Motivating Citizens to Participate in Public Policymaking: Identification, Trust and Cost-Benefit Analyses

Under what conditions do citizens of nations and states comply with governmental requests to participate in public policymaking? Drawing on the dual pathway model of collective action (Stürmer & Simon, 2004) but with a focus on compliance with the status quo, rather than participation in collective protest, two studies examined citizens’ motivation to participate in public policymaking. Study 1 (N = 169) was an MTurk hosted survey that recruited participants from California, while Study 2 (N = 198) was a field experiment that recruited participants in Sardinia, Italy. Study 1 measured cost-benefit analyses, societal identification, and willingness to participate in public policymaking. Study 2 repeated the same procedures, with the exception that we manipulated costs of participation, and also measured participants’ trust in government. Study 1 confirmed our initial hypotheses – fewer costs predicted more willingness to participate, as did stronger state identification. However, Study 2 found an interactive effect of costs, identification, and trust on willingness to participate in public policymaking. Results confirm our hypotheses by showing that both costs and identification independently influence willingness to participate in public policymaking. Results also add to the literature by showing that these additive pathways can be influenced by trust in the source of governance.

Initiatives that promote direct participation by citizens in governmental policies have been in place since the 1950s (Day, 1997), with the underlying assumption that citizens' direct participation in their government will promote more democratic and effective governance.Citizens are individuals who possess freedoms and rights within a certain governmental or community, while citizens' participation is when citizens enact these freedoms to better their community (Lister, 1998).Citizens' participation plays a prominent role in many community settings, ranging from work environments to public policies, as well as health and urban planning programs (Wandersman & Florin, 2000).The emergence of citizens' participation in public policymaking is largely dictated by the rationale that people desire policies that improve the quality of governmental decisions, and maximize community benefits (Box, 1997;Mannarini, Fedi, & Trippetti, 2010;Oldfield, 1990;Stivers, 1990).
Notwithstanding these premises, motivating citizens to participate in policymaking is still a challenge.This is evident within Western societies, where governments have observed declining rates of civic engagement -that is, declining rates of collaboration between citizens and their governmental institutions (e.g., Putnam, 1995;Stolle & Hooghe, 2005).Governments have also seen an erosion of citizens' trust in government and diminished willingness to participate in political actions (e.g., Alford, 2001;Hibbing & Theiss-Morse, 2002;Wattenberg, 2002).In this sense there is a paradox -declining participation in policymaking limits opportunities for governments to understand the needs and concerns of their constituents, yet it simultaneously accentuates the cleavage between citizens and politicians during a time of social change.
Given this paradox it is not surprising that many researchers have explored factors that motivate citizens to participate in policy making and collective action (e.g., Klandermans, 1997;van Zomeren, Postmes, & Spears, 2008).
Collective action is broadly defined as occurring when group members act on behalf of their group, with the action aimed at improving the group's overall conditions (Wright, Taylor, & Moghaddam, 1990;van Zomeren et al., 2008).
Social psychological research has previously examined bottom-up processes of collective action in which a smaller faction of people engage in collective action against a larger superordinate establishment (Papadopoulos, 1995).However, another process of collective action exists -a top-down process to collective action, in which people respond to an initiated request from their superordinate establishment to engage in collective action (Papadopoulos, 1995).While both sociologists and political scientists have examined top-down processes, social psychologists have not.
Drawing on social psychological research on collective action (Stürmer & Simon, 2004) and research on trust in governments and administrative institutions (Alford, 2001;Tyler, 1989;Uslaner & Brown, 2005), we examine people's willingness to participate in policymaking.

Citizens' Participation in Public Policymaking
Based on other psychological analyses of policymaking (e.g., Surel, 2000), we define public policymaking as a phenomenon in which citizens directly engage in governmental decision-making processes aimed at implementation of polices that affect citizens.Public policies are most often developed by politicians and larger government institutions, and then implemented for their citizens (Anderson, 2014).But while governments formulate these public policies, it is often citizens who decide whether or not a policy is actually implemented en masse.
Although governments promote their citizens to participate in policymaking, direct involvement in policymaking is often avoided (Antonini & Fini, 2011).One reason for little participation in policymaking is that despite enthusiasm for valuable and beneficial public policies, citizens frequently feel that their voice will simply not be heard, and Motivating Citizens in Public Policymaking 132 participation will provoke no changes within the government (Fowler & Kam, 2007).For example, in large populations the probability that a single act of participation will significantly affect the outcome is very small.From this rationalchoice perspective of participation, simply attaching value to the idea of democratic participation is not effective at fostering public engagement.
Rational choice scholars have instead approached people's motivation to participate in collective action and social movements by using models based on pure self-interest (e.g., Aldrich, 1993;Downs, 1957;Feddersen & Pesendorfer, 1996;Palfrey & Rosenthal, 1985) and expected value (e.g., Ajzen & Fishbein, 1980;Feather, 1982).These self-interest models argue that when participation either provides some type of personal benefits or implies low costs, citizens' willingness to participate is increased.Expectancy-value theory, however, states that a person's motivation to engage in political actions is a result of the value of the expected outcomes -the more likely the politically charged behavior will produce a valued outcome, the more likely they will engage.Olson's (1965) benefits/costs model falls in this tradition, such that the decision to engage in collection action is based on a rational evaluation of the benefits and personal costs due to participating.
Thus, some research on collective action suggests that citizens' willingness to participate in policymaking can be predicted by taking into account a rational, cost-benefit analysis of engagement.However, personal involvement in politics may not be solely ascribed to the role of rewards/costs expectancy of citizens' participation, suggesting that willingness to participate in policymaking could be influenced by other variables.

A Social Psychological Perspective on Public Participation
Both sociologists and social psychologists have effectively studied social movements.The former have analyzed people's participation in social movements as a function of the perceived costs and benefits of participation (e.g., Aldrich, 1993;Olson, 1965), while the latter have largely focused on the pivotal role played by social identity processes, specifically social identification, in collective action (e.g., Klandermans, 1997;Simon et al., 1998;Wright, 2001).
Social identity theory (Tajfel & Turner, 1979;Turner, Hogg, Oakes, Reicher, & Wetherell, 1987; also see Abrams & Hogg, 2010) claims that people's self-concept is derived from their knowledge of membership in social groups, and goes on to specify the cognitive processes that are involved in self-conception as a group member and in associated group behaviors and processes.Social identities are associated with distinct values and attitudes that influence how group members think and behave in various situations (Abrams & Hogg, 2010).Moreover, social identity research clearly shows that highly identifying group members engage in more group-oriented behavior than low identifiers, because for them the group is more self-relevant (e.g., Terry & Hogg, 1996;Terry, Hogg, & White, 1999).
Identification with a group plays a major role in fostering involvement in collective action when collective identity is salient and important.Individual cost-benefit analyses also play a role, but it is highly likely that the evaluation of relevant costs and benefits is itself configured by group identification, and shared identification lends uniformity to people's patterns of cost-benefit analyses (Kelly, 1993).
suggests that collective identification is a unique motivational pathway to participation, above and beyond the cost-benefit analysis.The relevant identity is a politicized collective identity that is ideologically grounded and associated with a disadvantaged collective within the political context (Simon & Klandermans, 2001;Stürmer & Simon, 2004).Lastly, this politicized collective identity is a dual identity, where people identify with both the marginalized subgroup (e.g., migrant workers) and the broader superordinate group within which they seek change (e.g., country of residence; Simon & Grabow, 2010;Simon & Ruhs, 2008).
Beyond this politicized identity, the dual pathway model shows additive, not interactive, effects of these pathways; specifically, both cost-benefits of participation and politicized identification independently predict collective action (Stürmer & Simon, 2004).This gives credence to the argument that costs-benefits expectations and collective identification processes are additive, rather than interactive, in their impact on political participation (cf., Kelly & Kelly, 1994).
Therefore, drawing on the two pathways that have consistently been shown to predict bottom-up collective action (Stürmer & Simon, 2004), we examined whether both a cost-benefits analysis and social identification can be applied to a top-down approach to collective action.Specifically, we examined whether the variables of the dualpathway model of collective action can be applied to citizens' participation in public policymaking based on a governmental request.

Citizenship: A Bridge Between Collective Action and Participation in Policymaking
Participation in social movements is enhanced by identification with a politicized disadvantaged or minority group (e.g., Simon et al., 1998;Simon & Klandermans, 2001;van Zomeren & Spears, 2009), and collective action tendencies are increased by perceived group-based inequality and injustice (e.g., van Zomeren, Postmes, & Spears, 2008).In these circumstances, people view collective action as the most effective way to address disadvantage.
However, perhaps the psychology is different in a top-down model of collective action, especially when applied to citizens' participation.This is because not all citizens who participate in public policymaking identify with a minority or a disadvantaged group; that is, not all group members who engage in this form of collective action are in the minority.Instead, all citizens will, to some degree, identify with the larger superordinate establishment (e.g., Gaertner & Dovidio, 2000).Hence, superordinate group identification, not politicized identification, may be a driving factor in top-down processes of collective action.
Moreover, this top-down approach is directly geared towards citizens of a particular territory.The social science literature on citizens' participation in politics (e.g., Gaventa, 2002;Lister, 1998;Mannarini, 2004) shows that people's sense of citizenship is closely related to their social identity, such that people's collective identity can be generated through a process of identification with geographical context of their residency (e.g., Altman, 1975;Sack, 1983;Stürmer & Kampmeier, 2003).Moreover, citizenship refers to the sense of belonging to a common socio-spatial entity that defines the behaviors and norms of those who identify with the territory (Mannarini, 2004) -in the language of social identity theory (Tajfel & Turner, 1979;Turner et al., 1987), a sense of belonging to a larger territorial community can strongly shape people's self-concept.Overall, while the cost-benefit's pathway is motivated by extrinsic rewards or costs avoidance, the identification-citizenship pathway is motivated by the internalization of group norms, values, and customs.
Overall, we believe citizens' participation in policymaking is influenced by both a cost-benefit dynamic and collective identification, as proposed by the dual pathway model of collective action (Stürmer & Simon, 2004).However, if the relationship between citizens and their government influences citizens' collective identification with the superordinate establishment, this will also influence their willingness to engage in collective behavior.Because topdown processes of collective action often involve responding to a request from one's government, we believe citizens' trust in their government can influence their willingness to engage in public policymaking.

Trust in the Government: A Bridge Between Two Pathways to Citizens' Participation
For many social movements, government administrations are perceived as obstacles to obtaining desired resources and at fault for many disadvantages.In turn, due to mistrust, the government itself often becomes a target of the social movement.This can be problematic in public policymaking, where governments often encourage their citizens to participate in policymaking.In this type of top-down collective action process, citizens need to trust their government administration in order to participate in public policymaking.Thus, because trust in authorities is an important psychological antecedent of some collective behavior (e.g., De Cremer & Van Vugt, 1999;Tyler, 1989;Van Vugt & De Cremer, 1999), we, lastly, examine the role played by trust in citizens' motivation to participate in public policymaking.
Citizens' trust in government benefits the government itself -policies, directives, 'moral suasion', tax collection and attempts at income redistribution are likely to be more effective if the population trusts the government to be working in their interests (Fjeldstad, 2004;Hetherington, 1998).Governments that face mistrust and suspicion may discover that their citizens ignore and resist their appeals, and are suspicious of their pronouncements and policies (e.g., Ayres & Braithwaite, 1992;Braithwaite & Makkai, 1994).
Previous literature shows that trust differently influences political participation, depending on whether it is a bottomup or top-down participation process.Specifically, in bottom-up processes of political participation it is distrust, not trust in the government, that elicits participatory behavior (Easton, 1967;Gamson, 1971).In line with research on collective action (van Zomeren et al., 2008), people who perceive their government to be oppressive or unfair may seek out social movements that protest government oppression and inequity (e.g., Gamson, 1971).
However, more relevant to this paper is a top-down process of political participation, which states that trust should elicit more participatory behavior than distrust (Almond, 1988;Uslaner & Brown, 2005).Specifically, citizens who trust their government are more willing to comply to a request from them in order to achieve "public results".Research on top-down processes of political participation shows that citizens are more compliant with demands and requests when perceiving their government to be trustworthy (e.g., Levi, 1989Levi, , 1997;;Levi & Stoker, 2000;Tyler, 1989Tyler, , 2006)).Overall, a strong link exists between citizens' trust in their government and their willingness to comply with requests from government.
Moreover, while trust is related to social identification (cf., Tyler, 1989), research suggests it is nonetheless a unique source of influence (cf., Tyler, Degoey, & Smith, 1996).Thus, along with examining the effects of costbenefits analyses and group identification, we examined how trust in government influences participation in policymaking.

Overview of Studies
Two studies were conducted simultaneously to test this general idea; Study 1 in California and Study 2 in Sardinia, Italy.These two socio-political contexts were chosen because they differ quite dramatically in the sort of relationship that citizens have with government, specifically in terms of the extent to which citizens have a direct say in policy.California's system has extensive ballot initiatives that promote its citizens to engage in policymaking, whereas these types of initiatives and promotions are very rare in Sardinia.Thus, to account for the different relationships people have with their government, we conducted our two studies in unique, cultural contexts.
Study 1 was a survey design to confirm our initial theorizing that the dual-pathway model of collective action can be applied to a top-down process of collective action; specifically, we measured both cost-benefits and state identification, with the hypotheses that lower costs of participation would predict more willingness to participate in public policymaking (H1a), as would stronger Californian identification (H1b).However, these pathways will be additive, such that these predictor variables will produce two main effects but no interactive effect.
After confirming our initial hypothesis that a top-down process of collective action can be examined with the dual pathway model, we conducted our second study -an experiment to examine whether trust in the government would transform these additive pathways into an interactive effect.Specifically, while high costs of participation were expected to be negatively related to willingness to participate, we hypothesized that high trust in government and high state (Sardinia) identification would cancel out this effect and instead increase willingness to participate in public policymaking (H2).

Participants and Design
Participants were residents of California who were recruited through Amazon's Mechanical Turk (MTurk) -an online crowdsourcing website that allows workers to complete surveys and tasks for monetary compensation.
Research shows that MTurk is a valid and reliable source of on-line data collection (Buhrmester, Kwang, & Gosling, 2011;Mason & Suri, 2012).All participants had a California IP address, resulting in 169 participants (106 males, 63 females), who were predominately European-American (60.40%;Asian-American 26.60%; Mexican-American 11.80; %; African-American 3.00%; Other 1.20%) with an average age of 31.71(SD = 11.43)years.Participants were told to role-play that the California government had requested them to participate in public policymaking.
Participants then completed measures of costs-benefits of participation, state identification, and willingness to engage in political participation.Demographic data (age, gender, and race) were collected at the end of the study before participants were debriefed and compensated 30 cents for their completion of the survey.

Procedure and Materials
After obtaining informed consent, participants were told that the study examined Californians attitudes towards political decision-making.After answering one question to confirm their Californian residency, participants were asked to carefully imagine the Californian government had asked them to vote in an upcoming public policymaking ballot.
Participants then responded to four statements (1 = not very much, 9 = very much) measuring the expected costsbenefits of participation in public policymaking.They were asked to think about the difficulties and costs involved in political participation before indicating how much each of four factors influenced their decision-making: (1) 'The feeling of frustration from lacking "real" results from participation'; (2) 'The need to give up personal and family members for participation'; (3) 'Interpersonal conflict with others during participation'; (4) 'Any other kind of costs you imagine' (M = 5.07, SD = 1.75; α = .83).These statements were adapted from previous research examining the costs of political participation (Wandersman, Florin, Friedmann, & Meier, 1987).
Participants then completed a measure of willingness to participate in public policymaking, by responding to three

Discussion
Study 1 examined whether the dual-pathway model of collective action could be applied to a top-down model of citizens' participation.We asked participants to complete measures of collective identification -which we conceptualized in terms of citizenship -and costs-benefits expectations before indicating their willingness to participate in public policymaking.
Consistent with Stürmer and Simon's (2004) model, stronger collective identification predicted willingness to participate in public policymaking (H1a).Simultaneously, a costs-benefits analysis (i.e., lower costs, higher benefits) uniquely predicted willingness to participate in public policymaking (H1b).Therefore, the two pathways showed an additive effect, such that they produced two main effects but no interactive effect on willingness to participate.
Overall, Study 1 shows that the dual-pathway model of collective action can be applied to top-down approaches of collective action and to citizenship participation.
Although we measured identification with the larger superordinate establishment (i.e., California) in Study 1, it is unclear whether participants' relationship with their government influenced willingness to participate in public policymaking.Within California, citizens are commonly requested by local and state governments to participate in public policymaking.Thus, even with low rates of participation, citizens can trust that participation in public policymaking is efficacious.
However, not all states or countries have similar governance to California.In many countries it is uncommon for citizens to receive participatory requests for public policymaking from their governments.Under these circumstances it is important to account for citizens' trust in their superordinate establishment (i.e., government).Thus, because Study 1 showed that the dual-pathway model of collective action (Stürmer & Simon, 2004) can effectively explicate top-down processes of citizens' participation, we conducted a second study to examine how trust in government effects this process in a setting where governmental requests for citizens' participation is uncommon: Sardinia, Italy.

Results
There was one manipulated variable (costs-benefits of participation), two continuous variables (identification, trust in government), and one continuous dependent variable, willingness to participate in public policymaking (see Table 3 for correlations between variables).The manipulated variable was coded as 0 (low costs of participation) and 1 (high costs of participation).Similar to Study 1, all of the predictor variables, both manipulated and measured, were mean-centered before creating interaction terms and conducting a multiple linear regression -See Table 3 for correlations.However, and as predicted, the three-way interaction of costs-benefits, identification and trust in government increased citizens' willingness to participate in public policymaking, β = .17,t(191) = 2.41, p = .02.Upon examining the simple slopes it was found that state identification and costs of participation were moderated by trust, such that high identification and high costs (low benefits) increased willingness to participate only when trust in govern-ment was high, β = .17,t(191) = 2.34, p = .02;however, high identification and low costs (high benefits) did not predict willingness to participate when trust in government was low, β = .06,t(191) = -.83,p = .40.Moreover, high costs (low benefits) decreased willingness to participate in public policymaking for high identifiers who had low trust in government, β = -.15,t(191) = -.2.04, p = .04.Lastly, even when having high trust in government, high costs (low benefits) decreased willingness to participate for low identifiers, β = -.18,t(191) = -2.47,p = .01.No other simple slopes were significant (ps > .20)-See Figure 1.

Discussion
Study 2 retested whether the dual-pathway model of collective action could be applied to citizens' participation in a different socio-political context: Sardinia, Italy.We also examined whether trust in government influenced willingness to participate in public policymaking.However, unlike Study 1 we manipulated cost-benefits of participation (high, low) before measuring collective identification and trust in government.We lastly measured willingness to participate in public policymaking as the dependent variable.
Consistent with Stürmer and Simon's (2004) model, stronger collective identification predicted willingness to participate in public policymaking (H1a).Simultaneously, a costs-benefits analysis (i.e., lower costs, higher benefits) uniquely predicted willingness to participate in public policymaking (H1b).However, trust in government had no predictive effect on willingness to participate in public policymaking.There was also no interaction between collective identification and costs-benefits.
Instead, and as predicted, a three-way interaction emerged between collective identification, cost-benefits analysis, and trust in government on willingness to participate in public policymaking.Results showed that when trust in government was high, perceiving high costs (low benefits) and having strong identification predicted more willingness to participate.However, neither cost-benefits nor identification predicted more willingness to participate when trust was low.Overall, Study 2 shows both that the dual-pathway model of collective action can be applied to top-down processes of political participation, and that trust in the government may influence this process.
how identification with a disadvantaged group predicts collective action; instead, our studies focused on superordinate (i.e., state/region) identification due to investigating a top-down approach (i.e., governmental request) to collective action.In this regard, we believe our work can be applied to top-down approaches to collective action.
Future research should examine whether these variables apply to other groups (e.g., workplace organizations, academic institutions) that request collective action and participatory behavior.
A second possible limitation relates to our costs-benefits of participation measure.We assessed the costs-benefits of participation with an adaptation of Wandersman and collaborators' scale (1987), which is particularly focused on costs evaluations.We also modified an item of this scale to ensure a focus on the benefits of participation.
Future research on the same topic could provide an improved operationalization of this variable.
Besides these limitations, we believe this research has implications for social psychological literature on collective action.Previous research has focused on how perceived injustices or politicized identities predict willingness to engage in collective action (Stürmer & Simon, 2004;van Zomeren et al., 2008).However, because collective action can apply to any group-based behavior performed by a member for their group (van Zomeren et al., 2008), we believe other types of identities can predict collection action.Thus, by showing stronger state identification predicted willingness to engage in collective action (i.e., participation in public policymaking), we believe this research adds to the literature on top-down models of collective action.
We also believe this research has far reaching implications for public policymaking, and can contribute to debates concerning the dynamics of citizens' participation in governance and policy.Governments should be aware of why their citizens do or do not comply with participatory requests for public policymaking.This awareness of citizens' participation can help governments and public administrators implement methods that effectively elicit participatory behavior.Our findings offer the practical suggestion for governments to minimize the costs of citizens' participation, and find ways to strengthen identification with the state/region where they live.Moreover, public institutions that encourage citizens' participation should account for whether their citizens trust them; specifically, if trust in the administration is low and participation in policymaking seems costly, citizens might tend to invest their time and energy elsewhere.Thus, governments need to find ways to address all three of these aspects simultaneously.
statements measuring likelihood of engaging in public policymaking (1 = not very much, 9 = very much): (1) 'I am willing to vote in California state elections'; (2) 'I feel motivated to keep myself updated on California propositions'; and (3) 'I am willing to vote in California ballot propositions' (M = 7.20, SD = 1.85; α = .90).As no research has directly examined participation in public policymaking, these questions were constructed for the purposes of this study.

Figure 1 .
Figure 1.Three-way interaction of identification, costs, and trust on willingness to participate in public policy making.Left: High trust; Right: Low Trust.Note.ID = Group identification (± 1 SD).
being given this role-playing prime, participants answered six questions measuring group identification (1 = not very much, 9 = very much): (1) 'To what extent do you feel you have strong ties with Californians'; (2) 'How similar do you feel you are to other Californians'; (3) 'How much do you feel you identify with California'; (4) 'How proud are you to be a Californian'; (5) 'To what extent do you feel a sense of belonging with California'; and (6)

Table 2
Study 1: Hierarchical Regression for Willingness to Participate in Public Policymaking

Table 3
Study 2: Correlations Between Predictor and Dependent Variables

Table 4 .
As predicted by the dual-pathway model, higher costs-high benefits of participation significantly predicted less willingness to participate in public policymaking, β = -.16,t(191) = -2.41,p = .02.Stronger Sardinian identification also predicted more willingness to participate in pubic policymaking, β = .34,t(191) = 4.76, p < .001.However, trust in the government did not predict of willingness to participate in public policymaking, nor did any of the two-

Table 4
Study 2: Hierarchical Regression for Willingness to Participate in Public Policymaking