Valuing Politics: Explaining Citizen’s Normative Conceptions of Citizenship

Civic duty is a critical feature of explanations of political participation, but why do individuals differ in their propensity to adopt particular civic duty norms? We argue that norms are likely to be adopted when congruent with underlying values and support this contention using evidence from the European Social Survey. We show that individuals who prioritize self-transcendence values (which capture altruistic goals) place much greater emphasis on norms positively related to participation while individuals who prioritize conservation values (which capture goals related to social stability) place much less emphasis on these norms. The substantive influence of personal values in explaining norm adoption is greater than that of education, age, and interpersonal trust. Our results thus provide new insights into how citizens form their normative conceptions of citizenship and also highlight the substantive role played by value orientations in conformity to social norms more generally.

Explorations of civic duty have traditionally focused on the duty to vote.However, many individuals also highlight acts other than voting as normatively important behaviors for citizens to perform (Conover et al., 1991(Conover et al., , 2004;;Lane, 1965;Theiss-Morse, 1993;Van Deth, 2007).For instance, some believe it is their duty to volunteer while others highlight the importance of obeying laws or displaying social solidarity.Acceptance of these norms, in turn, influences participation both at and outside the ballot box whereas the norm of voting promotes voter turnout but not necessarily non-electoral actions (Bolzendahl & Coffé, 2013;Dalton, 2008a;Theiss-Morse, 1993).The varied, but substantially important, relationship between discrete civic duty norms and political engagement motivates our core question: why do individuals differ in the civic duty norms they believe are important?
Existing work on civic duty norms has typically focused on the role of the social environment both as a source of norms via socialization experiences as well as a policer of norms via social sanctioning processes (Campbell, 2006a;Gerber et al., 2008;Hansen & Tyner, 2021;Knack, 1992;Letki, 2006).However, this elides the important question of why individuals vary regarding which norms they internalize, i.e. come to accept as important regardless of the presence of social sanctions.Campbell (2006b, p. 73) suggests that "as young people undergo socialization, they are imprinted with norms", but social influence processes are far from perfect and individuals vary in the lessons they learn from their environment.We thus develop a theory concerning the motivated acceptance of civic duty norms to complement existing approaches.We argue that individuals discriminate across civic duty norms by placing more importance on those norms congruent with their core personal values to satisfy motives related to maintaining the integrity of their personal identity. 2e support the foregoing claim with evidence from the European Social Survey.We find that individuals who prioritize self-transcendence values, which capture pro-social motivations, rather than self-enhancement motives, which capture ego-oriented goals such as success, place much greater emphasis on norms related to political participation, autonomy, and social solidarity.Meanwhile, individuals who prioritize conservation values that capture goals related to security and social conformity place less emphasis on these norms while instead accepting duty norms regarding the maintenance of the existing social order.The association between civic duty norms and these value dimensions is substantially greater than their association with commonly used predictors of civic duty including educational attainment, religiosity, and interpersonal trust.Our results thus provide new insight into how citizens form their normative conceptions of citizenship and highlight the substantive role played by value orientations in the acceptance of social norms and ultimately political participation.

3
Political Behavior (2023Behavior ( ) 45:1447Behavior ( -1466 The Multifaceted Nature of Citizenship Norms The norms associated with citizenship are multifaceted in nature, but can be placed into four categories: participatory, autonomy, social order, and social citizenship (Conover et al., 2004;Dalton, 2008a;Lane, 1965;Theiss-Morse, 1993;Zmerli, 2010).Participatory norms refer to behaviors such as voting and being active in political or social organizations.This is the dimension typically focused on in studies of duty and participation (Blais, 2000;Campbell, 2006aCampbell, , 2006b;;Riker & Ordeshook, 1968).Autonomy norms reference private actions such as forming one's own opinion about politics and being open-minded while doing so.Social order norms refer to acts such as obeying laws whose end is the maintenance of the existing social order.Finally, social citizenship norms stem from Marshall's (1950) concept of the same name and concern acts conducive to social solidarity such as helping the worse off.
The importance placed on these norms varies across both individuals and nations with the majority of variation occurring at the individual level (Denters & van der Kolk, 2008).High levels of support for social order norms and the participatory norm of voting can be found in nearly all nations where survey evidence exists (Dalton, 2008a;Denters et al., 2007;Van Deth, 2007).More variation exists for participatory norms focused on non-voting political acts and for social citizenship norms, with the former typically receiving the lowest level of aggregate support.Understanding variation in autonomy and non-voting participatory norms is especially important given that individuals who deem these norms important are widely active in the political process; these norms may operate as instigators of political activity even in the midst of declining acceptance of the norm of voting in elections (Blais & Rubenson, 2013;Bolzendahl & Coffé, 2013;Dalton, 2008aDalton, , 2008b;;Raney & Berdahl, 2009;Theiss-Morse, 1993). 3 More attention has been given to the consequences of citizenship norms than to their antecedents.Research concerning variation in the duty to vote has principally focused on the positive role of education as well as on aspects of the social environment, such as social capital and homogeneity, that facilitate norm transmission and policing (Blais, 2000;Campbell et al., 1960;Campbell, 2006a;Letki, 2006).The emerging literature on citizenship norms other than voting, meanwhile, has principally focused on demographic divisions and the role of social capital, with group differences largely explained as emanating from varying socialization experiences (Bolzendahl & Coffé, 2009;Coffé & Bolzendahl, 2011;Dalton, 2008aDalton, , 2008b;;Denters et al., 2007;Kotzian, 2014;Theiss-Morse, 1993; but see Dinesen et al., 2014 for work on personality traits and non-voting norms).Socialization experiences and the characteristics of one's social environment influence norm adoption by making some norms, rather than others, available to individuals to consider.Likewise, contextual factors matter insofar as they affect the ability of other agents 3 Of course, norms do not always translate into behavior.Individuals may fail to act on norms for a variety of reasons including a lack of resources such as time or money necessary for completing the action or, relatedly, a lack of efficacy (Brady et al., 1995;Fishbein & Ajzen, 2010).
to reward and/or punish the individual for taking norm-congruent behaviors (e.g., Gerber et al., 2008).However, such approaches do not fully account for individual variation among individuals exposed to the same types of environments.We thus need a theory to account for these 'bottom-up' processes as well (Jost et al., 2009).

The Motivated Adoption of Social Norms
A person's social environment may make a variety of norms available to them; what explains why an individual internalizes some rather than others?To answer this question we turn to the concept of motivated reasoning which posits that individuals process social information in light of their goals and needs (Kunda, 1990).Social norm adoption is no different with research on non-political norms suggesting the primacy of three motives and associated psychological needs: accuracy (need for efficient but sound decision making), other-related (need for social attachment), and self-related motives (need for self-integrity; Cialdini & Trost, 1998;Pool & Schwegler, 2007).Understanding normative conceptions of citizenship thus requires paying attention to people's underlying motives and how well they align with discrete duty norms.
Our focus is on self-related motives, which we describe in a moment.We believe that focusing on this motivation provides a unique vantage point for understanding the adoption of civic duty norms that both complements and builds on prior work which has highlighted the importance of social pressure and hence, implicitly, otherrelated motives.The social policing of norms is most effective when the behavior is public and the norm is 'crystallized', i.e. widely accepted and performed (Gerber et al., 2008;White et al., 2014).While this may characterize the norm of voting, it is far from clear that it describes other citizenship norms.Norms such as obeying the law or forming one's own opinion involve largely private behaviors.Meanwhile, behaviors related to participatory and social solidarity norms are less common than voting as can be seen from our data source, the 2002 European Social Survey.While 80.1% of ESS respondents indicated voting in the last election, just over half (51.3%) indicated performing none of nine other political acts.Likewise, approximately half (50.5%) of respondents reported being a member of zero out of eleven types of voluntary groups.Respondents were asked how often they helped others outside of their family, a behavior which should be relevant for solidarity norms; while 8.2% said "every day," a greater proportion said never (9.9%) and the modal response on the survey was "less often" than once a month (28.3%).This heterogeneity may weaken perceptions concerning whether performing these acts will yield social rewards, which means that personal motives may be especially important considerations in the adoption of these types of norms.Our focus is thus consonant with prior studies and takes them as a starting point.
By self-related motives we refer to cases where an individual connects a social norm to a core aspect of their self-concept.Individuals, particularly in western nations, possess a need to maintain the integrity of their self-concepts by acting in an identity-congruent manner (Sherman & Cohen, 2006).Individuals also possess a need for autonomy, i.e. to feel that they are acting in line with their own goals rather 1 3 Political Behavior (2023Behavior ( ) 45:1447Behavior ( -1466 than the externally enforced wishes of others (Ryan & Deci, 2000).The satisfaction of these psychological needs by the performance of identity-congruent norms activates positive emotional states, such as pride, while avoidance leads to negative states such as shame (Christensen et al., 2004;Wood et al., 1997).These emotional states help form a bond between the norm and the individual's identity, which in turn fosters the internalization of the former into the latter and a resulting intrinsic motivation to act in line with the norm.We thus expect that individuals will internalize norms they view as congruent with the core elements of their self-identity.

Personal Values and Citizenship Norms
Our central claim is that individuals will be more likely to adopt norms they perceive to be congruent with core aspects of their identity.Our specific focus is an individual's personal values.Values are positively evaluated goals that lie at the core of an individual's identity; they capture the ends an individual generally wishes to achieve (Caprara & Vecchione, 2013;Feldman, 2003b;Hitlin, 2003).Given their inherently normative content values operate as trans-situational guides for judging alternatives and have been frequently used to understand the adoption of more specific attitudes and behaviors (e.g., Feather, 1995;Goren et al., 2016Goren et al., , 2020;;Piurko et al., 2011;Rathbun et al., 2016).
Our focus is on personal values, i.e. values that describe the goals an individual wishes to achieve in their own lives.These values can be differentiated from more politically focused value constructs such egalitarianism or individualism in that the latter ask individuals about the importance of these goals for the community at large.We focus on personal values for several reasons.First, they are more basic than, and ultimately underlie, political values (Goren et al., 2016;Malka et al., 2014;Schwartz et al., 2010Schwartz et al., , 2014)).Second, personal values are a better conceptual fit given our focus on the self-related motives of norm adoption.As Vecchione et al., (2013, p. 17) show, personal values, unlike political values, are not reciprocally influenced by an individual's political choices because they are "more central to their self-identities" and thus "changing a basic [personal] value has implications for one's identity and for the whole range of specific attitudes, values, and behavior linked to it across life domains".Both arguments suggest that personal values are likely exogenous to citizenship norms in a manner that political values may not be.Finally, personal values help structure the roles (and role identities) an individual takes on, the social groups they identify with, and ultimately the nature of a person's beliefs about the world and thus may work both directly and indirectly on norm adoption (Conover & Feldman, 1984;Huddy, 2001, p. 144;Piurko et al., 2011).
The most widely adopted theory of personal values, and the one we utilize here, is the Schwartz Value Theory (SVT; Schwartz, 1992;Schwartz et al., 2014;Vecchione et al., 2013).The SVT identifies ten core personal values that capture distinct goals of individuals.These ten values can be reduced to two superordinate dimensions that focus attention on more fundamental conflicts, and motives, underlying human agency. 4The first dimension is bookended by the "self-transcendence" values of universalism and benevolence on the one end and the "self-enhancement" values of power and achievement on the other end.The second dimension, meanwhile, is bookended by the "conservation" values of security, conformity, and tradition on one end and the "openness-to-change" values of stimulation, self-direction, and hedonism on the other end.We discuss these dimensions in more detail and elaborate more specific hypotheses regarding their relationship to the citizenship norms reviewed earlier in the following sections.Table 1 provides an overview of our general expectations and how they map onto the specific measures of civic duty examined later.

Dimension 1: Self-transcendence and Self-orientation Values
The first dimension of the SVT is the self-transcendence/self-orientation dimension.Individuals who prioritize the self-transcendence end of the dimension emphasize altruistic goals, e.g.helping others, while individuals who prioritize the opposite end focus on goals related to their own welfare, i.e. "pursuing one's own success and dominance over others" as a means to avoid anxiety (Schwartz et al., 2014).This dimension of values thus focuses on a fundamental tension of group life: whether to focus on the welfare of others and hence the group(s) one belongs to or, instead, whether to focus on one's own welfare.
We expect that individuals that place more emphasis on self-transcendence relative to self-enhancement values will place greater importance on social citizenship, autonomy, and participatory norms.It seems likely that an individual who values treating people equally will likewise indicate that citizens ought to help others given the congruence between these beliefs.Autonomy norms, meanwhile, place an onus on the individual to be self-critical and open to other's perspectives, which should again cohere with the general goals of individuals who prioritize self-transcendence over self-enhancement goals.Finally, with regards to participatory norms, we note that the goals that comprise this dimension echo work on social preferences and participation (e.g., Fowler & Kam, 2007).As those studies demonstrate, individuals with altruist preferences are motivated to participate in the political arena to achieve instrumental benefits for other individuals, while egoists generally shun political activity as it an unlikely mechanism for obtaining instrumental benefits for oneself.We would add that there is also an identity component here.Altruistic individuals will believe they should participate in part to continue seeing themselves as the type of person who contributes to the public welfare.We thus expect a similar relationship for this dimension of values and participatory norms.
H1 Greater adherence to self-transcendent values (and thus less support for selforientation values) will be associated with greater acceptance of social citizenship, autonomy, and participatory norms, all else equal.

Dimension 2: Conservation and Openness-to-Change Values
The second dimension of the SVT is labeled conservation/openness-to-change.Individuals who prioritize the conservation end of this dimension are animated by a drive to minimize anxiety through the avoidance of "conflict, unpredictability and change by submission and passive acceptance of the status quo" (Schwartz et al., 2014, p. 905).Individuals who place greater emphasis on the goals characterizing the other end of the dimension, meanwhile, stress goals such as seeking excitement, novelty, and following one's "own intellectual and emotional interests in unpredictable and uncertain directions" (Schwartz, 1992, p. 43).A key concern of this dimension is another fundamental issue in group living: autonomy versus hierarchy (Feldman, 2003a).We expect that individuals that place more emphasis on conservation relative to openness-to-change values will also place greater importance on social order norms and less importance on autonomy norms.Social order norms are a natural 'fit' for such individuals as their performance is directly related to the goals of these values.Autonomy norms, on the other hand, stress actions that are less congruent with the goals of following received wisdom and should be given less importance.Individuals who place importance on openness-to-change values, meanwhile, believe that individuals should generally follow their own critical lights, which would seem to be a more natural 'fit' with autonomy norms than social order norms.
We expect this value dimension to have a nuanced relationship with participatory norms.First, we expect that greater adherence to conservation values will be associated with the norm of voting in elections.Voting in elections is an action explicitly promoted by private and public authority figures as a mark of good citizenship through socialization efforts and mass mobilization campaigns.Individuals who prioritize conservation values, per above, are expected to experience a general motivation to adopt such officially sanctioned participation norms.Conservation values, moreover, are associated with stronger national identification which should additionally strengthen the drive to adhere to norms espoused by national leaders (Roccas et al., 2010).On the other hand, we expect a negative relationship between conservation values and norms regarding other, less common and descriptively normative, forms of participation.Active participation outside of the ballot box is likely to bring with it the conflict and unpredictability adoption of these values is supposed to control in the first place all without the connection to expressing national identity that makes voting a desirable act for these individuals.Active participation should thus be less attractive to individuals emphasizing conservation values and more attractive to those placing greater weight on the opposite end of the dimension given their desire for sensation seeking, which is an end well suited for participation in voluntary activities and 'non-traditional' forms of participation (Kam, 2012).
H2 Greater adherence to conservation values (and thus less support to openness-tochange values) will be associated with greater acceptance of social order and voting norms, but lower acceptance of autonomy and active participatory norms, all else equal.

Data
We turn to the 2002 wave of the European Social Survey (ESS) to test our hypotheses.The ESS is a biennial cross-sectional survey of respondents from over twenty European nations.Respondents were recruited via random probability sampling procedures and the resulting country samples are representative of each country's adult population.Individuals were sampled from 22 countries in the 2002 ESS, but we only have full data for 20 countries because the items measuring value priorities were not asked in Italy and Luxembourg.Table OA1 provides the sample size for these countries as well as response rate per country.This survey is ideal for our purposes as it contains batteries of items tapping both the SVT and civic duty norms using representative samples from a wide array of national contexts.Indeed, it is the only survey known to us that contains the measures necessary to fully test our hypotheses.

Civic Duty Norms
Respondents were asked to rate the importance of six behaviors on a 0 (extremely unimportant) to 10 (extremely important) scale via the following question prompt: "To be a good citizen, how important would you say it is for a person to…".These six behaviors were: "vote in elections"; "always obey laws and regulations"; "form their own opinion, independently of others"; "be active in voluntary organizations"; "be active in politics"; and "support people who are worse off than themselves".These behaviors tap into the four-categories of civic duty norms discussed earlier (participatory: vote in elections, be active in politics, be active in voluntary 1 3 Political Behavior (2023) 45:1447-1466 organizations; social order: always obey laws and regulations; autonomy: form their own opinion; and social citizenship: support people who are worse off).
Table 2 provides summary statistics concerning the distribution of each item while Figure OA1 plots the distribution of each norm.Responses tend to be skewed toward the "important" end of the scale albeit with variation across acts with particularly high levels of support for norms related to obeying all laws, voting in elections, and forming one's own opinion and much weaker support for the norms of joining voluntary associations and being active in politics.The vast majority of variation in responses, meanwhile, is at the individual level with intra-class correlation statistics from multilevel models without any predictors ranging from 0.05 (opinions) to 0.08 (obey laws), which is consistent with prior work (Denters & van der Kolk, 2008).

Personal Values
Respondents' value priorities were assessed using the Portrait Values Questionnaire (PVQ).The PVQ includes twenty-one questions where each measure provides an individual with a description of a hypothetical person and the respondent reports their similarity to this person on a 1-6 scale (from "not like me at all" to "very much like me").For instance, someone who scores highly on self-transcendence values would indicate that they are like a person who "think[s] it is important that every person in the world should be treated equally" while someone who rated the opposite end of the scale (self-enhancement) very highly would instead place greater emphasis on statements such as "it is important to her/him to be rich".All question wordings are provided in Online Appendix Table OA2.
We scaled responses to questions measuring conservation and self-transcendence values such that higher scores indicated acceptance of the value in question, i.e. 1 = "not like me at all", 6 = "very much like me".Responses to questions measuring the other end of each dimension, openness-to-change and self-enhancement respectively, were scaled such that higher values indicated rejection of the value, i.e. 1 = "very much like me", 6 = "not like me at all". 5 We then averaged together the items for each dimension such that higher scores indicated both acceptance of one end and rejection of the other.Table 2 provides descriptive statistics for both measures: Self-Transcendence vs. Self-Enhancement and Conservation vs. Openness-to-Change.Our measures thus explicitly capture the relative priority the respondent gives to the values in question.
The foregoing method had the unfortunate consequence of producing scales with relatively low reliabilities (α conservation = 0.65, α self-transcendence = 0.53).We investigate an alternative method of measuring relative value priorities in Online Appendix B (Figures OB1 & OB2).There, we scale all individual items such that higher values indicate acceptance of the value in question, calculate indices separately for either end of the dimension (i.e.self-transcendence, self-enhancement, conservation, and openness-to-change), and then take the difference between these indices (i.e., selftranscendence-self-enhancement and conservation-openness-to-change; see Feldman, 2003a).The advantage of this method is that the four values have higher reliabilities.Ultimately this measurement choice does not affect our conclusions.

Models and Controls
Individual respondents are nested within countries in the ESS.Failing to take account of the hierarchical nature of the data would depress the size of standard errors and bias our results (Steenbergen & Jones, 2002).The models presented below are random-intercept/random-slope multi-level models where we predict the dependent variable based on a set of individual-level (i.e.Level 1) variables as well as country-level (i.e.Level 2) variables used to model country-specific intercepts.
We fit three multilevel models to test our hypotheses.All models use survey weights and exclude non-citizens.Model 1 only includes value priorities as predictors with their slopes allowed to vary across country-contexts.These variables (norms and values) were rescaled to range from 0-1 in all analyses as are subsequent control variables.This model thus provides the most straight-forward estimate of the role of values, but one likely biased by the exclusion of omitted variables.
In Models 2 and 3 we add individual and country-level control variables that may plausibly influence both value orientations and norm adoption due to social influence processes. 6The individual level controls added in Model 2 are: respondent education; father's education; gender; age; a categorical variable concerning whether the respondent was born in the country and, if not, how long the respondent had lived in the country; union membership; and whether the respondent indicated that they are a member of a minority group in the country.These variables speak to 1 3 Political Behavior (2023Behavior ( ) 45:1447Behavior ( -1466 the differential socializing environments that may affect the availability and policing of norms and hence a social route to their acceptance as well as environments that may affect value orientations as well.At the country-level, meanwhile, we include indicators for ethnic heterogeneity, the number of years of democratic rule in the country, and a dummy variable for whether the country was governed by a socialist/ Soviet government in the post-World War II era as these variables may influence the availability, transmission, acceptance, and meaning of both the civic duty norms and personal values in the wider society (Anderson & Paskeviciute, 2006;Peffley & Rohrschneider, 2003;Schwartz & Bardi, 1997). 7 Finally, in Model 3 we add further individual-level controls that have been investigated in previous studies of citizenship norms but whose relationship with values is more ambiguous.Specifically, Model 3 further controls for the respondent's religiosity, interpersonal trust, how often they report meeting with others socially, and residential stability (i.e.how long they have lived in their neighborhood).These variables may structure the social(izing) environment of individuals and thereby influence norm acceptance while potentially affecting value orientations themselves.For instance, religious individuals may be exposed to pro-solidarity norm messages and political mobilization campaigns with contents of religious sermons also potentially influencing respondent value orientations (Margolis, 2018).At the same time, some of these variables may plausibly be post-treatment to values themselves.Individuals who value tradition and security, for instance, may be less likely to move to new areas and hence live in the same area for longer and may be more likely to opt into continued religious practice than individuals who instead prioritize opennessto-change values.The estimates from Model 3 may thus be under-estimates of the influence of value priorities in guiding civic duty norm adoption insofar as these added controls are post rather than pre-treatment.

Analyses
We begin with Fig. 1 which plots the coefficients for the two value priorities indicators from each of the three models predicting norm acceptance; Tables OA3-OA8 provide full model results.In Hypothesis 1 we argued that individuals who prioritized self-transcendence values over self-enhancement values would indicate greater support for norms related to participation, autonomy, and social citizenship.Figure 1 provides supportive evidence on all counts.All coefficients are both positive and statistically significant regardless of model specification.The relationships on display are also substantively important as well with changes of between 1/10th ("be active") to 2/5th ("support others") of the range of the dependent variable.Figure 1 shows that self-transcendence values have a wide and deep influence on 7 We examine the case of Belgium in greater detail in Online Appendix B given that Belgium has a compulsory voting law that may affect norm acceptance.There, we find that Belgian and non-Belgian respondents do not significantly differ in acceptance of these duty norms and that our results are robust to excluding Belgian respondents.citizenship norms and thus appear to be a potent underlying source of their adoption and hence on political engagement more generally.
We argued in Hypothesis 2 that conservation vs. openness-to-change values would also be relevant with the prioritization of the former being associated with greater support for the norms of voting and obeying laws but less support for other types of participatory actions and autonomy norms.8 Figure 1 generally provides support for these claims.Moving from the minimum to the maximum on this variable was associated with increases of 6 (voting) and 40 (obeying laws) percentage points and a decline of 10 (being active in politics) and 7 (volunteering) percentage points on the dependent variable in question in Model 3.However, the relationship between this values dimension and volunteering norms was not statistically significant contrary to expectations.These values thus also influenced norm acceptance in the expected manner although the substantive importance of these relationships tended to be weaker than what is seen for selftranscendence vs. self-enhancement.
The relationship between personal values and citizenship norms in Fig. 1 is often quite sizable in nature and especially so for the value/norm combinations of self-transcendence/volunteering, self-transcendence/opinions, self-transcendence/supporting others, and conservation/obeying laws.How does the correlation of these value dimensions and civic duty compare to the other predictors in the models and particularly the lodestones of previous research on civic duty: education, age, and interpersonal trust?To answer this question we turn to Fig. 2 which plots the coefficients from the values and control variables in Model 3.For the categorical controls in Model 3 (own education, father's education, and when the individual came to the country if not born there) we focus on either min/ max comparisons (i.e. the difference in norm acceptance based on whether the respondent/their father had a tertiary degree versus less than a secondary degree) or on the largest coefficient (respondent came to the country less than 20 years ago vs. respondent was born in the country).
Figure 2 demonstrates that the value orientations are the most important correlates of the norms of obeying laws, forming opinions, and supporting others.Stability" = Residential Stability; "Came to Cntry < 20 years Ago" compares respondents not born in the country but emigrated there less than 20 years ago to those born in the country; "Teritary Ed" compares individuals with a tertiary degree against those with less than secondary education (or, individuals with fathers that have these characteristics) The influence of the value orientations-and particularly self-transcendence vs. self-enhancement-remains impressive even for the other norms.The association between self-transcendence vs. self-enhancement is greater than any other variables' for the norm of volunteering, only outstripped by age for voting, and just shy of the influence of religiosity and interpersonal trust for being active in politics. 9The relative impact of the conservation values dimension is more modest in scope as noted earlier.The value dimension of self-transcendence vs. selfenhancement thus appears to be the most consistently important correlate with civic duty of the analyzed variables.
One potential reaction to this foregoing point is that value orientations may mediate the relationship between duty norms and variables such as age, education, and gender.If so, then it should not be surprising that we see a large effect for the values.We investigate this point in Online Appendix B where we compare the coefficients for the two value orientation variables from Model 3 with the estimated influence of demographics from models that do not include the value orientations.Notably, we find that the substantive importance of the value correlations, and particularly selftranscendence, continues to outstrip the influence of the demographics even when we focus on estimates of the latter's influence that should not be biased downwards by post-treatment bias.

Conclusion
An individual's normative conception of citizenship powerfully impacts their engagement with the political and civic life of their communities (Blais, 2000;Blais & Achen, 2019;Dalton, 2008a;Raney & Berdahl, 2009).Existing work on the origins of civic duty provides an important window into the role of social influence in understanding why individuals conform to group norms but does not fully explain why individuals vary in the norms they internalize.Here we provide such a theory by drawing on research concerning motivated reasoning to argue that norms are accepted to the extent that they cohere with an individual's core sense of self as indexed by their value priorities.We supported this argument with evidence from twenty European nations.We showed that the more an individual prioritizes selftranscendence values, and hence the less they emphasize self-orientation values, the more importance they placed place on a wide array of politically relevant citizenship norms.Individuals falling into the former category adopt the mantle of the active citizen and are likely to engage in a wide array of political acts.Conservation values, on the other hand, had a more nuanced relationship with civic duty.The prioritization of social order over personal autonomy may bolster the internalization of widely shared norms such as obeying laws or voting in elections but undermines the acceptance of a more widespread conception of political engagement as normative.The size of the correlation between values and norms was substantial and either 1 3 Political Behavior (2023) 45:1447-1466 rivaled or outstripped the association between education, age, social, trust, religiosity and these norms.
Our evidence that personal values are substantially correlated with civic duty norms stems from cross-sectional evidence.There are two threats to inference in this case: (1) omitted variable bias and (2) reverse causality.We attempt to address the former via a wide array of control variables, but of course this attempt is circumscribed by the real possibility of unmeasured confounders.We cannot readily address the latter possibility with the data at hand.We can, however, examine the relationship between personality traits and the duty to vote using the combined British Household Panel Study/Understanding Society panel survey; see Online Appendix B.10 Personality traits are also a component of a person's broader personality and hence may give another window into the self-related motivations discussed here.We show in Online Appendix B that changes in the traits of agreeableness and extraversion are associated with higher and lower levels of civic duty.These traits tend to correlate with self-transcendence and self-enhancement values respectively and hence this relationship is consistent with our results concerning this value dimension in Fig. 1.Changes in openness-to-experience and conscientiousness, which have a modest relationship with the conservation/openness-to-change values dimension examined in Fig. 1, were not statistically significant on the other hand.We offer some caveats to these analyses.Traits and values are only modestly interrelated and are conceptually distinct from one another (Parks-Leduc et al., 2015;Vecchione et al., 2019).In particular, traits are descriptions of a person's general behavioral patterns while values are more abstract cognitive representations of people's goals and motivations.As a result, various studies find that "traits and values have differential relationships with the motivational processes of goal content and striving" (Parks-Leduc et al., 2015, p. 24).We are thus hesitant to place too much weight on these analyses as tests of the relationship between personal values, or underlying self-related motivations, and civic duty norms.Future work could advance our understanding on this subject via panel studies with measures of both values and norms, thereby enabling stronger causal inferences through the use of cross-lagged or individual-level fixed effect models.
We use survey data collected in Europe in 2002.One important question is thus whether we should expect these patterns to change over time and/or to also emerge in other societal contexts.There are two potential reasons why these patterns might vary across time or place.First, if the strength of self-related motivations varies across places, then we would expect differences in the strength of the relationships seen here to also vary.On this point there is some reason to expect variation between European and non-European samples as self-related motivations do appear stronger among WEIRD (European) samples (Henrich et al., 2010;Sherman & Cohen, 2006).Second, if the meaning of these norms were to change in some important way over time, then this could also affect the relationships on offer by changing which values would be most relevant for predicting a particular norm.The past two decades have witnessed the rise of right-wing populist movements in Europe that criticize existing elites as corrupt and attempt to mobilize 'the people' against them via political action.Insofar as conservation values are associated both with support for such movements and feelings of national identity, then it may be possible that the relationship between conservation values and the norm of "being active in politics" has diminished with individuals at both ends of the scale feeling motivated to adopt the norm.Ultimately, an examination of either possibility requires new data that broadens the temporal and geographic study of values and civic duty.
In the remaining space we will comment on the implications of our work for understanding political participation and norm acceptance more generally.In the former case our study contributes to recent efforts aimed at exploring the motives underlying participation.For instance, Miller (2013) provides evidence that individuals who believe that political engagement serves the motive of value-expression are significantly more likely to participate.Miller, however, does not unpack the concept of value-expression, i.e. she does not explore which values are likely to lead to this belief.Our evidence suggests that not all values are equal on this front and that self-transcendence values are particularly likely to foster this belief.An alternative line of work in this general area focuses on altruistic preferences akin to our selftranscendence values (i.e.Fowler, 2006;Fowler & Kam, 2007) as well as sensationseeking which captures goals similar to those underlying openness-to-change (Kam, 2012).However, these studies do not directly compare the relative influence of these goals.Our work suggests that altruistic motives are a far more substantial influence than sensation seeking.
We conclude by explicating the implications we feel this study has for work on social norms more generally.As we've seen, individuals discriminate across citizenship norms based on their underlying value priorities.We believe this should be the case with social norms in general.While the social environment helps influence the salience and content of the norms confronting individuals, we suspect that norm adoption in social settings is influenced by the same process of value-congruency described here.The result should be a push and pull between values and social environments with the latter working best when in congruence with the former (and vice versa).This highlights the importance of focusing on the interaction between motives; congruent motives are likely to strengthen one's likelihood to adopt a norm (or attitude, or behavior), while the possibility of conflicting motives raises the question of how individuals prioritize their goals.This latter question remains in need of empirical attention, as Leeper and Slothuus (2014, p. 139) note in their recent review of motivated reasoning: "The consequences of these motivational interac-tions…are the subject of some theorizing in psychological literature but have not faced considerable empirical scrutiny.If political psychological work continues down the current trajectory of motivated thinking as a central object of study, these matters must eventually come to the forefront of theory building."We cannot offer an answer to this question, but we do believe that our study provides further impetus to consider such possibilities and that personal values deserve a central place in future theorizing on the subject.

Fig. 1
Fig. 1 Value Orientations and Social Norm Acceptance.Markers provide the coefficient (with 95% confidence interval) for both value dimensions across model specifications (left y-axis) for each citizenship duty norm (right y-axis).The DV and values are all scaled to range from 0 to 1. Full model results can be found in Tables OA3-OA8 in the Online Appendix

Fig. 2
Fig. 2 Comparing the Influence of Values and Other Individual Predictors.Markers provide the coefficient (with 95% confidence interval) for both value dimensions and control variables for each citizenship duty norm (right y-axis).The DV and continuous variables are all scaled to range from 0 to 1. Full model results can be found in Tables OA3-OA8 in the Online Appendix.This figure draws from Model 3. "Res.Stability" = Residential Stability; "Came to Cntry < 20 years Ago" compares respondents not born in the country but emigrated there less than 20 years ago to those born in the country; "Teritary Ed" compares individuals with a tertiary degree against those with less than secondary education (or, individuals with fathers that have these characteristics)

Table 1
Overview of expectations

Table 2
Summary statistics