A Configurational Approach to Attracting Participation in Crowdsourcing Social Innovation: The Case of Openideo

Abstract

Crowdsourcing social innovation refers to utilization of crowdsourcing to solve social issues. It faces two organizational communication challenges to attract contributions: the public’s short attention span and concerns about a project’s feasibility. Guided by configurational thinking, we combine agenda setting theory and signaling theory to explore how combinations of four factors—media coverage, project duration, number of partners, and cross-sectoral partnership—can complement or substitute for one another to explain high and low crowd contributions solicited. With 53 cases from Openideo.com, we employ a fuzzy-set qualitative comparative analysis to identify two pathways to high contributions and two pathways resulting in low contributions. Implications on how organizations may design their crowdsourcing projects to attract more contributions are provided.

Keywords

crowdsourcing social innovation configurational thinking qualitative comparative analysis

Innovation and new idea development is a long-standing challenge for organizations. In recent years, crowdsourcing has been introduced as a problem-solving model for organizations, warranting further investigation (Wang, 2022). Crowdsourcing refers to the act of taking an innovation task faced by an organization to an online platform and calling for participation from people who are not members of the organization (Howe, 2006). Crowdsourcing has been adopted not only by the private sector to generate new product ideas (Brabham, 2012), but also by the public sector to encourage citizen participation in policymaking (Liu, 2017). This study pivots on the implementation of crowdsourcing for solving social issues: crowdsourcing social innovation.

Social innovation refers to “innovative activities and services that are motivated by the goal of meeting a social need and that are predominantly diffused through organizations whose primary purposes are social” (Mulgan, 2006, p. 146). Following Kohler and Chesbrough (2019) and Mergel and Desouza (2013), we conceptualize crowdsourcing social innovation as a collaborative model that leverages the collective intelligence of online communities to tackle complex social issues. The goal is to generate innovative ideas from individuals of diverse expertise and contribute to a social need (Mulgan, 2006). Organizations from any sector (public, private, or nonprofit) can host a project (Liu, 2017; Mergel, 2018). Well-known crowdsourcing social innovation platforms include the United Nations’s global pulse program on how to use big data for humanitarian actions, MIT’s Climate CoLab to solicit proposals on tackling climate change, and Openideo.com focusing on complex social problems such as water sanitation, urban recycling, higher education challenges, and widening inequity.

Crowdsourcing social innovation can be conceived of as an organizational communication phenomenon in which sponsoring organizations communicate with potential contributors to encourage participation in finding solutions (Guth & Brabham, 2017; Stohl, 2014). Organizational communication scholars have called for more research to investigate the relationship between crowds and collective problem-solving models in the contemporary media environment (Stohl, 2014). However, existing literature about crowdsourcing has been heavily guided by tech-optimism which emphasizes “the enormous benefits digital technologies could have, but tends to ignore the profound uncertainties and risks that come with technological innovation” (Lember et al., 2019, p. 1666). In other words, tech-optimism assumes that participation in crowdsourcing will naturally occur and there is no need to communicate with members of a crowd to facilitate participation. In a systematic review, Mu and Wang (2022) found that crowdsourcing social innovation often faces capacity- and technology-related barriers to promote more participation, suggesting that digital technologies do not necessarily encourage citizen contribution.

Communication from sponsoring organizations to the potential crowd is critical to the success of crowdsourcing social innovation. Crowdsourcing social innovation faces two major challenges to solicit contributions. First, with the influx of complex social issues that await solutions, the general public has a short attention span toward a particular issue (Cai et al., 2016). Second, crowdsourcing participants view their contribution as a private voluntary act and may have concerns regarding the feasibility and commitment of implementing the ideas they contributed. Crowd contribution to solve social issues constitutes their public participation; the more contributions, the more civically engaged they are (Yuan & Gasco-Hernandez, 2021). Examining what to communicate to members of a crowd to influence their participation will provide insights for hosting organizations that utilize crowdsourcing to solve social issues.

This study is guided by two theories: agenda setting theory and signaling theory. Agenda setting theory helps to examine communication variables related to issue salience, addressing the first challenge (i.e., how to stand out from other competing issues). Signaling theory helps to uncover how to reduce information asymmetry between crowdsourcing sponsors and participants. In particular, we focus on partnership signaling in crowdsourcing social innovation.

In existing literature, theories regarding crowdsourcing contributions are often examined independently. For example, Roberts et al., (2002) applied agenda setting theory to examine how issue salience and media coverage influence people’s discussion about a particular social issue online. Courtney et al. (2017) applied signaling theory to investigate how signals (e.g., project funder’s prior success, crowd comment sentiment) influence crowdfunding. The literature has thus examined competing theories rather than their integration. What is missing is a holistic approach to unpacking various pathways to crowdsourcing contributions. Informed by a configurational approach, this study integrates two theories to address the research question: How does issue salience combine with partnership signaling to attract contributions to crowdsourcing social innovation? A configuration approach helps unpack either a complementary (i.e., multiple factors work together in a mutually enhancing way) or a substitutive (i.e., two factors may replace each other in working with other factors) mechanism underlying different combinations.

Crowdsourcing Social Innovation

Crowdsourcing social innovation leverages the diversity of expertise and experiences among members of a crowd. Participants contribute by proposing ideas on how to solve a social issue. In a crowdsourcing process, participants engage in a series of interactions in which different assumptions and perspectives are discussed to resolve critical tradeoffs that were previously unresolvable. The initial pool of ideas generated thus affects the subsequent innovation process as it provides raw material for cross-fertilization of ideas across intellectual and organizational boundaries.

Implementing crowdsourcing social innovation projects typically involves four phases (Lakhani et al., 2013). A project, exemplified by Openideo.com, usually starts with the research phase during which crowd members post anecdotes, photos, videos, or existing solutions to develop empathy and understand people’s needs. The second stage is the ideation phase during which members post specific solution concepts based on what they have discussed at the research phrase. Then, the refinement phase involves testing prototypes with real people, incorporating their feedback, and then testing again. The final stage is the evaluation phase where top ideas are selected. Consider a recycling challenge sponsored by Coca-Cola as an example. During the research phase, Coca-Cola asked participants to collect stories about successful recycling in local communities, aiming to spark conversation about main barriers. Then during the ideation phase, individual members in a crowd either worked on their own or formed teams to propose actionable solutions on how to encourage people to recycle. In the refinement phase, all the contributors engaged in conversation and revised their proposed solutions. And, in the last phase, all the finalized proposals were evaluated by Coca-Cola and Openideo.

This study focuses on the ideation phase where ideas are being formalized into actionable proposals that can be tested and implemented in the real world—that is, the first phase where the effort of potential supporters is organized to contribute to real actions. Ideas in crowdsourcing social innovation constitute public participation from a crowd and reflect their contributions to produce public goods and empower civic engagement (Liu, 2017). Following Yuan and Gasco-Hernandez (2021), we argue that proposals yielded during the ideation stage can generate public value and benefit to the greater society.

Attracting a critical mass of contributors is essential for crowdsourcing projects to leverage the crowd’s creativity and knowledge. At the ideation phase, sufficient contributions to crowdsourcing social innovation are key to its quality and subsequent impacts, building a base for next stages of problem solving through revising and refining project proposals (Hartley et al., 2013; Mergel & Desouza, 2013). Yet the level of crowd contribution varies across projects and most projects have difficulty in sustaining engagement (Kohler & Rutzler, 2018). However, a paucity of empirical research has investigated what factors help crowdsourcing social innovation projects attract more and less contribution (Hartley et al., 2013). Examining what factors jointly influence crowdsourcing contributions would fill this gap and provide suggestions on how sponsoring organizations can effectively communicate with potential contributors to facilitate their participation and create public value.

Crowdsourcing social innovation differs from other types of crowdsourcing in three aspects. First, issue salience functions as an important motivator. Participation in social innovation is motivated by a social need instead of profit maximization (Mulgan, 2006). When an issue is framed as more salient, it can generate empathy and function as a personal motivator (Mergel & Desouza, 2013). Second, sponsorships organized by cross-sectoral partnerships create complexity (Kohler & Rutzler, 2018). Examples of such partnerships include collaboration and coordination among policymakers, sponsoring organizations, and local communities. These partnerships are necessary, because social innovation does not just rely on monetary resources but also requires access to constrained resources such as political recognition or philanthropic commitment (Cooper, 2021). Third, the success of crowdsourcing social innovation is difficult to measure as these social innovation challenges do not always bring tangible benefits. Therefore, organizations behind crowdsourcing projects need to show their ability to deliver ideas or products.

A gap exists in the literature regarding what factors hinder or facilitate crowdsourcing social innovation (Mergel, 2018). Kohler and Rutzler (2018) conducted a longitudinal experiment on a crowdsourcing platform and found that many social innovation challenges failed to reach a critical mass, hence calling for implementing organizations to “provide the appropriate structure and incentives to architecture participation” (p. 65). Their study focused on type of innovation, motivation, and platform business model as designing principles. The current research looks into the same question but from a different perspective by considering challenge (project)-level characteristics.

A crowdsourcing community often faces participation inequality due to different motivations and incentives. We argue that a user’s contribution to crowdsourcing social innovation is subject to two major communication challenges. First, there is an influx of social issues competing for the public’s attention while at the same time the general public has a short attention span (Cai et al., 2016). The challenge is how to make a particular crowdsourcing social innovation project stand out. This is consistent with the concept of philanthropic insufficiency, which encapsulates the inability of voluntary contributors “to generate resources on a scale that is both adequate enough and reliable enough to cope with the human-service problems of an advanced industrial society” (Salamon, 1987, p. 39). The second challenge is that potential contributors may have concerns about the feasibility and the commitment of implementing their ideas due to the information asymmetry. It is thus important to uncover what communication factors could attract individuals to contribute.

We draw on agenda setting theory and signaling theory to identify multiple project-based conditions that may affect the crowd’s participation in the social problem-solving process. Agenda setting theory is utilized to address the challenge of issue salience, whereas signaling theory is used to uncover how to reduce information asymmetry between crowdsourcing sponsors and participants.

Social Issue Salience and Crowdsourcing Contribution

Public attention is defined as the coverage social issues receive within public arenas or organized channels of communication (Hilgartner & Bosk, 1988). Social issues vary in the level of attention they receive by institutional actors and their ability to shape the public’s awareness and understanding. Public attention influences individuals’ perceptions of a particular issue through the sensemaking process, i.e., people giving meanings to what they do, which determines people’s behaviors (Weick et al., 2005). According to agenda setting theory, mass media play an important role in informing the public of what to pay attention to.

Agenda setting theory has been tested in the contexts of presidential election (McCombs & Shaw, 1972), corporate communication (Tam, 2015), and public sector innovation (Vigoda-Gadot et al., 2005). At the theory’s core is issue salience. Salience, defined as perceived importance, posits that the frequency with which mass media cover an issue can significantly dictate how the audience views the importance of the issue to a society (Vargo et al., 2018). Agenda setting theory does not assume that the public simply accepts the news blindly; instead, it focuses on mass media setting the public salience for an issue. For example, when mainstream news media cover climate change on a daily basis, the general public is expected to pay more attention to the issue and view it as an important topic to think about. To uncover how issue salience influences crowdsourcing contribution, we take into consideration media coverage and project duration, and how these factors may interact to influence crowdsourcing contributions.

Media Coverage

We theorize that when a particular social issue a crowdsourcing project tackles gets more media coverage, it is more likely the public will view the issue as salient. The media coverage as a critical carrier of attention highlights critical issues, events, and sense-making (Nigam & Ocasio, 2010). Public attention to a social issue can trigger contributors to make sense of the issue. Extensive media coverage indicates that a social issue has reached a critical mass where a sufficient number of individuals in a society are paying attention. Higher salience could make an issue stand out more, and attract more attention from potential contributors. Therefore, the high salience attributed to more media coverage helps to address the challenge of the influx of social issues competing for the public attention. For example, event-driven crowdsourcing (such as during the time of a political election, social movement, and natural disaster) is likely to generate a large volume of contributions (Bailard & Livingston, 2014). Therefore, we would expect crowdsourcing projects addressing a social issue that has received extensive media coverage will attract more contributions from a crowd.

Project Duration

Another factor to account for when evaluating the effect of issue salience is project duration. The reason is twofold. First, project duration affects how media coverage is measured through a particular time frame. Thus, the effects of both factors should be examined simultaneously. Second, guided by configurational thinking, the combination of project duration and media coverage could result in different pathways and either high or low contributions. The optimal project duration is can be a dilemma for crowdsourcing social innovation due to the short attention span from potential participants (Cai et al., 2016). Most crowdsourcing projects have an intermediate project duration, persisting only as long as current funding permits. For a crowdsourcing project to reach a critical mass, it has to last long enough to grab the public’s attention. Particularly in social innovation projects, it may take a while for the call for contribution to be communicated to diverse social groups and to solicit ideas. On one hand, long project duration allows participants to engage in conversation and generate iterations of innovative proposals. On the other hand, if a project lasts too long, it increases the cost for project management and may lose attention from the public.

Reducing Information Asymmetry by Signaling in Crowdsourcing Social Innovation

Another barrier to generating contributions is information asymmetry between members of a crowd and the hosting organization—a situation in which one party has more or better information than the other. Information asymmetry could occur either through a lack of information or a lack of credible information (Courtney et al., 2017). A typical crowdsourcing project describes the goal, duration, and partners who provide funding and access to local communities to test ideas (Lakhani et al., 2013). Potential participants could comment on the homepage of a project and interact with fellow participants. Participants may face the risk of investing intellectually in something that can never be completed due to limited funding, and hosting organizations may face the challenge of how to provide credible information about the cause to be addressed or the process of crowdsourcing (Chen & Aitamurto, 2019). Participants may have concerns about the feasibility of the project. Therefore, the uncertainty embedded in information asymmetry can discourage people from contributing their ideas.

Signaling theory examines what signals can be sent by the informed party (i.e., hosting organizations) to the less informed party (i.e., members of a crowd) to reduce information asymmetry (Moleskis et al., 2019). Effective signals need to be observable and costly (Fischer & Reuber, 2007). Observability of a signal refers to the extent to which the signal is noticeable by outsiders. For example, partners’ names shared on the challenge Web site are observable signals. Costly signals indicate quality, reliability, or genuineness, and thus are hard to fake (Connelly et al., 2011; Moss et al., 2015). For example, a costly signal would be an endorsement from a celebrity for the recycling challenge. To obtain this endorsement, the sponsoring organization needs to pay the celebrity as a spokesperson for the project; once the signal is established, it is more likely to be perceived as reliable and genuine.

Effective signals separate a project from competitors and are essential for activating the credibility of the project (Hasson, 1997). In crowdsourcing social innovation, what motivates people to contribute is the legitimacy of the projects, requiring hosting organizations to communicate sufficient information to the crowd. Guided by signaling theory, we argue that effective signals could help endorse the credibility of projects to reduce the uncertainty potential contributors may have.

Literature on crowdsourcing has applied signaling theory in two research areas: (a) how certain characteristics of the crowd (e.g., expertise, experience) function as effective signals to attract attention and increase bargaining power (Durward et al., 2016), and (b) how features of a project (e.g., feedback provided or not, monetary or non-monetary incentives) can affect the decision making of a crowd (Glaeser et al., 2016). However, no research has investigated what signals embedded in project characteristics could influence the amount of contribution from a crowd (see exceptions in crowdfunding, e.g., Courtney et al., 2017; Horvát et al., 2018; Moleskis et al., 2019).

We focus on effective signals related to partnerships (Courtney et al., 2017). Partner organizations may play different roles throughout a project. They often provide financial support, develop criteria in evaluating ideas, and test proposals in communities (Lakhani et al., 2013). Two partnership-related factors can influence the number of contributions: the number of partners and cross-sectoral partnerships.

Number of Partners

Interorganizational partnerships are created to mobilize resources to collectively achieve a goal. Partnerships embedded in a crowdsourcing project serve as a signal of third-party endorsement for a project to succeed (Courtney et al., 2017; Mergel, 2018). Having more partners shows a strong signal about how the sponsoring organization can secure resources from partnering organizations to mitigate the problems of information asymmetry (Courtney et al., 2017). In other words, having a larger number of partners helps to provide credible information about the project. Therefore, potential contributors will have more confidence in the feasibility of the project.

Cross-Sectoral Partnerships

Organizations from different sectors have sponsored crowdsourcing social innovation. In the public sector, well-known examples include the New Zealand government’s flag referendum, the Crowdsourcing Act of the Philippine, and Iceland’s Kitchenware revolution to develop a new constitution. In the private sector, crowdsourcing has been implemented by companies such as Starbucks, Dell, and IBM.

The question remains as to what types of partnerships affiliated with crowdsourcing social innovation are more effective signals. Cross-sectoral partnerships are built upon the notion that complex social issues transcend the capacities of individual organizations or sectors to deal with in silo (Austin & Seitanidi, 2012). Cross-sectoral partnerships are expected to create significant economic, social, and environmental value to benefit a greater society. Following this line of research (Bryson et al., 2015), we propose that partnership characteristics could function as effective signals to endorse the credibility of a crowdsourcing project and to verify the resources available through project partners, influencing the number of contributions.

We also acknowledge that cross-sectoral collaboration does not occur without challenges. Stephens et al. (2009) demonstrated that trust could be an issue if partners had no past relationships. Different sectors bring their own agenda into the collaboration, which may lead to conflicting goals and higher coordinating cost (Clarke & MacDonald, 2019). Furthermore, different sectors are unique in organizational structures, which may negatively influence information sharing and collaboration processes. Therefore, the effect of cross-sectoral partnerships is contingent on how such partnerships are combined with other factors.

A Configurational Approach to Crowdsourcing Contribution

A conventional “reductionist” approach views a social phenomenon as decomposable into independent elements that can be examined individually (Van de Ven et al., 2013). However, it is important to consider our research topic in a holistic way to examine what configurations of the factors—media coverage, project duration, number of partners, and cross-sectoral partnerships—may attract high or low contributions. A configuration is a specific combination of causal variables that explains an outcome. For example, third-party endorsement does not function as an effective signal in isolation and often operates with other signals (Courtney et al., 2017).

A qualitative comparative analysis (QCA) is a configurational approach appropriate for our study. Three aspects of QCA are noteworthy (Schneider & Wagemann, 2012). The first of these is conjunctural causation. In regression-oriented approaches, each independent variable is held constant at its average to isolate the independent effect of that variable. Whereas these approaches conceal the ways factors may interact with each other to affect the ultimate outcome, QCA assumes a multidimensional combination of contingencies that commonly occur together and focuses on how a system is designed from the interaction of its constituent elements taken together as a whole. While multiplicative interaction terms can be used to test joint effects in regressions, they typically do not include more than two independent variables. Also, adding multiple interaction terms requires a large data set to alleviate the issues of multicollinearity and stretching degree of freedom. Qualitative comparative analysis can analyze a large number of interactions/combinations of more than two conditions for small-to-medium-size cases.

A configuration refers to a constellation of factors bound together by some internal mechanisms and thus needs to specify how different factors may interact to bring about the outcome. The present study pays attention to two combinational mechanisms: complementary and substitutive. A complementary mechanism indicates that multiple factors work together in a mutually enhancing way (McEvily et al., 2014; Misangyi & Acharya, 2014). For example, when a project has a greater number of partners to provide financial and technical support and it lasts long enough to gather sufficient public attention, this project will likely solicit more contributions. The complementary mechanism essentially suggests that all positive or negative relationships among variables and their combinations lead to a synergy of positive or negative effects on the outcome. In contrast, a substitutive mechanism posits that two factors may replace each other in working with other factors (Misangyi & Acharya, 2014). For example, consider the two signals related to partnerships: the number of partners and the existence of cross-sectoral partnerships. These factors can both indicate the availability of sufficient resources to manage the crowdsourcing process, sending positive signals to reduce information. It is also possible that they could both negatively affect contributions due to expected high coordination cost. Therefore, their effects could substitute each other when coupling with other factors.

A second key aspect of QCA is equifinality. Regression analyses are unifinal in assuming linear and additive effects of explanatory variables on the outcome. A number of statistically significant independent variables are assumed to constitute a one-size-fits-all optimal solution to the outcome. Qualitative comparative analysis posits that organizational problems can be resolved by multiple feasible, equally effective options. Different conjunctural combinations of various conditions represent multiple pathways toward the same outcome. For example, there might be two pathways to attracting contributions to crowdsourcing social innovation: either a combination of extensive media coverage and a large number of partners or a combination of longer project duration and cross-sectoral partnership.

A third key aspect of QCA is asymmetric causation. That is, the articulation of a configurational rationale includes an epistemological assumption of non-symmetric causation. The dominant correlation-based approaches have channeled research efforts toward symmetric causality in which the presence or a high value of A leads to the presence or a high value of B. Following this, conventional symmetric thinking posits that the absence or a low value of A will result in the absence or a low value of B. Yet in complex reality, combinations of factors leading to presence of an outcome may differ from those leading to lack of the outcome. Using QCA, we explore combinations of conditions leading to high contribution, as well as other combinations of conditions resulting in low contribution.

The QCA method is subject to two shortcomings. First, the number of explanatory conditions is limited because the number of possible configurations increases exponentially with the number of conditions considered. This is why researchers should focus on theoretically relevant conditions. Second, the method is not well equipped to analyze case dynamics over time. This study applies two theories to guiding the QCA analysis with cross-sectional data collected from Openideo.com.

Data Collection and A Fuzzy-Set Qualitative Comparative Analysis Method

We employed QCA to conduct a configurational analysis of the factors that lead to high and low contribution to crowdsourcing social innovation. Using a set theory and Boolean algebra, QCA as a case-oriented approach unpacks the complex interactions among different factors with a limited number of empirical cases (Ragin, 2008).

Constructing Empirical Cases: Crowdsourcing Social Innovation at Openideo.com

A flow chart is shown in Figure 1 to indicate the detailed steps of QCA.¹ The first step in any QCA studies is to use the outcome of interest to identify the “theoretically defined” population of cases (Ragin, 2008, p. 4). Our comparative case data were collected from a global crowdsourcing community, Openideo.com, founded in August 2010 to host social innovation projects. Participation on Openideo is voluntary with no reward expected. We chose Openideo.com for two reasons. First, it is one of the largest crowdsourcing platforms for social innovation. Since its founding, Openideo has attracted people in over 200 countries and territories, generating nearly 19,000 ideas for good. This allowed us to gather and analyze a large set of crowdsourcing social innovation cases. Second, we adopted a comparative logic close to a “most similar system design” to select a set of crowdsourcing cases (Przeworski & Teune, 1970) by comparing them as similar as possible, those that differ only in those “independent” variables of our theoretical interest (“causal conditions,” in QCA terms), which should explain the variation of the “dependent” variable (“outcome”). We were primarily interested in examining the interplay of issue salience and partnership signals. By focusing on one platform, we effectively controlled some platform design features that may influence the crowd contribution; for example, charging hosting fees and enabling user voting (Wen & Lin, 2016, innovation type, motivation, and platform business models (Kohler & Rutzler, 2018).

Figure 1.

Steps of performing qualitative comparative analysis.

Our dataset included a total of 53 social challenge projects from August 2010 to August 2017. These cases covered a variety of social issues including humanitarian crisis, business innovation, climate change, education, public health, and community well-being. Openideo projects are created through collaboration between Openideo and sponsors which can be a nonprofit organization, a government agency, or a for-profit business (see Lakhani et al. (2013) and Wang (2020) for more detail on how Openideo operates).

Calibration

The second step of QCA is “calibration,” the process of transforming raw data into cases’ membership scores on the outcome and conditions (Ragin, 2008). This study employed a fuzzy-set approach that converts all interval conditions into a continuous scale ranging from 0.0 to 1.0 (Ragin, 2008). We relied on the direct method of calibration, a process of fuzzy-set calibration determining which raw condition values constitute full membership (“in,” the 95th percentile) in a respective category (e.g., high contribution), full non-membership (“out,” the fifth percentile, e.g., low contribution), and the crossover point (neither “in” nor “out,” the 50th percentile, e.g., neither high nor low contribution) (Ragin, 2008). Then, the Fuzzy-Set Qualitative Comparative Analysis (fsQCA) 3.0 software transformed raw data into fuzzy membership scores using the three benchmarks based on the log odds of full membership.

Number of Contributions

Number of contributions was measured as the total number of proposals submitted to each case at the ideation stage, ranging from 7 and 660 (M = 249.70, SD = 177.16). Based on the distribution of raw data, we computed the high contribution benchmark as 611 submitted proposals at the 95^th percentile, the crossover point as 175.9 submissions (the 50^th percentile), and the low contribution level as 53.6 proposals (the fifth percentile). For example, any case with the number of submitted proposals equal or more than 611, would be assigned a full membership score of 1, with 175.9 submissions a membership score of 0.5, and with 53.6 or less submissions a full non-membership score of 0. If a project received either 240 (between the high contribution benchmark and the crossover point) or 151 proposals (between the crossover point and the low contribution benchmark), it would be computed as a membership score of either 0.61 or 0.35.

Media Coverage

Following Lim et al. (2016), we first identified keywords to describe what social issue a project focused on and then searched in the LexisNexis dataset to capture the number of news articles containing the keywords. To accurately record media coverage, we limited our search to only the period when the project was hosted to accept proposals. This measure ranged from 16 to 93,092 (M = 18,894.28, SD = 25,559.07). In the same vein, we calibrated the full membership at 8,0730.1 news articles, the crossover point at 4322.9, and non-membership at 66.3.

Project Duration

This condition indicates how many months a project allowed submission of proposals. A typical project on Openideo lasted between three and 5 months. We recorded the beginning and ending months of each case. The duration ranged from 1 month to 6 months (M = 2.83, SD = 0.91). We set the threshold for long duration (i.e., fully in) to 4.3 months, 2.9 months as neither in or out, and 1.7 month as fully out.

Number of Partners

This condition counts how many unique partners sponsor each project acknowledged in its main page. It ranged from 1 to 13 (M = 3, SD = 3.02). The thresholds of full membership were set to 5.3 sponsors, the crossover point to 1.9 sponsors, and non-membership to 1 sponsor.

Cross-Sectoral Partnerships

This condition was calibrated as binary where organizations from multiple sectors, including for-profit, were acknowledged as sponsors (n = 31), and 0 = single sector (n = 22).

Analyses

We used fsQCA 3.0 to create a “truth table,” a data matrix summarizing $2^{k}$ logically possible combinations, where k is the number of causal conditions under consideration. In our study, the truth table includes $2^{4} = 16$ combinations. In QCA, the truth table is first consolidated based on (a) a threshold of at least two cases for a given configuration to be considered and (b) the minimum level of consistency of 0.80 for a given solution (Fiss, 2011; Ragin, 2008). Consistency refers to the degree to which cases that share a combination of conditions consistently produce the key outcome. Low consistency indicates that a given configuration is not reliably related to the outcome; high consistency means that a given configuration of factors almost always leads to the outcome of interest (Misangyi & Acharya, 2014). We also later ensured that the overall final solution consistency was above the 0.80 threshold (Crilly, 2011).

The truth table configurations were then simplified or logically reduced using the software’s Boolean algorithm. The final analysis generated three solutions: complex, parsimonious, and intermediate. In an intermediate solution, the algorithm removes causal conditions from the complex solution that are inconsistent with existing knowledge (Ragin, 2008). In contrast, the parsimonious solution is based on all simplifying assumptions, representing the most reduced form of the complex solution. A combination of the parsimonious and intermediate solutions is recommended for interpreting the QCA results. We followed extant research (Fiss, 2011; Misangyi & Acharya, 2014) to determine which conditions were core or peripheral to the configurations. Core conditions are part of both the intermediate and the parsimonious solution, indicating a strong causal relationship with the outcome, while peripheral conditions are absent from the most reduced, parsimonious solution and only present in the intermediate solution, suggesting a weak relationship.

Finally, asymmetric causation required that we repeat the analyses to produce the configurations of conditions that lead to the low crowdsourcing contribution.

Results

Table 1 summarizes the configurations generated by our analyses. We follow the notation applied by Fiss (2011), where “●” represents the presence of a condition, “⊗” represents its absence, and a blank space indicates a “don’t care” situation, meaning that a given condition is not causally related to the outcome. Moreover, larger symbols indicate that the condition is core to a given configuration, while smaller symbols point to peripheral conditions. The four theoretically important factors yielded an equal number of empirically supported pathways in Table 1: Pathways 1–2 reflect the constellations of factors attracting high contribution, while Pathways 3–4 reflect the constellations of factors attracting low contribution.

Table 1.

Pathways to High and Low Crowdsourcing Contribution.

	High Contribution		Low contribution
	1	2	3	4
Issue Salience
Media Coverage		●	$\otimes$	$\otimes$
Project Duration	●		$\otimes$	$\otimes$
Information Asymmetry
Number of partners	●	●	$\otimes$
Cross-Sectoral Partnership		$\otimes$		●
Consistency	0.81	0.91	0.87	0.82
Row Coverage	0.62	0.23	0.43	0.32
Unique Coverage	0.44	0.05	0.24	0.13
Overall Solution Consistency	0.81		0.83
Overall Solution Coverage	0.67		0.57

Note. ●:Core condition; ●:Peripheral Condition; $\otimes$ : Absence of Core Condition; $\otimes$ : Absence of peripheral Condition; Blank space indicates it is not relevant to the outcome.

We include measures of consistency and coverage for each pathway and for the solution as a whole. Coverage assesses the degree to which instances of the outcome of interest are accounted for by a given path, and by the solution as a whole. All of our individual pathways exhibit acceptable consistency levels (perfect consistency being 1), but they also have varying degrees of coverage. Somewhat analogous to the way variance explained is partitioned in multiple regression, coverage can be further divided into “raw” and “unique” portions. Unique coverage which explains memberships in the outcome not covered by other pathways points to the relative empirical “weight” of each path (Ragin, 2008).

Across the four pathways, the factors related to increasing social issue salience combined with ones reducing information asymmetry. Pathway 1 includes longer project duration and larger number of partners. Pathway 2 exhibits some differences, as it includes extensive media coverage of the social issue and the absence of cross-sectoral sponsorship, in addition to a larger number of partners. In both configurations, a larger number of partners is a peripheral condition. Pathway 3 and 4 are empirically similar. Pathway 3 points to three core conditions: less extensive media coverage, shorter project duration, and smaller number of partners. Pathway 4 also includes less media coverage and absence of longer project duration (in other words, shorter project duration), but a presence of cross-sectoral partnerships, in lieu of the lack of a larger number of sponsors.

Case Narratives

In accordance with our configurational focus, we labeled the pathways as “long-lived favorite,” “well-received,” “poorly-received,” and “short-lived and wrongly-perceived” with typical case narratives from Openideo. See Table 2 for a summary.

Table 2.

A Summary of Pathways and Cases.

	High Crowdsourcing Contribution		Low Crowdsourcing Contribution
	Pathway 1: Long-lived favorite	Pathway 2: Well-received	Pathway 3: Poorly-received	Pathway 4: Short-lived and wrongly perceived
Combination of conditions	Longer project duration and larger number of partners	Extensive media coverage of the issues, larger number of partners, and the absence of cross-sector partnerships	Fewer media coverage, shorter project duration, and smaller number of partners	Fewer media coverage, shorter project duration, and existence of cross-sectoral partnerships
Configurational mechanism	Complementary	Complementary	Complementary & substitutive	Complementary & substitutive
Typical cases	The end of life	The urban resilience and climate change	The openideo social impact	The social business
	Recycling habits at home	Plastic waste	Social challenge stories	Inspirational logo for openideo
	How do low-income communities ensure children thrive in their first 5 years?	Plastic waste		Inspirational logo for openideo
	Agriculture tech and climate change	Reducing stigma and increasing opportunities for people with disabilities		Combat health threats like zika, SARS, ebola and malaria

Pathway 1: Long-Lived Favorite

The first pathway to more contribution contained a combination of longer project duration and larger number of partners. This configuration showcases a long-lived favorite. The two conditions identified suggest a complementary mechanism: cases that are long-lived in conjunction with being well-supported by more partners are more likely to attract more contributions. The longer duration addresses the challenge of issue salience by retaining the potential contributors’ attention, and by making the social issue tackled stand out. Larger number of partners addresses the information asymmetry challenge by signaling project endorsements, and by demonstrating sufficient resources. Potential contributors will see that they are not alone in making a contribution.

One of typical cases in this pathway is the End of Life, which focused on reimagining the end-of-life experience. The ideation stage was relatively long between May 2016 and July 2016, above the average project duration of all the sampled challenges (M = 2.83). This case had four partners including two health care providers, one foundation, and Openideo. The partners helped to recruit participants with different experiences and skillsets, and developed evaluation criteria (Openideo, 2019). The combination of longer project duration (3 months) to keep participants engaged in conversation and reach a critical mass of ideation, and sufficient number of partners to provide resources leads to a higher number of contributions (n = 300).

Pathway 2: Well-Received

The second pathway to higher contribution included extensive media coverage of the issues, more partners, and the absence of cross-sector partnerships. This configuration is an example of a well-received project. Three conditions in this configuration suggest a complementary mechanism as well, i.e., extensive media coverage enhances the salience of an issue, coupled by larger number of same sector partners that endorses the credibility of the project, mixing with the absence of cross-sectoral partnerships that may have helped to ease the coordination process and cost and provide signals on the feasibility.

One typical case in the well-received pathway is Urban Resilience which solicited new ideas for urban slum communities to adapt, transform, and thrive given the effects of climate change. The topic had a lot of media coverage (n = 69,663) using key words “climate change” and “urban living.” It had three nonprofit partners, all of which had overlapping interests in helping vulnerable populations and building a more resilient future. Single-sectoral partnerships indicate that all the partner organizations had a focused niche, which helped to align the collective goal and mobilize resources toward a collective agenda.

Pathway 3: Poorly-Received

The first pathway leading to lower contribution had the following conditions: fewer media coverage, shorter project duration, and smaller number of partners. This configuration captured projects that are poorly received in the media and lack sufficient resources and time allocation to reach a critical mass of contributions. The challenges of how to make the issue stand out and how to address potential contributors’ short attention span have not been addressed jointly. In addition, the single host may further send a signal that this project was merely for self-serving purposes, instead of for achieving greater social good. Therefore, the information asymmetry and the resulting trust issue remain unsolved. In other words, potential contributors do not trust the sponsoring organization and are not willing to provide contributions. The combined conditions in this configuration fit the complementary mechanism due to the consistent overall negative effects.

The Openideo Social Impact is one of typical cases aligning with this configuration. With a focus on generating innovative ideas about how Openideo can increase its individual and collective social impact, its topic is very specific and abstract which did not attract a lot of media coverage (n = 1,271, using key words “social impact” and “Openideo”). The ideation stage lasted only 2 months. It was solely supported by Openideo. The lack of issue salience, the limited focus, and sole partner resulted in a low number of contributions (n = 91).

Pathway 4: Short-Lived and Wrongly-Perceived

The second configuration leading to lower contribution entailed a mix of following conditions: less media coverage, shorter project duration, and existence of cross-sectoral partnerships. Cross-sectoral partnerships in the study took the following forms: nonprofits working with government agencies, or nonprofits working with businesses. Given that the government sector and the private sector tend to receive less public trust (Edelman, 2015), the cross-sectoral partnerships may be “wrongly-perceived” with suspicion raised among potential contributors. The combination of short-lived projects, few media coverage, and cross-sectoral partnerships fails to address the issue salience and information asymmetry challenges, leading to fewer contributions. This pathway also highlights complementary mechanisms.

One example of this pathway is the Social Business, which focused on using social business to improve health in low-income communities. The topic it tackled (social business) was not a trendy theme in media coverage, with only 371 relevant news articles. The ideation stage lasted 2 months. This project was supported by cross-sectoral partnerships formed among Openideo, a local government in Colombia, a social enterprise organization (Grameen Creative Lab). This case suggests that for social innovation projects that are short-lived with less intensive media coverage, it is hard to reach a critical mass of participants (96 contributions were received). Potential participants were reluctant to contribute, as the issue being tackled did not reach a significant level of salience for them. The issue did not stand out in the large pool of social issues that were competing for their attention. Furthermore, cross-sectoral partnerships may help to endorse project credibility, yet they may also generate obstacles to facilitate the involvement of stakeholders and flow of information.

Pathways 3 and 4, despite their unique coverage, exhibited a fair amount of empirical overlap, highlighting shared elements across low-contribution innovation challenges. Two pathways reflecting the outcome of low contribution shared two core conditions of less extensive media coverage and shorter project duration and only differed in that pathway 3 had a smaller number of sponsoring partners as a core condition and pathway 4 had the presence of cross-sectoral partnerships also as a core condition. These two conditions (smaller number of sponsors and presence of cross-sectoral partnerships) can substitute each other when matched with two factors related to issue salience: less extensive media coverage and shorter project duration. This finding suggests that when a project is less well discussed in the media and has a pressing timeline to attract contribution, either a small number of partners or partners that are all from the same sector can make the situation worse. Based on the different project durations identified in Pathways 1, 3, and 4, we argue that there might be a sweet spot for the appropriate project duration between 3 - 4.3 months.

Discussion

Participation in crowdsourcing social innovation constitutes citizens’ public participation in solving complex social issues. More participation entails a higher level of commitment and contribution to the greater society (Liu, 2017). This study utilized QCA as a configurational approach to examine factors derived from agenda setting theory and signaling theory to identify different mechanisms underlying crowd contribution. The results showed that crowdsourcing does not necessarily result in more contributions, which stands in contrast to the tech-optimism argument.

Two findings are worth noting. First, this paper demonstrates that the effects of issue salience and partnership signaling on the crowdsourcing outcome are not linear and independent. This challenges the conventional assumption that there is an optimal way to motivate crowds to contribute their ideas to the collective knowledge pool. Social innovation challenges can be resolved by considering multiple feasible and equally effective ways. Results showed that different factors match and mix with one another through complementary and substitutive mechanisms. Consider media coverage as an example. The literature has primarily documented the positive effect of media coverage on soliciting contributions from individuals (Vargo et al., 2018). This study showed that its effect is more complex than that: in high contribution scenarios media coverage had no effect if a project was hosted for a long period of time to reach a critical mass of attention and at the same time had a large number of partners as supporters. In other words, when sponsoring organizations launch a challenge tackling a less well-known social issue, they need to host it for longer (longer project duration and extensive media coverage were revealed as complementary mechanisms).

Second, the QCA analysis in this study uncovered the complexity of explaining the number of contributions to social innovation projects by challenging the traditional view that a particular factor alone has a consistent and equal effect on an outcome. For example, we identified one condition (larger number of partners) as a peripheral condition in high contribution cases but its absence as a core condition in low contribution cases, depending on its combination with other factors. Another example is the effect of cross-sectoral partnerships. In high contribution scenarios, their absence was a core condition; while their presence was a core condition in low contribution scenarios. This finding challenges the value of cross-sectoral partnerships advocated in the literature (Austin & Seitanidi, 2012). Cross-sectoral collaboration may entail tension related to community problem-solving (Cooper, 2021; Cooper & Shumate, 2012) and its complexity may signal more uncertainty in crowdsourcing social innovation (Kohler & Rutzler 2018).

Findings from this study have important implications on how to design a crowdsourcing social innovation project to attract more contributions. First, strategies can be created by focusing on a social issue that attracts more media coverage (pathway 2) or hosting a project for an extended period of time to reach a critical mass when the issue does not receive wide media coverage (pathway 1). These two conditions do not need to co-exist to result in more contribution. Social innovation issues range in terms of coverage from legacy media organizations and project funders that are interested in less popular social issues may utilize social media to raise the awareness and to increase salience of the issue to potential contributors. The key is to create a higher level of salience. This finding resonates with findings from the literature demonstrating that framing an issue as salient can function as personal motivators (Mergel, 2018). Second, the hosting organization should be more careful in making cross-sectoral partnership decisions and provide effective signals about the feasibility of the crowdsourcing process. The results showed that cross-sectoral partnerships do not have a linear relationship with the outcome variable. As highlighted by Kohler and Rutzler (2018), complexity related to cross-sectoral partnerships is a main barrier to social innovation. Our finding shows that the communication of cross-sectoral partnerships does not necessarily motivate more contribution. Combined with less issue salience and shorter project duration, such communication could impede contribution. Overall, practitioners are encouraged to carefully address the concerns potential participants may have regarding issue salience and information asymmetry in a holistic manner. We acknowledge that organizations may not have the luxury of selecting partners. In that situation, an alternative pathway is available for generating more contributions (i.e., Path 1 where partnership type does not matter).

This paper makes two major contributions to the literature on crowdsourcing social innovation. First, it examines the interplay of two theoretical frameworks (agenda setting theory and signaling theory) to unpack combinations of conditions that influence public participation in social innovation. Guided by configurational thinking, the current study differs from the existing literature that tends to examine these theories independently (Robert et al., 2002; Courtney et al., 2017) and fails to take on a holistic approach to uncovering different pathways leading to high and low contributions. This study identified complementary and supplementary mechanisms with factors from both theories and offered implications for sponsoring organizations to address the challenges of issue salience and partnership signaling. The results demonstrated that the relationships examined among variables should not be viewed as linear and independent. Crowdsourcing projects do not have to only tackle social issues that have been heavily covered in the mass media or rely on cross-sectoral partnerships to persuade potential contributors.

Second, this study combined a fsQCA with typical case narratives with 53 cases from a global crowdsourcing platform to uncover multiple pathways to less or more contribution in crowdsourcing social innovation. The complex interactions among issue salience and partnership signaling stand in contrast to the tech-optimism argument and show that digital platforms do not necessarily lead to a critical mass of contributions. The study thus answers the recent call from organizational communication scholars (Guth & Brabham, 2017; Stohl, 2014) to further examine the importance of communication between crowdsourcing sponsoring organizations and crowd members in encouraging public participation.

Four limitations of the study warrant further research. First, the goal of QCA is to explore configurations of the independent variables from a small-to-medium size of cases, not to yield generalizable findings from large-N studies. This study relied on data collected from a single crowdsourcing platform. While we controlled the platform design features, the results should not be easily generalized. Crowdsourcing platforms vary by designing features. We call for future research to replicate our approach in other platforms. Second, we believe that the outcome examined should move beyond the simple contribution number. The initial number of idea contributions serves as an important starting point. Future research needs to examine how project signals and issue salience interact with one another to influence the quality and the impact of ideas generated. Third, this study measured issue salience with media coverage. This variable was a proxy measurement about social issues a project tackled. It is possible that large scale catastrophic events that received more media coverage may simply indicate a compassion issue leading to more engagement, which was not accounted for in this paper. Future research should consider direct measurements about specific projects; for example, publicity efforts from the hosting organizations on social media platforms. Social media campaigns could function as a more proactive communication strategy to increase issue salience and to solicit more contributions (Madianou, 2013). Finally, QCA only allows limited variables in the analysis due to the complex relationships it may unpack. We acknowledge that additional factors may play important roles to influence crowdsourcing social innovation contribution (Mergel, 2018). One useful theory to draw from in future research is diffusion of innovations theory (Rogers, 2003). Future research could examine the effects of opinion leaders, network exposure, and perceived innovativeness on crowdsourcing social innovation contribution.

Conclusion

This study offers a case study of crowdsourcing social innovation as a collaborative model that leverages collective intelligence of online communities to tackle complex social issues. Qualitative comparative analysis analysis showed two pathways leading to more contribution: (1) long-lived favorite (longer project duration and larger number of partners) and (2) well-received (extensive media coverage, larger number of partners, and same-sector partnerships). Two pathways were also identified to result in less contribution: (1) poorly-received (fewer media coverage, shorter project duration, and smaller number of partners and (2) short-lived and wrongly perceived (fewer media coverage, shorter project duration, and cross-sector partnerships). The results reported only begin to scratch the surface of theoretical exploration that examines how to motivate the crowd to contribute by accounting for configurations of issue salience and partnership signaling. This study offers theoretical and practical implications on how different factors match and mix to jointly influence the number of crowdsourcing contributions. As crowdsourcing brings both new opportunities and challenges to the field of social innovation, we call for more research to go beyond tech-optimism to further uncover the mechanisms of utilizing the collective intelligence to contribute to social innovation.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Network for Nonprofit and Social Impact, Northwestern University.

ORCID iD

Rong Wang

Note

Author Biographies

Rong Wang (PhD, University of Southern California) is an Assistant Professor at the Department of Human and Organizational Development at Peabody College of Education and Human Development, Vanderbilt University. Her main research interests include collaborative crowdsourcing, inter-organizational alliances, collective action, and nonprofit management, with a focus on how to leverage network dynamics to solve complex social issues.

Bin Chen (PhD, University of Southern California) is a professor in the Austin W. Marxe School of Public and International Affairs, Baruch College, and a doctoral faculty at The Graduate Center, The City University of New York, USA. His research spans collaborative governance in public policy implementation, government-nonprofit relations, regional networked governance, and comparative public administration and policy, with methodological interests on social network analysis, qualitative comparative analysis, and necessary condition analysis.

References

Austin

J. E.

Seitanidi

M. M.

(2012). Collaborative value creation: A review of partnering between nonprofits and businesses: Part I. Value creation spectrum and collaboration stages. Nonprofit and Voluntary Sector Quarterly, 41(5), 726–758. https://doi.org/10.1177/0899764012450777

Bailard

C. S.

Livingston

(2014). Crowdsourcing accountability in a Nigerian election. Journal of Information Technology & Politics, 11(4), 349–367. https://doi.org/10.1080/19331681.2014.947056

Brabham

D. C.

(2012). Motivations for participation in a crowdsourcing application to improve public engagement in transit planning. Journal of Applied Communication Research, 40(3), 307–328. https://doi.org/10.1080/00909882.2012.693940

Bryson

J. M.

Crosby

B. C.

Stone

M. M.

(2015). Designing and implementing cross sector collaborations: Needed and challenging. Public Administration Review, 75(5), 647–663. https://doi.org/10.1111/puar.12432

Cai

C. J.

Iqbal

S. T.

Teevan

(2016). Chain reactions: The impact of order on microtask chains. In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems, San Jose, CA, 123143–163154 May, 2016. ACM, pp. –. https://doi.org/10.1145/2858036.2858237

Chen

Aitamurto

(2019). Barriers for crowd’s Impact in crowdsourced policymaking: Civic data overload and filter hierarchy. International Public Management Journal, 22(1), 99–126. https://doi.org/10.1080/10967494.2018.1488780

Clarke

MacDonald

(2019). Outcomes to partners in multi-stakeholder cross-sector partnerships: A resource-based view. Business & Society, 58(2), 298–332. https://doi.org/10.1177/0007650316660534

Connelly

B. L.

Certo

S. T.

Ireland

R. D.

Reutzel

C. R.

(2011). Signaling theory: A review and assessment. Journal of Management, 37(1), 39–67. https://doi.org/10.1177/0149206310388419

Cooper

K. R.

(2021). For Everyone” means “For No One”: Membership tensions in community collaboration. Management Communication Quarterly, 35(3), 392–417. https://doi.org/10.1177/08933189211000324

10.

Cooper

K. R.

Shumate

(2012). Interorganizational collaboration explored through the bona fide network perspective. Management Communication Quarterly, 26(4), 623–654. https://doi.org/10.1177/0893318912462014

11.

Courtney

Dutta

(2017). Resolving information asymmetry: Signaling , endorsement, and crowdfunding success. Entrepreneurship Theory and Practice, 41(2), 265–290. https://doi.org/10.1111/etap.12267

12.

Crilly

(2011). Predicting stakeholder orientation in the multinational enterprise: A mid-range theory. Journal of International Business Studies, 42(5), 694–717. https://doi.org/10.1057/jibs.2010.57

13.

Durward

Blohm

Leimeister

J. M.

(2016). Rags to riches - how signaling behaviour causes a power shift in crowdsourcing markets. In Proceedings of European Conference on Information Systems (ECIS), Istanbul, Turkey, 12–15 June, 2016.

14.

Edelman (2015). Edelman trust barometer report. https://www.edelman.com/research/2015-edelman-trust-barometer

15.

Fischer

Reuber

(2007). The good, the bad, and the unfamiliar: The challenges of reputation formation facing new firms. Entrepreneurship Theory and Practice, 31(1), 53–75. https://doi.org/10.1111/j.1540-6520.2007.00163.x

16.

Fiss

P. C.

(2011). Building better causal theories: A fuzzy set approach to typologies in organization research. Academy of Management Journal, 54(2), 393–420. 10.5465/AMJ.2011.60263120

17.

Glaeser

E. L.

Hillis

Kominers

S. D.

Luca

(2016). Crowdsourcing city government: Using tournaments to improve inspection accuracy. American Economic Review, 106(5), 114–118. https://doi.org/10.1257/aer.p20161027

18.

Guth

K. L.

Brabham

D. C.

(2017). Finding the diamond in the rough: Exploring communication and platform in crowdsourcing performance. Communication Monographs, 84(4), 510–533. https://doi.org/10.1080/03637751.2017.1359748

19.

Hartley

Sørensen

Torfing

(2013). Collaborative innovation: A viable alternative to market competition and organizational entrepreneurship. Public Administration Review, 73(6), 821–830. https://doi.org/10.1111/puar.12136

20.

Hasson

(1997). Towards a general theory of biological signaling. Journal of Theoretical Biology, 185(2), 139–156. https://doi.org/10.1006/jtbi.1996.0258

21.

Hilgartner

Bosk

C. L.

(1988). The rise and fall of social problems: A public arena model. American Journal of Sociology, 94(1), 53–78. https://doi.org/10.1086/228951

22.

Horvát

E. Á.

Wachs

Wang

Hannák

(2018). The role of novelty in securing investors for equity crowdfunding campaigns. In Sixth AAAI Conference on Human Computation and Crowdsourcing. Zurich, 5–8 July, 2018.

23.

Howe

(2006). The rise of crowdsourcing. Wired Magazine, 14(6), 1–4. https://doi.org/10.1300/j021v26n03_01

24.

Kohler

Chesbrough

(2019). From collaborative community to competitive market: The quest to build a crowdsourcing platform for social innovation. R&D Management, 49(3), 356–368. https://doi.org/10.1111/radm.12372

25.

Kohler

Rutzler

(2018). Crowdsourcing social innovation in tourism: Insights on platform design. In Information and Communication Technologies in Tourism 2018 (pp. 64-76). Cham: Springer.

26.

Lakhani

K. R.

Fayard

A. L.

Levina

Pokrywa

S. H.

(2013). OpenIDEO. Harvard Business School Technology & Operations Management. Unit Case No. 612-066.

27.

Lember

Brandsen

Tõnurist

(2019). The potential impacts of digital technologies on co-production and co-creation. Public Management Review, 21(11), 1665–1686. https://doi.org/10.1080/14719037.2019.1619807

28.

Lim

Berry

F. S.

Lee

K. H.

(2016). Stakeholders in the same bed with different dreams: Semantic network analysis of issue interpretation in risk policy related to Mad Cow disease. Journal of Public Administration Research and Theory, 26(1), 79–93. https://doi.org/10.1093/jopart/muu052

29.

Liu

H. K.

(2017). Crowdsourcing government: Lessons from multiple disciplines. Public Administration Review, 77(5), 656–667. https://doi.org/10.1111/puar.12808

30.

Madianou

(2013). Humanitarian campaigns in social media: Network architectures and polymedia events. Journalism Studies, 14(2), 249–266. https://doi.org/10.1080/1461670X.2012.718558

31.

McCombs

M. E.

Shaw

D. L.

(1972). The agenda-setting function of mass media. Public Opinion Quarterly, 36(2), 176–187. https://doi.org/10.1086/267990

32.

McEvily

Giuseppe

Tortoriello

(2014). More formally: Rediscovering the missing link between formal organization and informal social structure. Academy of Management Annals, 8(1), 299–345. https://doi.org/10.5465/19416520.2014.885252

33.

Mergel

(2018). Open innovation in the public sector: Drivers and barriers for the adoption of Challenge.gov. Public Management Review, 20(5), 726–745. https://doi.org/10.1080/14719037.2017.1320044

34.

Mergel

Desouza

K. C.

(2013). Implementing open innovation in the public sector: The case of Challenge.gov. Public Administration Review, 73(6), 882–890. https://doi.org/10.1111/puar.12141

35.

Misangyi

V. F.

Acharya

A. G.

(2014). Substitutes or complements? A configurational examination of corporate governance mechanisms. Academy of Management Journal, 57(6), 1681–1705. https://doi.org/10.5465/amj.2012.0728

36.

Moleskis

Alegre

Canela

M. A.

(2019). Crowdfunding entrepreneurial or humanitarian needs? The influence of signals and biases on decisions. Nonprofit and Voluntary Sector Quarterly, 48(3), 552–571. https://doi.org/10.1177/0899764018802367

37.

Moss

T. W.

Neubaum

D. O.

Meyskens

(2015). The effect of virtuous and entrepreneurial orientations on microfinance lending and repayment: A signaling theory perspective. Entrepreneurship Theory and Practice, 39(1), 27–52. https://doi.org/10.1111/etap.12110

38.

Wang

(2022). A systematic literature review of open innovation in the public sector: Comparing barriers and governance strategies of digital and non-digital open innovation. Public Management Review, 24(4), 489–511. https://doi.org/10.1080/14719037.2020.1838787

39.

Mulgan

(2006). The process of social innovation. Innovations: Technology, Governance, Globalization, 1(2), 145–162. https://doi.org/10.1162/itgg.2006.1.2.145

40.

Nigam

Ocasio

(2010). Event attention, environmental sensemaking, and change in institutional logics: An inductive analysis of the effects of public attention to Clinton’s health care reform initiative. Organization Science, 21(4), 823–841. https://doi.org/10.1287/orsc.1090.0490

41.

Openideo (2019). How might we reimagine the end-of-life experience for ourselves and our loved ones? https://challenges.openideo.com/challenge/end-of-life/brief

42.

Przeworski

Teune

(1970). The logic of comparative social inquiry. Wiley-Interscience.

43.

Ragin

C. C.

(2008). Redesigning social inquiry: Fuzzy sets and beyond. University of Chicago Press.

44.

Roberts

Wanta

Dzwo

T. H.

(2002). Agenda setting and issue salience online. Communication Research, 29(4), 452–465. https://doi.org/10.1177/0093650202029004004

45.

Rogers

E. M.

(2003). Diffusion of innovations (5th ed.). Free Press.

46.

Salamon

(1987). Of market failure, voluntary failure, and third-party government. Journal of Voluntary Action Research, 16(1), 29–49. https://doi.org/10.1177/089976408701600104

47.

Schneider

C. Q.

Wagemann

(2012). Set-theoretic methods for the social sciences: A guide to qualitative comparative analysis. Cambridge University Press.

48.

Stephens

K. J.

Fulk

Monge

P. R.

(2009). Constrained choices in alliance formations: Cupids and organizational marriages. Human Relations, 62(4), 501–536. https://doi.org/10.1177/0018726708101982

49.

Stohl

(2014). Crowds, clouds, and community. Journal of Communication, 64(1), 1–19. https://doi.org/10.1111/jcom.12075

50.

Tam

(2015). News about corporate social responsibility (CSR): The interplay of intermedia agenda setting influences between corporate news releases and press coverage. Asian Journal of Business Ethics, 4(2), 117–130. https://doi.org/10.1007/s13520-015-0046-y

51.

Van De Ven

A. H.

Ganco

Hinings

C. R.

(2013). Returning to the frontier of contingency theory of organizational and institutional designs. Academy of Management Annals, 7(1), 393–440. https://doi.org/10.1080/19416520.2013.774981

52.

Vargo

C. J.

Guo

Amazeen

M. A.

(2018). The agenda-setting power of fake news: A big data analysis of the online media landscape from 2014 to 2016. New Media & Society, 20(5), 2028–2049. https://doi.org/10.1177/1461444817712086

53.

Vigoda-Gadot

Shoham

Schwabsky

Ruvio

(2005). Public sector innovation for the managerial and the post-managerial era: Promises and realities in a globalizing public administration. International Public Management Journal, 8(1), 57–81

54.

Wang

(2020). Marginality and team building in collaborative crowdsourcing. Online Information Review, 44(4), 827–846. https://doi.org/10.1108/oir-09-2018-0269

55.

Wang

(2022). Team diversity and team success in collaborative crowdsourcing. Communication Studies, 73(1), 68–84. https://doi.org/10.1080/10510974.2021.2011355

56.

Weick

K. E.

Sutcliffe

K. M.

Obstfeld

(2005). Organizing and the process of sensemaking. Organization Science, 16(4), 409–421. https://doi.org/10.1287/orsc.1050.0133

57.

Wen

Lin

(2016). Optimal fee structures of crowdsourcing platforms. Decision Sciences, 47(5), 820–850. https://doi.org/10.1111/deci.12201

58.

Yuan

Gasco-Hernandez

(2021). Open innovation in the public sector: Creating public value through civic hackathons. Public Management Review, 4(23), 523–544. https://doi.org/10.1080/14719037.2019.1695884