How citizen coproducers cope with public value creation conflicts: a survey experiment

Abstract

Purpose

In coproduction, citizens may be confronted with a conflict between creating user value and a more collective understanding of public value creation. In order to deal with conflicts experienced as trade-off situations, coproducers follow various coping strategies leading to different results. This study aims to gain insight into what drives the choices for coping strategies, which are valuable for understanding the role of citizen coproducers in public value creation.

Design/methodology/approach

This article studies the effects of citizens' external efficacy and trust in public servants on citizen coproducers' preferences for coping strategies. The study presents a vignette experiment among n = 257 citizens involved in the temporary use of vacant spaces in Flanders, Belgium.

Findings

No statistically significant effects of external efficacy and trust in public servants on respondents' preferences for coping strategies are found. The results show that irrespective of the level of external efficacy or trust in public servants, citizen coproducers prefer to ask for help from the public servant involved in the project.

Originality/value

This result draws attention to the need for facilitation and guidance from public servants and the servants' organizations to help citizen coproducers balance out these otherwise paralyzing value conflicts. Moreover, the lack of statistically significant effects of trust and external efficacy is a valuable finding for literature. The result shows that, in the drivers of coproduction behavior, there is no consistent relationship between citizen trust in government or external efficacy and coproduction behavior.

Citation

Jaspers, S. and Migchelbrink, K. (2023), "How citizen coproducers cope with public value creation conflicts: a survey experiment", International Journal of Public Sector Management, Vol. 36 No. 3, pp. 236-252. https://doi.org/10.1108/IJPSM-04-2022-0089

Introduction

Public servants invite citizens to coproduce services to create better outcomes and public value from public service delivery (Bovaird, 2007; Bovaird and Loeffler, 2012). One of the remaining conundrums in public value creation theory is “the balance between individual and social/public value in public service delivery” (Osborne, 2017, p. 229), as these are potentially at conflict and regular service providers are burdened with the task of balancing both when aiming to create better outcomes (Bovaird, 2007; Needham and Carr, 2009; Brandsen and Helderman, 2012). More recently, empirical studies on this balance have grown (e.g. Rossi and Sanna, 2019). Studies show that in coproducing public services, citizens, similar to other service producers, can be confronted with a conflict between user value for the coproducer and social value for the broader community (Jaspers and Steen, 2021). However, little is known about how citizen coproducers cope with these conflicts and insights into this process could be helpful for theory and practice. This article studies the preferences of citizens for four often-mentioned coping strategies: two trade-off strategies, an asking-for-help strategy and an avoidance strategy (Steenhuisen, 2009; de Graaf et al., 2016).

In the organizational and political psychology literature, scholars have focused on personality traits and individual differences to explain how individuals behave when facing value conflicts (e.g. Van Lange, 1999). We extend this line of research and examine two personality traits as determinants for coping strategies of citizen coproducers in value conflicts: trust in the public servant facilitating the coproduction activity and external efficacy. First, trust is a beneficial psychological construct in solving value conflicts (Dawes, 1980; Kramer et al., 1996; De Cremer et al., 2001). Second, external efficacy, or an individual's perceived impact on participation, is identified as a driver of pursuing a desired outcome (e.g. Verba and Nie, 1991) when faced with value conflicts.

This article examines the effects of citizen coproducers' trust in public servants and external efficacy on their choice for coping strategies when dealing with a public value conflict between user value and social value.

We present the findings of a vignette experiment on the effects of external efficacy and trust in a public servant on citizen coproducers' preference for coping strategies when dealing with a perceived conflict during coproduction activities. We surveyed (n = 257) citizen coproducers from these projects and randomly assigned them to evaluate a vignette of a coproduction value conflict between creating user value and social value in which levels of external efficacy and trust in public servants were systematically manipulated. Participants were asked to rank four coping options: a trade-off in favor of social value creation, a trade-off in favor of user value creation, asking for help from a public servant and avoiding the conflict situation. We analyzed the results of this experiment using rank-ordered logistic regression (ROL) analysis. The results show that irrespective of the level of external efficacy or trust in public servants, citizen coproducers prefer to ask for help from the public servant involved in the project.

Coproduction and the cocreation of public value

Brandsen and Honingh (2015, p. 428), define coproduction as “the direct input of citizens in the individual design and delivery of a service during the production phase”. We follow their definition. Scholars have identified various forms of coproduction activities (Bovaird, 2007; Meijer, 2014; Brandsen and Honingh, 2015; Nabatchi et al., 2017). Although scholars distinguish between “coproduction”, taking place during the implementation phase of a public service and “cocreation”, taking place in the service design phase (Voorberg et al., 2014; Brandsen and Honingh, 2018, p. 13), we follow Bovaird and Loeffler's (2012) suggestion to use the term coproduction for citizen-government collaboration throughout the policy cycle (when an initiative includes both the codesign and the codelivery of a service). We also follow the conceptualization of enhanced coproduction, which conceptualizes coproduction where citizens and service users are given a more fundamental role for codesigning and cocreating services (cf. Strokosch and Osborne, 2016).

The aim of coproduction is similar to that of regular service delivery: to create public value (cf. Nabatchi et al., 2017). The concept of public value (singular) (cf. Moore, 1995) represents the added value created through the activities of public organizations and officials and is “sometimes presented in terms of normative aspirations for a ‘good society’” (Hartley et al., 2017, p. 672; Stoker, 2006; Rutgers, 2015). It can be distinguished from the concept of “public values”, which refers to values that represent the “normative consensus about the rights [and obligations of citizens …] and the principles on which governments and policies should be based” (Bozeman, 2007, p. 13). Following the definitions of coproduction and public value creation, we refer to public services as being coproduced and public value as being (co)created.

Coproduction brings a specific context to public value creation. Multiple actors, including citizens, cocreate public value in the coproduction of a service (Bryson et al., 2017). Bovaird and Loeffler (2012, pp. 1126–1127) provide a framework for public value in the context of coproduction where they identify different dimensions of public value created by the public sector:

1. User value;

2. Value to wider groups (such as family or friends of service users or individuals who are indirectly affected);

3. Social value (creation of social cohesion or support for social interaction) and

4. Environmental value (ensuring environmental sustainability of all policies).

5. Political value (support to democratic process, e.g. through coplanning of services with users and other stakeholders)

According to Bovaird and Loeffler (2012), it is likely that all of these dimensions are part of the desires and motivations of citizen coproducers for creating value. Citizen coproducers could be motivated to coproduce because of their desire to ensure high levels of user value. Likewise, environmentally conscious coproducers may be motivated to ensure the environmental sustainability of a particular service. Conversely, community-conscious coproducers may be focused on facilitating social inclusion and producing outcomes that benefit the “widest possible range of local community members” (p. 1127). These desires for public value creation may be experienced as conflicting.

Value conflicts

Bovaird and Loeffler (2012) argue that politicians need to balance and prioritize the abovementioned different dimensions of public value, as conflicts with certain interest groups are likely to occur. Just as public servants have been found to experience value conflicts between creating user value for specific (groups of) individuals and creating other dimensions of public value in regular service provision (Rainey, 2009; Alford, 2014, 2016; Osborne et al., 2016; Farr, 2016), citizen coproducers also face similar dilemmas arising from the coproduction process (Jaspers and Steen, 2021). Coproducers engaged in coproduction can have different goals or mixed desires about the type of public value they wish to create (e.g. Van Eijk and Steen, 2014; Uzochukwu and Thomas, 2017). This can lead to experiencing some of these aims as incompatible. Individuals experience a conflict between the dimensions of public value when they perceive that creating different facets of value is not possible (Brandsen and Helderman, 2012).

Coping

When individual co-producers experience value conflicts do they arrive in a state of paralysis? Recently, studies have shown that co-producing actors, including project coordinators and citizen co-producers adhere to similar coping strategies as those found and identified among street level bureaucrats (Aschhoff and Vogel, 2018; Jaspers and Steen, 2019). The concept of “coping” refers to the strategy people adopt to deal with value conflicts (Lipsky, 1980; de Graaf et al., 2016). Steenhuisen (2009, p. 20) defines coping as “a response to competing values that takes form in […] actions and decisions”. The street-level bureaucracy literature identifies a number of ways in which public servants deal with value conflicts on a day-to-day basis (e.g. Thacher and Rein, 2004; Stewart, 2006; de Graaf et al., 2016). We can group these strategies into “trade-off strategies” and “balancing strategies”. Trade-off strategies are mono-value solutions where the focus is placed on the creation of just one value. Balancing strategies are multivalue solutions in which two conflicting value dimensions are created (Steenhuisen and de Bruijne, 2009).

Thacher and Rein (2004) and Stewart (2006) provide several examples of trade-off and balancing strategies. Thacher and Rein (2004) first identified a casuistry strategy (where individuals trade-off a value based on how they traded off during a previous conflict experience). Stewart (2006) then identifies two more trade-off strategies: a bias strategy (where one value is preferred over the other) and an incrementalism strategy (where individuals gradually increase the emphasis on one value). With regard to balancing strategies, Thacher and Rein (2004) identify a strategy of building firewalls (where individuals and organizations appoint different institutions or positions to support each value) and a cycling strategy (where coping is carried out by paying sequential attention to each value). Finally, Stewart (2006) identified a hybridization strategy where individuals seek a balance between competing value dimensions.

The literature (Tetlock, 2000; de Graaf et al., 2016) also points to other coping strategies that cannot straightforwardly be grouped into trade-off or balancing strategies. We add two additional coping strategies based on that literature: an asking-for-help strategy and an avoidance strategy. Asking for help is a strategy by which people “escalate” questions about competing values to a higher administrative or legislative authority (Tetlock, 2000; de Graaf et al., 2016). Theoretically, escalating goes further than help-seeking behavior (e.g. Bamberger, 2009) and implies “passing the buck”, meaning an individual will ascribe one's own responsibility to another person or group (Steenhuisen, 2009). Alternatively, an avoidance coping strategy prevents citizen coproducers from dealing with the conflict completely (Endler and Parker, 1990). Since some citizens feel that they lack the capacity to cope with the conflict, they may opt to avoid dealing with it and consciously decide to ignore that the value tension (Steenhuisen, 2009). This can be done by putting off addressing the conflict and avoiding making a decision, for example by removing oneself from the situation where the conflict emerges (Endler and Parker, 1990).

What drives the choice of coping strategies?

In the early coping literature, Hirschman (1970) identified two conditions for following strategies that aim at actively solving a conflict situation rather than neglecting it: a person's perceived ability to have an impact, or external efficacy and a person's loyalty or trust toward an authority. We chose to further test these variables because they are both drivers of coproductive behavior (Parrado et al., 2013; Van Eijk and Steen, 2014; Bovaird et al., 2015; Thomsen, 2017; Clark and Brudney, 2019) and are well recognized in the psychology literature for driving behavior in situations that are stressful or characterized by conflict (Bandura, 1986; Lewicki et al., 1998).

External efficacy

External efficacy refers to individuals' “perception of the results of their participation in the service provision” (Kristensen et al., 2012, pp. 4–5). In discussing the external efficacy of citizens, Jo and Nabatchi (2018, p. 235) refer to an individual's “belief that professionals will make room for participation and be responsive to input”. External efficacy is a situational trait-like disposition (Schneider et al., 2014), which means that individuals already involved in coproduction may reevaluate their perceived impact based on new developments or situations in the coproduction process. This reevaluation provides the individual with new information about their perceived impact, which may influence their behavior (Ertiö et al., 2014; Clark and Shurik, 2016).

Van Eijk and Steen (2016) identify external efficacy as one of the main factors driving citizens' efforts to coproduce. They find that low external efficacy is related to low motivation to coproduce. This is in line with political psychology literature describing that people participate little in policy-making if they frequently feel that their voices are not heard (Fowler and Kam, 2007; see also: Iyengar, 1980). Thus, external efficacy is the situational self-measure about the room given to create an impact. When a person considers that their impact is not getting much “room” or their voice is not heard, they may choose to avoid putting in more effort, which leads to dropping out or perhaps neglecting the issue (Fowler and Kam, 2007). Applying this to coping with value conflicts in coproduction, citizens are low externally efficacious when they believe their coping strategy will not have an impact. Therefore, we formulate the following hypothesis:

H1.

Citizen coproducers' preference for avoidance as a coping strategy is higher when their external efficacy is low.

The political participation literature conceptualizes external efficacy as a psychological resource that equips people to engage in more help-seeking behavior (Verba and Nie, 1991). In this study, we examine asking for help from a specific administrative authority, a public servant, since in coproduction projects, the public servant is often the administrative authority (i.e. he/she holds some power, but not always, over administrative coordination, ownership over the coproduction location, or assigning the budget) with whom citizens directly interact. In other words, individuals who are highly external efficacious, are more likely to take matter into their own hands, rather than leaving the problem unresolved. Following this logic of highly efficacious co-producers behaving more as policy entrepreneurs, they are more likely to ask for help (Schneider et al., 2014; Clark and Shurik, 2016; Jo and Nabatchi, 2018; Callaghan and Sylvester, 2019). In this case, asking for help enhances the pursuit of outcomes as citizens actively engage in the process, for example by asking for help (Verba and Nie, 1991; Van Eijk and Steen, 2016). We expect that external efficacy may have a positive effect on individuals' preferences to seek help from this public authority. We thus propose the following hypothesis:

H2.

Citizen coproducers' preference for asking for help from a public servant as a coping strategy is higher when their external efficacy is high.

Trust

Trust is seen as one of the key conditions for collaborative practices as it reduces complexity and transaction costs (Yamagishi and Cook, 1993; Ostrom, 1988). The trust literature indicates that trust is an indicator of coping behavior in conflict situations (Kramer et al., 1996; Balliet and Van Lange, 2013). Following Mayer et al. (1995, p. 712), trust can be defined as “the willingness of a party to be vulnerable to the actions of another party based on the expectations that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party”. Thus, trust in others plays a potential role in cooperative behavior in mixed-motive situations (De Cremer et al., 2001).

First, the psychology literature claims that trust is positively related to cooperation, especially in the case of a conflict between user value and other dimensions of public value (Balliet and Van Lange, 2013). When trust is high, people have confidence in each other's goodwill and engage in reciprocal cooperation (e.g. Ring and van de Ven, 1994). In other words, trusting people are more willing to compromise and collaborate with others because of a feeling that they can trust the others. Balliet and Van Lange (2013, p. 1102) show that trust matters in a conflict between self-interest and benevolent motives and does so increasingly in cases of increasing conflict. Trusting the other person and believing that one can rely on the benevolent intentions of the trusted person, thus, are predictors of cooperation in conflict situations (Balliet and Van Lange, 2013). Moreover, trust enhances the desire of people to strengthen and include new members in existing social networks pointing towards to the desire to create social value (Fukuyama, 1995). Trust then acts as a mechanism to be more welcoming to the desires of others, and in case of an experienced conflict between social value creation and individual value creation, citizens with higher trust can be more likely to opt for the social value creation than for their own individual value creation. Combining both the psychology literature and the political participation literature, we hypothesize the following:

H3.

Citizen coproducers' preference for a trade-off in favor of social value creation as a coping strategy is higher when their trust in the public servant is high.

Another characteristic of trust is that it involves the expectations of the trusted person (in our case, a public servant) to act benevolently toward the trusting person when there is a conflict between oneself and the collective interests (e.g. Mayer et al., 1995; Rousseau et al., 1998; Yamagishi, 2011). This means that in the case of conflict, trust enhances the risk-taking behavior of the trusting person. This risk-taking behavior is characterized by vulnerability, such as help-seeking behavior from a more trustworthy authority. Therefore, we hypothesize the following:

H4.

Citizen coproducers' preference for asking for help from a public servant as a coping strategy is higher when their trust in the public servant is high.

Finally, building on the conditions that Hirschman (1970) formulated, when trust in a specific authority is absent, individuals may show neglect and have a higher likelihood of coping using an avoidance strategy. Organizational behavior studies have corroborated this mechanism. Lewicki et al. (1998) state that people with low trust levels believe that other group members provide little insight into or relief from the project, which results in the avoidance of voicing concerns. This means that people with lower levels of trust do not think that others are able to help and therefore opt for an avoidance strategy. We hypothesize the following:

H5.

Citizen coproducers' preference for avoidance as a coping strategy is higher when their trust in the public servant is low.

Methods

Temporary coproduction in vacant spaces

A lot of coproductive activity takes place in the coproduction of temporary vacant spaces. In the temporary use of vacant spaces, the government often, but not solely, makes empty buildings and lots available or acts as a legal mediator when the building is made available by a private actor. Here, many different types of public services are coproduced, such as the coproduction of urban planning and spatial policies, community building, social care programs, youth policy and innovative social experiments. Often, these spaces encompass a variety of activities, all designed and implemented by citizen coproducers. It forms an ideal case for studying coping with value conflicts because these conflicts are likely to arise due to the plurality of aims of coproduction in the temporary use of vacant spaces (see also Jaspers and Steen, 2020).

Experimental design

The objective of this study is to test the effects of external efficacy and trust in public servants on citizen coproducers' preferences for coping strategies when confronted with value conflicts between creating private user value and social value. Experimental research is well suited to studying individual preferences at the micro-level since they allow us to study the effects of people's attitudes and behaviors (James et al., 2017). Therefore, we choose a robust, non-observational experimental procedure to test the effects of external efficacy and trust in public servants on preferences for coping strategies. To apply this method to a large and diverse sample, we chose an online vignette experiment.

The vignette

Vignette experiments allow us to test the effects of personal attitudes without having to measure actual behavior (Bevir, 2006). A vignette is “a short textual description of a situation that represents a systematic combination of theoretically determined characteristics” (Atzmüller and Steiner, 2010, p. 128). Our experiment contains four vignettes with identical textual descriptions. We include two factors in the vignettes that represent citizen coproducers' external efficacy and trust in public servants and manipulate these characteristics at two levels (high vs. low; e.g. a 2*2 full-factorial design). Because of this use of multiple factors in a single experimental design, vignette studies enable more valid and realistic scenarios to be tested than classic single-factor experiments (Mee, 2009). This design allows us to estimate the effects of external efficacy and trust in public servants independent of each other in the same experiment.

The respondents were randomly assigned to evaluate one vignette in which a fictional citizen coproducer protagonist named Sam was confronted with a value conflict between creating user value and social value, characterized by either high external efficacy and high trust in public servants, low external efficacy and high trust in public servants, high external efficacy and low trust in public servants, or low external efficacy and low trust in public servants. In the design of the vignette, we present a hypothetical situation featuring a fictional character to probe the opinion of the respondents (Hughes, 1998). In (Figure 1), the protagonist Sam is involved in coproducing a temporary gardening project in which he is personally invested (creation of user value) and is confronted with the demand to organize more activities for the people of the neighborhood (creation of social value). The vignette continues by illustrating how Sam is experiencing this as a conflict between the creation of user value and social value. The respondents are then invited to imagine how they would behave if they were in the same situation that Sam is in by ranking four coping strategies.

The response categories were presented in random order. A trade-off in favor of user value creation was operationalized as follows: “In Sam's situation, I would not organize additional events for the neighbors because it would not directly benefit my gardening project”. Trading-off in favor of social value creation was operationalized as: “In Sam's situation, I would try to organize more events for the neighborhood because it is important, even though this is at the expense of my own gardening project”. Asking for help was operationalized by emphasizing asking for help from a public servant: “In Sam's situation, I would ask advice from a public servant for how to cope with the conflict between organizing events for the neighborhood and executing my own project”. Finally, for the avoidance strategy, we opted for a realistic operationalization of abstaining from coping: “In Sam's situation, I would do nothing and wait”. Before conducting the full experiment, we piloted the vignette experiment using hard copies (N = 9) [2] and an online pilot survey among a subsample of the target population (N = 55) [3]. The survey experiment was conducted using Qualtrics (2019). The respondents provided their informed consent before they could participate in the study.

Independent variables

We operationalized high external efficacy as “Sam has the feeling that people listen well to what he has to say” and low external efficacy as “Sam has the feeling that people do not listen to what he has to say”. We operationalized trust as the trust an individual has toward a specific person. Trust is therefore operationalized as “Sam trusts the public servant's intentions”. Low trust levels are operationalized as “Sam does not trust the public servant's intentions”.

Data sampling

We compiled a unique dataset of Flemish citizens who were engaged in the coproduction of temporary use of vacant spaces projects between March and November 2019. We contacted all 300 Flemish municipalities and cities and asked them via e-mail and/or telephone whether projects concerning the temporary coproduction of vacant spaces existed in their municipality to obtain a contact list of all citizen coproducers of temporary spaces in Flanders. We obtained the contact details of the project coordinators of the reported temporary use projects (n = 95) and relied on their collaboration (due to privacy regulations) to forward the invitations to the experiment to all citizen coproducers in their project. Most project coordinators of the identified temporary use projects responded positively and we asked the coordinators to report how many temporary users they had forwarded an invitation to take part in our survey questionnaire (n = 1,069) to determine the sampling frame.

Data analysis

Statistically, we examined the effects of external efficacy and trust in public servants on respondents' preference for a coping strategy using a ROL analysis (Croissant, 2020). By asking respondents to provide a complete ranking of all four presented coping alternatives instead of only the most preferred strategy, we can make a more efficient estimation of respondents' preferences for coping strategies than we would have done by using a standard discrete-choice model in which only the most preferred alternative is analyzed (Fok et al., 2010; Croissant, 2012). We recoded the vignette manipulations in two orthogonal indicator variables (one for trust in the public servant and one for external efficacy). Following Mee's (2009) recommendations for the analysis of full-factorial vignette experiments, we coded each factor's high-level manipulation as 1 and each factor's low-level manipulation as −1. The design of the experiment has been approved by the institutional review board (Sociaal-Maatschappelijke Ethische Commissie) of the KU Leuven under project number {G – 2017 08,892}.

Fielding

We fielded the vignettes from October 1 to November 13, 2019. Of the n = 1,069 respondents invited to participate, n = 257 participants completed the vignette experiment, which is a response rate of 24.04%. This final dataset contained the responses of 112 women and 119 men, 71% of whom completed either a vocational or university education. The mean age of the respondents was 40 years. The number of respondents assigned per vignette ranged from n = 60 to n = 68 (Table 1). We conducted five balanced tests to assess whether randomization produced statistically equivalent groups of respondents for each vignette. Based on respondents' age, gender, education, actor type and involvement in cocreation, we found no such differences in the demographics between the four groups of respondents (Appendix Table A1).

Results

Do external efficacy and trust in public servants affect respondents' preference for coping strategies in conflicts between the dimensions of public value in coproduction? Table 2 displays the mean ranks per coping strategy for each of the four vignettes. The ranks per coping strategy range from 1 (most preferred coping strategy) to 4 (least preferred coping strategy). The mean ranks display the mean of the preferential ranks given to a specific coping strategy. The observed mean ranks range between 3.03 for trading-off in favor of user value creation as the least preferred coping strategy and 1.8 for asking for help from a public servant as the most preferred coping strategy. This shows that asking for help from public servants was most often selected as being the most preferred strategy.

Our results indicate that irrespective of the treatment combinations, asking for help from a public servant appears to be the most preferred coping strategy for coping with conflicts between the user value and social value dimensions of public value creation. Similarly, irrespective of the treatment combination, trading-off in favor of user creation appears to be the least preferred coping strategy for dealing with conflicts between the dimensions of public value.

The distributions of the mean ranks in the intermediate treatment combinations (either high or low external efficacy or trust in public servants) appear interchangeable. In both cases, trading off in favor of social value creation appears to be the second most preferred coping strategy, while trading off in favor of user value creation and avoidance are the least preferred coping strategies. For vignettes in which both external efficacy and trust in public servants are either high or low, the results diverge. For the treatment combination in which external efficacy and trust in public servants are high, trading off in favor of social value creation appears to be the second most preferred coping strategy, followed by avoidance and trading off in favor of user value creation. For the treatment combination in which external efficacy and trust in public servants are low, respondents indicate that avoidance is the second most preferred coping strategy, followed by trading-off in favor of social value creation and trading-off in favor of user creation.

To determine whether the differences in the preferred coping strategies are significant and how external efficacy and trust in public servants affect these preferences for coping strategies, we conducted ROL-regression analysis. The results of this analysis are presented in odds ratios in Table 3.

There is no statistically significant difference between coping through avoidance and trading-off in favor of user value creation. The respondents are estimated to be 152% more likely to opt for asking for help from a public servant than to do nothing (avoidance) and an estimated 29% more likely to opt for trading-off in favor of social value creation than to do nothing.

To present the effects of external efficacy and trust in public servants on the likelihood function for the separate coping strategies, we present their average marginal effects (AMEs) in Figure 2. The results confirm that neither external efficacy nor trust in public servants significantly affects respondents' preferences for one of the four coping strategies. Based on these results, we cannot support our hypotheses. Furthermore, we find no statistically significant evidence for an interaction effect between trust in public servants and external efficacy on respondents' preferences for coping strategies.

Discussion

Contrary to our expectations, we find no statistically significant effects of external efficacy and trust in public servants on respondents' preferences for coping strategies. Our results show that irrespective of the levels of external efficacy and trust in public servants, citizen coproducers prefer to resolve the conflict between creating user value and social value by asking for help from a public servant. An earlier article by Fung (2015, p. 517) stated that when “decisions involve important ethical or material trade-offs, citizens may be best placed to adjudicate those decisions [perhaps through coproduction]”. However, our findings indicate that citizen coproducers can have difficulties taking responsibility for making ethical trade-offs. The asking-for-help strategy offers an opportunity to balance out value conflicts through communication and discussion with the public servant. Citizens are looking for and desire assistance from public servants who facilitate the public value creation in coproduction and, as such, apply an adaptive strategy of seeking assistance (Newman, 2008). Asking for help from a public servant allows citizen coproducers to avoid making a trade-off between the dimensions of public value while maintaining their voice. In this respect, the popularity of the asking-for-help coping strategy shows that public servants are seen as “credible leaders” (Fledderus, 2014, p. 561). Future qualitative research could investigate citizen coproducers' preference for this coping strategy in more detail.

This study should be considered in light of several limitations. First, although we conducted our experiment using an original dataset of citizen coproducers and the temporary use of vacant spaces that include multiple coproduction services, our findings are not necessarily valid for cases outside of our study. Furthermore, in Flanders, municipalities are responsible for inviting direct participation in these areas as they see fit, whereas in other contexts, the rules, regulations and opportunities for coproduction in temporary vacant spaces are different (Refill, 2018). A replication of this study for specific coproduction services in different administrative contexts would help to further strengthen the external validity of our results. Second, there is a potential bias in the data because of a high concentration of respondents that have a higher education diploma as compared to the general population in Flanders. This can be explained by the fact that co-production activities generally attract higher educated citizens and that the participants in this study are recruited via a self-selection strategy. Preselection can happen unintentionally because certain citizens know their way around the service delivery processes better than others (de Graaf, 2007). Similarly, the results of most quantitative studies on coproduction are only representative for the respondents in the study, not for all citizens involved in temporary co-production projects everywhere and not for citizen co-producers in general. Another possible explanation for this is a selection bias in the respondents, associated with problems in attaining lists of participants. We relied on the cooperation of municipal co-production workers to bring us into contact with citizen-coproducers. In conducting the sampling, considerable effort was made to avoid selection bias. However, due to the new General Data Protection Regulation in the European Union there was no possibility to check for this. Third, we tested the effects of two factors on citizen coproducers' preferences for coping strategies when dealing with conflicts between creating user value and social value. Although the random assignment of participants to treatment conditions mediated the effects of observed and unobserved covariates, other relevant variables, such as self-efficacy, self-identity and social identity (e.g. De Cremer and Van Vugt, 1999), could also play roles in citizen coproducers' preferences for coping strategies and could be examined by future research.

Conclusion

The evidence presented here offers some valuable insights into the coping strategies of citizen coproducers. First, citizens preferring to ask for help indicate the need for facilitation and guidance from public servants and their organizations to help citizen coproducers balance out these otherwise paralyzing value conflicts. After citizens exhausted the help of public servants in dealing with the conflict, citizen coproducers preferred creating social value over avoidance, doing nothing or waiting. They are thus not only coproducers but also active cocreators of public value, not mere bystanders.

Second, the lack of statistically significant effects of trust and external efficacy should not go unnoticed. It shows that, in the drivers of coproduction behavior, there is no consistent relationship between citizen trust in government and coproduction behavior such as coping (Parrado et al., 2013; Van Ryzin et al., 2017). Similarly, external efficacy may not always be a driver of behavior in conflict situations (cf. Van Eijk and Steen, 2016), and it is valuable for scholars that the literature takes note of nonsignificant effects. Future research could replicate this study with larger sample sizes and further examine the effects of the core independent variables used in this study in other policy sectors. Additionally, our study also shows the need to think about other factors that can impact coping strategies of citizen co-producers, for example other personal traits, such as extroversion versus introversion (e.g. Luo and Shiang Chen, 1996), or interpersonal characteristics, such as types and intensity of social interaction (e.g. Emerson et al., 2011).

Finally, this study provides some managerial implications. Public officials and project coordinators of coproduction projects must be aware that conflicts between the dimensions of public value cocreation are also experienced by citizen coproducers and that they have a role in mediating these conflicts, which also occur at an intrapersonal level. Coping with these conflicts is how citizens and public servants cocreate public value in the coproduction of public services together. For this reason, it is important for public servants to remain available and approachable when citizen coproducers require feedback and guidance throughout the coproduction process. In doing so, practitioners should be open minded and innovative about encouraging citizens in their coping.

Figures

Figure 1

The vignette and the manipulated influential factors [1]

Figure 2

Average marginal effects

Table 1

Vignette population

Vignette	External efficacy	Trust in the public servant	N
1	High	High	60
2	Low	High	68
3	High	Low	63
4	Low	Low	66
		Total	257

Table 2

Mean ranks per coping strategy, grouped per vignette

Vignette	Coping strategy	Mean rank
Vignette 1	Asking for help from the public official	1.800
	Trade-off in favor of social value	2.450
	Trade-off in favor of user value	3.033
	Avoidance	2.717
Vignette 2	Asking for help from the public official	2.809
	Trade-off in favor of social value	2.559
	Trade-off in favor of user value	1.824
	Avoidance	2.809
Vignette 3	Asking for help from the public official	1.825
	Trade-off in favor of social value	2.492
	Trade-off in favor of user value	2.814
	Avoidance	2.841
Vignette 4	Asking for help from the public official	1.894
	Trade-off in favor of social value	2.712
	Trade-off in favor of user value	2.833
	Avoidance	2.561

Table 3

ROLs-regression analysis parameter estimates (with standard errors)

Parameters	Estimates
	Odds ratios	Std. err
Asking for help from public servant - (intercept)	2.516	(0.300)***
Trading-off in favor of social value creation - (intercept)	1.285	(0.144)*
Trading-off in favor of user value creation - (intercept)	0.873	(0.097)
Avoidance – (intercept)	–
Asking for help from public servant–trust†	1.165	(0.139)
Trading-off in favor of social value creation–trust	1.174	(0.131)
Trading-off in favor of user value creation–trust	1.092	(0.122)
Avoidance–trust	–
Asking for help from public servant–efficacy †	1.067	(0.127)
Trading-off in favor of social value creation–efficacy	1.115	(0.125)
Trading-off in favor of user value creation–efficacy	0.967	(0.108)
Avoidance–efficacy	–
Asking for help from public servant * trust: efficacy	0.919	(0.110)
Trading-off in favor of social value creation – trust * efficacy	0.927	(0.104)
Trading-off in favor of user value creation – trust * efficacy	0.856	(0.095)
Avoidance – trust * efficacy	–

Note(s): Significance levels: ***p < 0.001, **p = 0.01 and *p = 0.05

Source(s): †Trust in this table refers to trust in public servants and efficacy in this table refers to external efficacy

Table A1

Balance tests†

Parameter	χ2	df	p-value
Gender	0.409	3	0.938
Age	2.069	3	0.558
Activity assessment	0.0433	3	0.933
Education	5.289	3	0.152
Actor_new	1.758	3	0.624

Note(s): †We test whether the dependent parameter has an identical data distribution across the four vignettes. The null hypothesis states that the parameter is identically distributed across the vignettes. We test this hypothesis using the kruskal test function in R for five demographic parameters (gender, age, activity, education and actor). The results produce p-values that are well above 0.05, indicating that we cannot reject the null hypothesis for any of the five parameters. Our conclusion is that there is no statistical evidence to support the claim that the vignettes are not balanced, i.e. randomization succeeded in producing comparable groups

Notes

1. Translated from Dutch, which was the original language of the vignette.

2. The first author interviewed four public administration scholars, four citizens who were involved in volunteer work and one temporary use project coordinator (the project coordinator was asked not to fill out the final survey). They verbally answered the questions and explained their answers, ensuring that the vignette was understood and that the manipulations came through.

3. Approximately 55 respondents answered the test survey. The test survey was sent to people who were currently or in the past involved in volunteer work and were not part of the population sample. They coproduced services and could easily place themselves in the hypothetical situation of the vignette.

Funding: This work was funded by the Research Foundation – Flanders under (No: 11C2418N).