Life satisfaction research has increasingly drawn the academic attention of social scientists in the last several decades. Past research shows that citizens’ life satisfaction has important political implications as it correlates with regime support or enhances the likelihood of political participation across various types of regimes, including in the well-established Western democraciesFootnote 1 and the younger democracies of Latin American countries,Footnote 2 as well as in the former Soviet UnionFootnote 3 and contemporary authoritarian China.Footnote 4 Past research also shows that subjective life satisfaction assessments are determined by institutional-level political featuresFootnote 5 and individual-level characteristics.Footnote 6 However, there has been scant research on how citizen life satisfaction judgements are affected by individuals’ fundamental political values in non-democratic countries.
In this study, we focus on the theoretical links between citizens’ fundamental political values and their life satisfaction assessments in an authoritarian context by using recent national public survey data to empirically examine how two key ingredients of individual political values – authoritarianism and national attachment – shape citizens’ life satisfaction in contemporary China. Our data analyses with different model specifications consistently demonstrate that Chinese citizens with stronger authoritarian ideology and national attachment are likely to report a higher level of life satisfaction.
Our study contributes to the literature by highlighting the importance of political values in shaping citizens’ life satisfaction. It also advances our understanding of the origins of social stability and political sustainability in authoritarian regimes where citizen life satisfaction usually represents a major source of state legitimacyFootnote 7 and government satisfaction.Footnote 8
Theoretical Expectations and Hypotheses
Life satisfaction refers to a cognitive judgement of a person's overall quality of life as a whole at a particular time according to his or her criteria.Footnote 9 Previous research shows that life satisfaction judgements are shaped by both institutional-level political features, such as the complexion of government ideology,Footnote 10 the size of government,Footnote 11 and the quality of governance,Footnote 12 as well as individual-level factors, such as citizens’ issue attention and preferences, personal values, judgement standards and top-down biases.Footnote 13 The limited research on life satisfaction in authoritarian contexts mainly focuses on social and economic indicators or psychological and biological traits,Footnote 14 while the relationships between citizens’ political values and their life satisfaction assessments tend to be overlooked or underexplored.
This study advances the research on life satisfaction by focusing on the effects of two key ingredients of individuals’ fundamental political values in an authoritarian context: authoritarian ideology and national attachment. Examining the effects of these political values can supplement the existing socio-economic or psychological explanations of the sources of citizen life satisfaction and contribute to a deeper understanding of the origins of social and political stability in authoritarian regimes, as life satisfaction has important implications for state legitimacy and political support in these countries.Footnote 15
The compatibility of citizens’ political ideology with political climate can affect their evaluation of life satisfaction. Previous research based on US samples shows that political climate can activate the cognitive and affective reactions of citizens and that citizens are more optimistic about their lives when the political climate is congruent with their political ideology.Footnote 16 Following this logic, in an authoritarian context citizens with stronger authoritarian beliefs should have higher life satisfaction than citizens with weaker authoritarian beliefs. When what citizens believe is consistent with what they see, they will suffer little from cognitive dissonance.Footnote 17 Particularly, authoritarian beliefs tend to lead to obedience, conformity to group norms and the underrating of freedom.Footnote 18 As citizens in an authoritarian country are typically socialized by the state-owned propaganda machine, authoritarian beliefs provide them with the psychological benefit of social worth.Footnote 19 For these reasons, we formulate the first hypothesis:
H1: Individuals with stronger authoritarian ideology will have a higher level of life satisfaction.
As an important ingredient of social and political values, national attachment refers to citizens’ positive evaluations of their national identity.Footnote 20 National attachment may have a positive impact on citizens’ life satisfaction. Psychologists have found that “human beings have an innate need for attachment and recognition.”Footnote 21 Citizens with stronger national attachment have a stronger sense of security, identity and belongingness.Footnote 22 They tend to have fewer aspirations to pursue goals incompatible with national, societal or group interests and are found to be more satisfied with their country or society.Footnote 23 Given this logic, national attachment is often emphasized by governments in order to maintain political stability and regime legitimacy in an authoritarian context.Footnote 24 Therefore, the positive relationship between national attachment and life satisfaction should be particularly salient in an authoritarian context. This leads to our second hypothesis:
H2: Individuals with stronger national attachment will have a higher level of life satisfaction.
It is important to note that we do not assert that the direction of influence is exclusively from individuals’ political values to their life satisfaction assessments, as it is possible that the level of life satisfaction may also affect individuals’ political values. However, we speculate two reasons that the direction of causality should be primarily from fundamental political values to life satisfaction. First, as discussed earlier, individual life satisfaction is a subjective evaluation of one's overall quality of life at a particular time, while one's fundamental political values have long been developed and formed through political socialization processes since childhood.Footnote 25 Fundamental values reflect abstract beliefs, principles and orientations absorbed in accumulated experiences, and they most likely pre-exist one's life satisfaction evaluation. Second, fundamental political values are relatively stable over time,Footnote 26 as people hardly change their political ideologies and beliefs, while life satisfaction evaluation is relatively fluid and contingent, depending on the changing environment and specific life circumstances (for example, good or bad health, income shocks, etc.). In political science and psychology, these relatively stable political values and ideologies are theoretically considered and empirically shown as information-processing heuristics or motivated reasoning mechanisms,Footnote 27 guiding and steering people's contingent evaluations of subjective well-being and life satisfaction.Footnote 28 Therefore, it is more plausible that fundamental political values influence life satisfaction assessments, rather than the other way around.Footnote 29
Survey Data and Variable Measurement
Data used in this study are from one of the China Governance and Public Policy Surveys (CGPPS) conducted between 28 October and 5 December 2016 by the Survey and Research Center for China Households (SRCCH) at Southwestern University of Finance and Economics (Chengdu, China) through a random dialling computer-aided telephone interviewing system. The sample of this CGPPS was drawn from the SRCCH's 2015 national survey panel. This probability-based panel employed a stratified three-stage probability proportion to size (PPS) random sample design with additional onsite global positioning system/ geographic information system (GPS/GIS) remote sensing sampling strategy to draw a sample of Chinese adults aged 18 and older in mainland China (excluding Xinjiang, Tibet, Hong Kong and Macau). The SRCCH maintained a list of panel members’ primary contact numbers (either cell telephone or landline telephone). For this CGPPS survey, 7,298 individuals from 29 provinces and 353 county-level administrative divisions were sampled and then contacted via their primary contact number (cell phone or landline phone). Of those contacted, 6,138 respondents agreed to take the survey (an 84.1 per cent response rate), and 3,972 participants completed the interviews (a 64.7 per cent completion rate). All interviews were conducted in Chinese, with the average completion time taking about 14 minutes.
Dependent variable
The dependent variable in this study is Life Satisfaction, which was measured by a respondent's answer to a question in the survey asking to what extent the respondent is currently satisfied with his or her life as a whole. The original response category had five scales (very unsatisfied = 1; unsatisfied = 2; neither unsatisfied nor satisfied = 3; satisfied = 4; and very satisfied = 5). Since the central tendency of the original five-scale responses was strongly skewed towards the right side of the scale (27.83 per cent were very satisfied and 45.3 per cent satisfied), and very few respondents fell under the subcategories on the left side of the scale (2.21 per cent were very unsatisfied and 5.33 per cent unsatisfied), we rescaled the variable for analysis by collapsing the original scales from 1 to 3 and creating a non-satisfied subcategory while leaving the two other most common subcategories (satisfied and very satisfied) alone. We then recoded: non-satisfied = 1; satisfied = 2; and very satisfied = 3. This rescaling corrects for the skewed distribution in the original scale and avoids estimation issues related to the sparse responses in the raw data. After the rescaling and recoding, the distribution of the dependent variable Life Satisfaction is much more balanced, as shown in Figure 1.
Figure 1: Distribution of the Recoded Life Satisfaction Scale (Three Subcategories)
Source:
2016 China Governance and Public Policy Survey.
Key explanatory variables
The key explanatory variables in this study are the two ingredients of individuals’ fundamental political values, authoritarian ideology and national attachment. To measure individuals’ authoritarian ideology, we used respondents’ answers to two questions in the survey about their agreement or disagreement with the following statements: (1) “Protests and demonstrations should be forbidden,” and (2) “Freedom of the press shall be under the direction of the Party and the government,” where 1 = completely agree; 2 = agree; 3 = neither agree nor disagree; 4 = disagree; 5 = completely disagree. We reversed the coding of their answers and created two variables, Protest Ban and Press Control, to reflect the strength of respondents’ authoritarian ideology, where a higher score indicates a stronger authoritarian belief.
To measure the degree of respondents’ national attachment, we used a set of standard questions commonly used in extant literature to gauge individual citizens’ national identity.Footnote 30 In the survey, respondents were asked, on a 0–10 scale, with 0 indicating strongly disagree and 10 indicating strongly agree, to what extent did they disagree or agree with the following statements:
(1) “Overall, China is better than other countries”;
(2) “I love China”;
(3) “The world would be better if foreigners were more similar to Chinese people”;
(4) “Chinese culture is the best culture in the world”;
(5) “I'm really proud to be a Chinese when Chinese athletes have good achievements in competitions”;
(6) “I'm proud to be a Chinese”;
(7) “I would be a Chinese again even if I could choose to be a citizen of whichever country I want”;
(8) “Emotionally, I regard myself as a Chinese first, rather than a person from a specific province or city.”
Past literature suggests there are generally two types of national attachment: patriotism and nationalism.Footnote 31 Patriotism refers to the non-competitive love of and attachment to one's country, which is neutral with regard to the evaluation of others,Footnote 32 while nationalism is related to a sentiment that one's country is superior to others.Footnote 33 To determine whether these two distinct dimensions – i.e. patriotism and nationalism – are embedded among Chinese citizens, we conducted factor analysis (with principal component extraction and varimax rotation methods) using survey participants’ responses to the above eight items. The results, reported in Table 1, indicate that the eight question items had the strongest loading on two separate dimensions or factors. Factor 1 appears to reflect a patriotic dimension as it includes the five questions that are typically considered as indicators of patriotism (i.e. non-competitive pride of and attachment to one's country), while Factor 2 seems to represent a nationalism dimension as it clearly loads together the three questions with a nationalistic sentiment of “superiority.” Based on the factor analysis, two factor scores were imputed to reflect respondents’ national attachment on each dimension: Patriotism and Nationalism.
Table 1: Factor Loading
Source:
2016 China Governance and Public Policy Survey.
Control variables
We included a series of control variables in our statistical models to rule out potential confounding effects. First, we used two variables to control for individual citizens’ attitudes towards the status quo in China, as past research suggests that life satisfaction may be correlated with citizens’ unfavourable perceptions of current social conditions or issues such as pollution and unemployment.Footnote 34 In the CGPPS, one question asked respondents about how serious they think the issues of civil liberty and political democracy are in China. Another question asked about how serious they think the issue of environmental protection is in China. Responses to both questions were on a 0–10 scale, with 0 being not serious at all and 10 being extremely serious. Based on their responses, two variables, Liberty–Democracy Concern and Environmental Concern, were used as proxy measures for their perceptions of some aspects of the status quo in China, with a higher score indicating a stronger unfavourable attitude towards the current political and environmental conditions, respectively.
Second, we included standard socio-demographic variables and socio-economic status in our models. The CGPPS collected the following background information from survey participants: Female (1 = female; 0 = male); Age (respondents’ actual age); Education (9 levels, ranging from no school at all = 1, to doctorate degree = 9); Income (logged annual household per capita income); Urban (urban resident = 1, otherwise = 0); and CCP (member of the Chinese Communist Party = 1, otherwise = 0). These socio-demographic background and socio-economic status variables were used as control variables in our regression analysis, as previous studies find that some of these variables also affect Chinese citizens’ life satisfaction.Footnote 35
Descriptive statistics of all the variables are reported in Table 2.Footnote 36 The bivariate correlation matrix is reported in the Appendix.
Table 2: Descriptive Statistics
Source:
2016 China Governance and Public Policy Survey.
Data Analysis and Result
Since the dependent variable, Life Satisfaction, is a three-scale ordinal variable (1 = not satisfied, 2 = satisfied, and 3 = very satisfied), we used ordered logit regression to assess the relationships.
We estimated four separate regression models. Model 1 regresses Life Satisfaction with the four variables that reflect respondents’ political values, including two authoritarian ideology variables (Protest Ban and Press Control) and two national attachment variables (Patriotism and Nationalism). Model 2 adds all the individual-level control variables (respondents’ perception of status quo, demographic information and socio-economic status) to Model 1. In Model 3, we included province fixed effects to account for unobservable factors (tradition, culture, economy, politics and policies, etc.) that vary across provinces. By including provincial fixed effects (defined as a dummy variable for each province, minus one as a reference), Model 3 controls for the average differences across provinces and reduces the risk of omitted variable bias.Footnote 37 Finally, in Model 4, we constructed 20 imputed datasets, creating imputed values for the missing observations in the independent variables and analysing the results of multiple imputation estimates.Footnote 38
Table 3 shows the empirical results. In Table 3, the percentage change in odd ratios for one unit increase in each independent variable and t statistics in parentheses are provided. The two-tailed p values are shown in separate columns. Moreover, Table 3 reports the McFadden R-squared, Akaike information criterion (AIC) and the Bayesian information criterion (BIC). We used the robust standard errors adjusted for clustering by county in all models (not reported in Table 3) to account for the potential heteroskedasticity and correlation between respondents within a county.
Table 3: Ordered Logit Model of Life Satisfaction
Source:
2016 China Governance and Public Policy Survey.
Note:
Ordered logit models with two-tailed p values. The dependent variables are measured with a three-point scale, ranging from not satisfied (1), satisfied (2) to very satisfied (3). Robust standard errors clustered by counties (not reported).
Consistent with our key theoretical expectations, our test corroborates that political values significantly predict citizens’ life satisfaction in China. Model 1 shows that the McFadden R-square of political values is 0.068. A comparison of Model 1 and Model 2 shows that the McFadden R-square of individual-level socio-economic variables is 0.031 (i.e. 0.099–0.068). A comparison of Model 2 and Model 3 further shows that the McFadden R-square of province-level factors is 0.006 (i.e. 0.105–0.099). These results suggest that compared to individual-level socio-economic variables or province-level factors, the measures of political values have much higher explanatory power for life satisfaction.
Owing to the hundreds of missing values in the measures of key independent variables, the first three models may suffer from the potential bias caused by list-wise deletion. Therefore, in Model 4, we imputed these variables by employing the most commonly used multivariate normal regression.Footnote 39 The results of the multiple imputation estimates in Model 4 are consistent with the regression results based on list-wise deletion in the first three models. Substantively, in Model 4, all else being equal, one unit increase in citizens’ support for the ban on protests is associated with a 6.1 per cent increase in the odds of life satisfaction. Similarly, one unit increase in citizens’ support for government control of the press is associated with a 45.9 per cent increase in the odds of life satisfaction. These results suggest that authoritarian ideology has a positive impact on citizen life satisfaction in China, which is consistent with our first hypothesis.
Moreover, the results also show that, all else being equal, one unit increase in citizens’ patriotism is associated with a 15.8 per cent increase in the odds of their life satisfaction and one unit increase in citizens’ nationalism is associated with a 26.3 per cent increase in the odds of their life satisfaction. These findings are supportive of our second hypothesis, suggesting that national attachment is positively associated with citizens’ life satisfaction in China.
In addition, the regression results in Table 3 reveal some interesting estimated effects of the control variables on Chinese citizens’ life satisfaction. According to Model 4, one unit increase in citizens’ liberty–democracy concern is associated with a 6.1 per cent decrease in the odds of their life satisfaction, and one unit increase in citizens’ concern about the environment is associated with a 3.3 per cent decrease in the odds of their life satisfaction. These findings indicate that Chinese citizens who are more concerned about current social issues and conditions tend to have lower levels of life satisfaction. With regard to the demographic and socio-economic status variables, the results show that when compared to male citizens, female Chinese are 16.5 per cent less likely to be satisfied with their life, which is not surprising given the long history of gender inequality in China.Footnote 40 The results in Table 3 also show that older or richer people are more satisfied with their life than younger or poorer people – a finding consistent with past researchFootnote 41 – and that urban residents are significantly less satisfied with their life than rural residents. Nevertheless, we do not find strong evidence to support the potential effects of education or Party membership on the life satisfaction of citizens in China.Footnote 42
Conclusion and Discussion
In this study, we attempted to address one underexplored question in the past research literature: the links between individuals’ political values and their life satisfaction evaluation in the context of an authoritarian regime. Most of the previous studies have been conducted in well-developed democracies. In the limited research on life satisfaction in authoritarian or developing contexts, scholars mainly focus on the effects of socio-economic factors and psychological or biological traits rather than the fundamental political values of citizens.
To shed new light on this question, this work theorizes that two fundamental political values held by individual citizens, authoritarian ideology and national attachment, will significantly shape their life satisfaction in an authoritarian context. The empirical results based on a recent public survey in China consistently show that relative to their counterparts, citizens with stronger authoritarian ideology and national attachment tend to have a greater level of life satisfaction.
Our work represents one of the first attempts to systematically examine the role of political values in citizens’ life satisfaction evaluation in an authoritarian context. Our research findings show that political values have more substantial explanatory power compared to the traditionally used measures of socio-economic status in authoritarian China. As life satisfaction may significantly shape the state legitimacy,Footnote 43 our research suggests that political values of citizens have important broader implications for the regime survival and stability as well as for political reform and change. Although this work is based on a single survey conducted in China, the key findings suggest that the role of fundamental political values deserves more attention in life satisfaction research. Future research may utilize common or similar question items from other surveys with fine-grained information to compare and corroborate findings,Footnote 44 and more research should be conducted in other authoritarian contexts to verify the generalizability of the findings in this study.
Overall, the principal conclusion emerging from our study is clear: there are strong links between citizens’ fundamental political values and their life satisfaction assessments. While in the theoretical section we have discussed the reasons why we think that the causal path is likely to flow primarily from citizens’ fundamental political values to their life satisfaction evaluations, it is worth noting that the statistical relationships shown in our regression analyses are correlations rather than empirical causations. To continuously advance scholarly knowledge on this topic, future research should consider using an experimental design, panel data (when available), or other statistical techniques (for example, instrumental variable) to further verify the direction of causality between political values and life satisfaction.
Acknowledgements
The material used in this study is based on the research project, China Governance and Public Policy Survey (CGPPS), which is supported by the Bush School of Government and Public Service and by the Institute for Science, Technology and Public Policy at Texas A&M University. The findings, statements and conclusions are solely those of the authors. We thank the anonymous reviewers for their insightful comments and helpful suggestions.
Conflict of interest
None.
Biographical notes
Xinsheng LIU is a research scientist for the Institute for Science, Technology and Public Policy in the George Bush School of Government and Public Service, Texas A&M University. He is also affiliated with Inner Mongolia University's School of Public Administration as a Changjiang scholar visiting chair professor and Peking University's Institute of State Governance Studies as a research fellow.
Youlang ZHANG is an assistant professor in the School of Public Administration and Policy, Renmin University of China, Beijing. He is also a research fellow of the Institute for Science, Technology and Public Policy in the George Bush School of Government and Public Service, Texas A&M University.
Arnold VEDLITZ is a professor and distinguished research scholar of the Institute for Science, Technology and Public Policy, and Bob Bullock chair in government and public policy in the George Bush School of Government and Public Service, Texas A&M University.
Appendix: Bivariate Correlation Matrix
