto tiene confianza en los partidos políticos? .614 .262 .214 .160 -.172
    Hasta qué punto tiene usted confianza en la Corte Suprema de .662 .214 .429 .223 -.067
    To what extent do you trust the  municipality? .407 .201 .304 .758 -.022
    Satisfaction with local government services .142 .060 .192 .847 .044
    How well do you believe President Arias… Fights poverty? .351 .104 .785 .194 .037
    Protects democratic principles? .363 .191 .810 .095 .147
    Fights corruption? .264 .200 .767 .076 .133

    Extraction Method: Principal Component Analysis.
    Rotation Method: Oblimin with Kaiser Normalization.

    Component Correlation Matrix 2004

    Component 1 2 3 4 5
    Political Institutions Regime (Democratic) Principles Regime Actors & Regime (Economic) Performance Local Government Political Community
    1 Political Institutions -        
    2 Regime (Democratic) Principles .166 -      
    3 Regime Actors and Regime      (Economic) Performance .345 .113 -    
    4 Local Government .174 -.004 .227 -  
    5 Political Community .044 .123 -.021 -.044  

    Extraction Method: Principal Component Analysis.
    Rotation Method: Oblimin with Kaiser Normalization.

    Table IB.  Factor Structure of Legitimacy Attitudes in Costa Rica, 2008.

    Variable Component
    1 2 3 4 5 6
    Political Institutions  I Regime (Democratic) Principles Regime (Economic) Performance Political Community &  Institutions Local Govern-ment Political Actors
    To what extent are you proud of being a Costa Rican? -.080 .079 .108 -.549 .194 -.179
    Despite our differences, we nationality have many things that unite us as a country. -.013 .503 -.063 -.554 .047 -.080
    How much do you approve or disapprove: Of people participating in legal demonstrations? .010 .720 .067 .081 -.064 .045
    Of people participating in an organization or group to try to solve community problem> -.018 .792 -.041 -.258 .036 -.034
    Of people working for campaigns for a political party or candidate? .061 .790 -.073 -.254 .050 -.168
    How would you describe the country’s economic situation? (very bad … very good) .156 -.144 .724 .048 .066 -.299
    How would you describe your own economic situation? (very bad … very good) .057 .034 .714 .093 .014 -.111
    Do you consider your own economic situation to be worse, the same, or better than it was 12 months ago? .060 .024 .696 -.136 .026 .019
    Do you think the country’s economic situation is worse, the same, or better than it was 12 months ago? .083 -.011 .702 -.148 .109 -.143
    To what extent do you respect the political institutions of Costa Rica? .310 .259 .025 -.671 .055 -.070
    To what extent do you feel proud of living under the political system of Costa Rica? .554 .103 .096 -.604 .126 -.417
    To what extent do you think that one should support the political system of Costa Rica? .481 .055 .113 -.584 .106 -.459
    To what extent do you think the courts in Costa Rica guarantee a fair trial? .655 -.088 .137 -.039 .124 -.195
    To what extent do you think that citizens’ basic rights are well protected by the political system of Costa Rica? .615 -.076 .169 -.338 .192 -.359
    To what extent do you trust the Supreme Electoral Tribunal? .670 .150 .139 -.296 .195 -.474
    To what extent do you trust the Legislative Assembly? .745 .016 .054 -.084 .208 -.354
    To what extent do you trust the political parties? .590 .000 .077 .120 .235 -.479
    To what extent do you trust the Supreme Court? .742 .140 .051 -.173 .240 -.402
    To what extent do you trust the  municipality? .492 .083 .032 -.150 .737 -.338
    Satisfaction with local government services .076 -.049 .079 -.043 .891 -.149
    How well do you believe President Arias… Fights poverty? .358 .062 .173 -.137 .199 -.854
    Protects democratic principles? .385 .125 .137 -.412 .181 -.749
    Fights corruption? .357 .009 .147 -.066 .292 -.825

    Extraction Method: Principal Component Analysis.
    Rotation Method: Oblimin with Kaiser Normalization.

    Component Correlation Matrix 2008

    Component 1 2 3 4 5 6
    Political Institutions  I Regime (Democratic) Principles Regime (Economic) Performance Political Community & Instituitions II Local   Govern- ment Political Actors
    1 Political Institutions I -          
    2 Regime (Democratic) Principles .018 -        
    3 Regime (Economic) Performance .102 -.038 -      
    4 Political Community & Institutions II -.146 -.184 -.045 -    
    5 Local Government .185 .004 .063 -.098 -  
    6 Political Actors -.399 -.041 -.150 .181 -.246 -

    Extraction Method: Principal Component Analysis.
    Rotation Method: Oblimin with Kaiser Normalization.

    Appendix II: Dimension Evaluation Level for Both Surveys

    Descriptive Statistics Costa Rica 2004
      N Minimum Maximum Mean Std. Deviation
    Political Community 1500 1.00 98.81 91.55 10.05227
    Regime Principles 1500 7.54 96.58 73.3529 15.66912
    Regime Institutions 1500 8.98 96.47 58.9352 15.69817
    Regime Performance 1500 7.67 94.01 46.4979 15.13530
    Political Actors 1500 5.34 94.34 54.9664 20.31454
    Local Government 1500 1.00 97.89 45.0033 18.57000
    Valid N (listwise) 1500        
    Descriptive Statistics Costa Rica 2008
      N Minimum Maximum Mean Std. Deviation
    Political Community 1441 8.33 100.00 88.67 15.76128
    Regime Principles 1473 .00 100.00 75.6430 22.78899
    Political institutions 1493 .00 100.00 55.3034 19.71846
    Political Actors 1488 .00 100.00 52.9349 24.37795
    Local Government 1489 .00 100.00 51.2313 23.28135
    Regime Performance 1497 .00 100.00 47.7385 21.46796
    Valid N (listwise) 1420