Do Nonprofit Responses to Social Movements Influence Public Perceptions and Willingness to Support? An Experimental Study

2.1 Nonprofit Advocacy and Social Movements

The nonprofit sector encompasses a diverse array of organizations in the United States and around the world. Early attempts to define and classify the nonprofit sector led to the creation of several typologies and classification systems, such as the NTEE classification of exempt entities created by the Urban Institute (Urban Institute 2019). One conceptual typology defines nonprofits according to two dimensions: supply-side driven versus demand-side driven and instrumental versus expressive (Frumkin 2002). The latter dimension deals with the primary function of the nonprofit in its day-to-day activities and operations. Instrumental nonprofits primarily provide services to individuals and groups, such as human service organizations like homeless shelters and food pantries. By contrast, expressive nonprofits primarily give their stakeholders opportunities to express and promote their values, beliefs, or identities; examples include religious congregations and associations serving particular ethnic groups. Expressive nonprofits also often advocate for issues that affect their stakeholders through education and outreach and involvement in the local, state, and federal policy process. Thus, a nonprofit whose primary mission and activities involve direct advocacy is also an example of an expressive nonprofit.

As Calderon and colleagues (2021) point out, “Today, it is accepted that advocacy, at least in some form, is a necessary function for the human service sector at-large” (p. 288). Thus, nonprofit organizations across this spectrum of functions and domains often challenge the status quo and push for systemic change, particularly during times of political and social unrest when public attention and support are high (e.g. Lamothe and Lavastida 2020). By engaging in or responding to social movements, nonprofits are seen to take a public stance on a particular issue, both affecting and affected by public policy (Bies et al. 2013). In this wider policy environment, nonprofits engaging primarily in instrumental functions tend to enjoy a favorable status as compared to nonprofits engaging primarily in expressive functions (Appe 2019). Moreover, the former is typically viewed as less politicized than the latter. However, advocacy can occur in both types and through both functions (Chin 2018). Nonprofit advocacy extends not only to their stakeholders but also to the causes relevant to them and their wider community (Alexander and Fernandez 2021), emerging as a “critical component” of the work of the sector as a whole (Guo and Saxton 2010).

Nonprofits frequently advocate for underrepresented client groups by amplifying their voices and promoting social justice (LeRoux 2014). Past research has shown that nonprofit organizations serving “specialized populations” are more likely to engage in advocacy (MacIndoe 2014, p. 309) and therefore may play a unique role in times of political and social unrest as they seek to represent their constituencies. For example, in a study of ethnic Korean nonprofits in Los Angeles in the aftermath of the 1992 Los Angeles riots, Chung (2005) highlights the importance of nonprofits’ political activism and solidarity-building as a means to mediate relationships both between the mainstream and the ethnic minority group and within members of the ethnic group. While these organizations may take different actions in service of their strategic objectives (e.g. indirectly through programmatic activities or directly through public protests), both approaches emerge as political in nature and in response to a period of extreme social unrest (Chung 2005). An earlier study by Coston, Cooper, and Sundeen (1993) examining the role of community organizations following the Los Angeles riots also confirms the importance of these localized nonprofits in mobilizing resources, connecting community members to needed services and assistance, and acting as mediators between individuals and other nonprofit and public agencies seeking to respond to crisis.

In terms of advocacy activities, community-based nonprofits in marginalized neighborhoods engage in both civic activities and social movements (Majic 2011). These actions include communicating with local or state governments on behalf of their clients, organizing campaigns, providing education, or supporting political protests (LeRoux 2014). These expressive forms of advocacy make issues more visible to society and call for immediate action by disrupting the status quo (Mosley 2011).

Although nonprofits advocate and mobilize their stakeholders around political and social issues, the 501(c)(3) status of public charities prevents them from engaging directly or “substantially” in some political activities, such as lobbying and campaigning for political candidates, with legal and practical ambiguity about the definition of “substantial” further inhibiting nonprofit involvement in these areas (Berry 2020). Nonprofits also face a perceptual barrier to direct and controversial political involvement in that they wish to avoid alienating potential donors who may hold opposing political views (Bell, Fryar, and Johnson 2021; LeRoux 2014). Moreover, Alexander and Fernandez (2021) contend that recent trends, such as professionalization and marketization of the sector, have compromised nonprofits’ ability “to perform essential political roles” (368) by rendering them more adverse to risk and controversy. Likewise, King and Griffin (2019) note that managerialism leads nonprofits to pursue seemingly apolitical instrumental objectives rather than political ones.

Despite these trends identified in the literature, recent and historical events point to nonprofits’ embeddedness at the intersection of social movements, policy processes, and crisis response, with their political role activated through both instrumental and expressive functions. In some cases, nonprofit organizations evolve from social movements, actively organizing and supporting advocacy activities by themselves as well as mobilizing others for political actions (Cress 1997). Examples include organizations for the homeless (Cress 1997) and feminist groups (Hyde 2000) that actively participate in political activities including protests to fight for their socially neglected client groups, as well as nonprofit coalitions resulting from broader social movements (Wiley 2022).

2.2 Public Perceptions of Nonprofit Organizations Engaging in Social Movements

Like any organization, nonprofits must be concerned with public perceptions of their work as a whole. With regard to nonprofit advocacy, a favorable image is critical in shaping public attitudes toward an issue or movement, while expectations about the public’s reaction – and any potential financial or political fallout – shape nonprofits’ decisions about how to respond (Bell, Fryar, and Johnson 2021; Lee 2021). Examining nonprofit involvement in the realm of climate change policy, Bies et al. (2013) investigate how nonprofits influence awareness, attitudes, beliefs, values, and knowledge of the public through their various activities. According to the authors, “In the modern world where instant communications, the internet, and an almost infinite array of communicating options are available, advocacy nonprofit institutions may be poised to be a significant cue giver to their members and others who trust them or their message” (Bies et al. 2013, 10). The authors find that the public’s perception of organizational legitimacy and trust is the most important factor in influencing their concern about the environment, their belief that their actions affected the environment, and their support for policies to mitigate climate change (Bies et al. 2013). Thus, nonprofits may be well-positioned to influence individuals’ beliefs, concerns, and behaviors, provided they are viewed as legitimate and trustworthy by the public. The degree to which the public perceives a nonprofit favorably and may therefore be willing to support both the cause in question and the nonprofit in particular will depend on the strategies (actions) employed by the organization. We explore this theme further in the sections below.

2.3 Public Perceptions of Political Protests and Support for the Organization

The literature on nonprofit advocacy is expansive, with several scholars outlining the various types of nonprofit advocacy, from direct lobbying to grassroots efforts (e.g. Guo and Saxton 2010; Riegel and Mumford 2022). However, there remains a gap in our understanding of how public perceptions of nonprofit organizations vary depending on the type of advocacy approaches utilized. In particular, we were unable to locate any existing studies of how nonprofits’ involvement in direct actions like protests shape public perceptions of that organization, and how these direct actions compare with other forms of advocacy.

At the individual level, protests are one of the major forms of political participation – a way for citizens to organize and influence their elected officials in addition to participation in institutional politics through voting or political campaigns (DiGrazia 2014). Attitudes and participation are often catalyzed by current events and shaped by social movements (Lamothe and Lavastida 2020). Over the past decades, protests have received growing scholarly attention, as we have witnessed low voting rates and decreasing levels of public confidence and trust in politics and governments (Ekman and Amnå 2012).

In the past, protests were believed to be organized mainly by marginalized groups of people, such as racial minorities and underserved communities, but the motivations for organizing protests have been broadened and come to be seen as a strategic action by a broad range of social groups (McVeigh and Smith 1999). Collective groups of protestors aim to achieve broader social and political change mainly by affecting public opinion (Louis 2009; Thomas and Louis 2014). Specifically, protests lead the public to perceive the current situation as illegitimate and unsatisfactory, which makes them want social change and sometimes motivates them to join in the protests (Leuprecht et al. 2010). Thus, it is important to understand how public perceptions are shaped by political protests and to consider successful strategies that shape positive and supportive perceptions of protests.

In general, the public is supportive of protests when they perceive social injustice, when they perceive the protests as effective in making actual changes, and when they identify themselves with the protesters (Van Zomeren, Postmes, and Spears 2008). Public opinion theorists argue that social movements led by minorities serve as a bottom-up stimulus that shifts how the public perceives contentious issues such as racism (Lee 2002). In particular, sudden, unpredictable, and highly visible events such as Black Lives Matter movement have an instant effect of politicizing public opinion toward issues such as racism and decrease favorability toward law enforcement (Reny and Newman 2021). In these situations, the public may be more supportive of protests organized to fight perceived injustices. Additionally, protest potential, or the willingness to engage in protests, is influenced by past participation, such that those who have participated in the past may be more willing to do so again in the future (Kwak 2022).

However, the public is not always supportive of protests. Extreme and violent forms of protest can cause negative public perceptions because the extremism and violence dilute individuals’ identification with the protestors and increase the perceived immorality of the movement (Feinberg, Willer, and Kovacheff 2020). When behaviors are viewed as violating basic social norms by causing harm to other people, individuals are likely to perceive those behaviors as immoral, leading to decreased levels of support (Schein and Gray 2018). If protests include blocking roads, damaging private or public property, or physically hurting others, individuals will begin to view the movement as immoral, and their potential positive emotional connection to the protesting group will be diminished (Feinberg, Willer, and Kovacheff 2020).

By contrast, scholars argue that peaceful and nonviolent protests are more effective than violent protests in cultivating widespread public appeal (Edwards and Arnon 2021). Peaceful protests such as sit-ins and strikes have lower thresholds for buy-in from bystanders since such protests cause fewer conflicts with authorities, and potential repression from authorities can create public sympathy for protesters (Chenoweth and Stephan 2011). Evidence also shows that nonviolent protests are more effective than violent protests in promoting public support of political movements and dealing with extremist groups (Jones and Libicki 2008; Stephan and Chenoweth 2008).

Because the extant literature indicates that public perceptions vary according to the type of action taken, we expect support for a nonprofit that engages in some action on the part of a social movement will depend on the specific action taken. We consider support as both overall support for the nonprofit’s actions in this specific situation and willingness to support the organization through donations of time (volunteering) and money. With regard to overall support for the nonprofit’s actions, we hypothesize the following:

As charitable organizations, nonprofits rely on direct public support in the form of monetary donations and volunteering to continue their work. As the trend of giving declines despite overall nonprofit sector growth (Giving USA 2024), it has become even more competitive for nonprofit organizations to secure individual giving. Empirical findings also indicate that when individuals view nonprofits favorably, they are more willing to give and volunteer (Meijer 2009; Peng, Kim, and Deat 2019; Sarstedt and Schloderer 2010; Willems et al. 2017). Thus, whether and how nonprofits’ stance toward political protests affects direct public support is of critical importance for nonprofits. We hypothesize the following:

2.4 Perceptions of the Organization as Socially Responsible

In addition to examining various facets of individual support, this study will also examine individuals’ perceptions of an organization as socially responsible. Organizational social responsibility (SR), more commonly known as corporate social responsibility (CSR), refers to organizations holding moral obligations to society in which they aim to behave ethically in terms of their operation, as well as how they deal with their stakeholders (Snider, Hill, and Martin 2003, 175). Beyond the fact that social responsibility captures one aspect of stakeholders’ perception and evaluation of organizations, studies also show that an organization’s social contribution is positively associated with organizational reputation (Fombrun and Shanley 1990). Moreover, empirical evidence suggests that organizational social responsibility influences how stakeholders view the reputation of an organization more strongly than other factors, such as financial performance or leadership (Eisenegger and Schranz 2011). The perceived social responsibility even mitigates the negative stakeholder perceptions that might have been caused by reputational shocks like organizational scandals (Minor and Morgan 2011). Organizational social responsibility also leads to a greater level of stakeholder support including trustworthiness (Sisodia, Wolfe, and Sheth 2003), customer loyalty (Sen, Bhattacharya, and Korschun 2006), and employee commitment and satisfaction (Ellemers et al. 2011).

In the context of nonprofit organizations, organizational social responsibility might seem like an obvious virtue that their stakeholders expect, but as mission-based organizations, nonprofit organizations always need to hone their approach in responding to their social calling to provide essential services and goods to the population mostly neglected by the government and business (Carroll 1991). In addition to this social obligation, nonprofits also need to abide by the moral obligation to operate within laws and regulations (Pope et al. 2018). Finally, organizational social responsibility highlights the importance of nonprofits being held accountable for their various stakeholders by making their internal information public and having their performance audited by the external entities (Gibelman and Gelman 2004; Gugerty and Prakash 2010). Given the significance of perceived social responsibility for nonprofit organizations, this study will further investigate this concept and how it is shaped by nonprofits’ responses to social movements by testing the following hypotheses:

Public support for protests tends to increase as actions are taken, provided that those actions are perceived as peaceful, legal, and in line with expected norms. However, public support may take a downturn if actions cross the line of what is considered reasonable and acceptable. Thus, we expect that the public’s perceptions of a nonprofit will vary depending on the type of action it takes in response to a social justice movement.

3.1 Sampling, Recruitment, and Randomization

We designed the experiment in Qualtrics and recruited panels through an online survey platform, Amazon Mechanical Turk (MTurk). We restricted the pool of respondents to those who were 18 years or older and resided in the U.S. We fielded the first round of the experiment with the total number of 161 respondents based on the power analysis that we conducted on G*Power program. The program recommended 138 respondents to find significant results, but we collected an additional 42 participants to ensure data quality by filtering inattentive responses. We then ran a second round of the experiment, which included additional questions, such as willingness to donate, and generated a total of 202 additional responses. The final sample size was 363 participants.

The experiment included a vignette along with four different experimental manipulations randomly assigned to the participants. The manipulations presented different scenarios for how a fictional nonprofit responded to a social justice movement advocating for police reform in their local community, amidst growing national attention and concerns about the use of excessive force by police against African Americans. The vignette presented information about a community-based organization called Community Connects that provides services in an area where the majority of residents are racial minorities. The vignette also gave basic information about the nonprofit’s mission, services, target population, and budget. After reading the vignette, participants were randomly assigned to one of four treatment groups outlining various organizational responses to the social justice movement. At the end of the survey, we debriefed participants about the purpose of the experiment and the rationale for using a fictional organization. (See the Supplementary Document for the full text of the base scenario and vignettes.)

MTurk accommodates random assignment by recruiting individual survey participants through their platform and then randomly assigning experimental manipulations to their panel. To ensure this process functioned properly, we conducted balance tests, which showed that samples across the four treatment groups did not have statistically significant differences in demographic characteristics, signaling successful randomization of our study sample.

3.2 Data Quality Checks

Public and nonprofit management scholars are increasingly using online survey platforms to study various topics, particularly those related to individuals’ cognitive processes (e.g. Pedersen, Stritch, and Taggart 2017; Tremblay-Boire and Prakash 2019). MTurk is one survey platform that offers low cost and high accessibility, as well as a large pool of potential respondents to enable efficient and timely data collection. However, as with other experimental tools, MTurk poses some potential challenges identified by previous research. First, it has been suggested that survey participants spend too little time taking surveys so they can take as many surveys as possible (Stritch, Pedersen, and Taggart 2017). Second, participants may seek to take the same survey multiple times, looking for ways to bypass restrictions that researchers set through virtual provider servers (VPS) and virtual provider networks (VPN). Third, participants may also try to work around eligibility criteria, such as geographic location, taking the survey even if they do not qualify for participation. Finally, there are also cases in which researchers find that automated completion scripts (e.g. bots) are used to complete surveys, not actual human participants.

To address such issues with data quality and unidentifiable responses, we adopted various screening methods and attention checks to ensure the data’s validity. First, the experiment implemented the screening protocols Burleigh, Kennedy, and Clifford (2018) developed to filter out respondents who use VPS or VPN and those participating in the survey outside the U.S. These measures help to eliminate the issue of participants taking the same survey more than once, as well as ensuring that participants meet the eligibility criterion of living within the U.S. Second, the survey included a Recaptcha question to screen out automated scripts such as bots. Third, an attention check question on the mission of the nonprofit was presented after the experimental interventions to test whether the respondents paid appropriate attention to the vignette and manipulation, as well as to add another bot check through the use of an open-ended question. The final sample after poor quality responses were eliminated through these quality control checks was 180 respondents. In terms of representativeness, the demographic characteristics of our sample overrepresent male (62 % vs. 49.5 %), white (82 % vs. 59.3 %), bachelor’s degree (61 % vs. 23.5 %), and low-to-mid income (76 % vs. 52.3 %) groups of respondents (see Table B for a full comparison).

3.3 Measures

The independent variable in this study is the experimental intervention on how nonprofits engaged with a social justice movement centered on police reform in their community. We randomly assigned survey participants to four different levels of a nonprofit’s involvement in the social movement, ranging from radical involvement to no actions at all. The first group is the group in which the nonprofit did not take any action. In the second group, the nonprofit addressed the issues raised by the social justice movement through the creation of new programming within their organization. In the third treatment group, the nonprofit engaged in direct action in support of the social justice movement through organizing and conducting permitted demonstrations and public events. In the final treatment group, the nonprofit engaged in direct action through participation in unpermitted protests involving more extreme actions such as blocking traffic and confronting local authorities.

This study includes multiple dependent variables encompassing stakeholder perceptions and support. First, we measured overall support by asking survey participants how much they supported the organization’s actions on a 5-point Likert scale (from 1 = strongly disagree to 5 = strongly agree). Second, we asked respondents about their willingness to give by asking them to choose the amount they are willing to donate on a slider that ranges from 0 to 100 US dollars. We had some missing responses, resulting in a slightly smaller sample size (N = 168). Due to the outliers, we logged this measure for the analysis. Third, we asked about their willingness to volunteer using a 5-point Likert scale (from 1 = strongly unlikely to 5 = strongly likely). Finally, we asked participants about their perception of the organization as socially responsible. Participants were asked to indicate their agreement with the statement that “the organization is socially responsible” on a 5-point Likert scale (from 1 = strongly disagree to 5 = strongly agree). Finally, we included a moderator on whether or not individuals had previously participated in protests (1 = yes, 0 = no). Details on the operationalization of the independent, dependent, and control variables examined in this study are available in the Supplementary Document.

5.1 Limitations and Directions for Future Research

As one of the first attempts to examine nonprofits’ role in social movements in relation to public perceptions and support, we acknowledge the limitations of our study and suggest ways that future research might address them. First, this paper contains a single case study of a hypothetical community-based nonprofit organization with a relatively small sample size. Future studies might find more significant effects of nonprofits’ stance on social movements in a spectrum of intensity on other types of public support and perceptions with a larger dataset. Moreover, given that the study utilized an online experiment with convenience sampling via MTurk, the generalizability of the findings beyond the study context, which describes a human services-related nonprofit responding to a specific social issue at a single point in time, may be limited. As mentioned in the data section, we have attempted to mitigate some of the challenges associated with using MTurk by conducting a series of rigorous quality checks. Nevertheless, future research is needed to investigate nonprofits’ involvement in advocacy activities in other service domains and policy settings to see whether the results are consistent with this study with a larger and more representative sample.

Second, this study asked participants about their intent to support the organization and did not measure actual actions undertaken in support of a nonprofit (e.g. donations or volunteering). Some scholars have noted the limitations of asking people about their intentions rather than their actual behaviors, arguing that the two may not align (Nederhof 1985). Future research could try to mitigate this issue by providing participants with a small amount of money to donate to a real organization, or by asking actual donors or volunteers about their perceptions of a nonprofit’s advocacy work.

Third, although the current study had three different levels of intensity in terms of nonprofits’ political actions, the scope of intensity could be broken down even further by including more extreme types of protest. For example, future research could investigate how more extreme types of action in support of social movements influence public perceptions and willingness support, as existing studies in political science suggest such activities are perceived strongly negatively by the public.

Finally, in an effort to examine public perceptions of nonprofit responses to a social movement, we chose to conduct the experiment using a fictional nonprofit organization that was deliberately broad in scope so as to make it relatable to a general audience. However, the use of a hypothetical case organization creates some additional limitations that leave room for future exploration. For instance, while we were able to investigate the moderating effect of participants’ prior involvement in protests in general, individuals’ personal connections and past involvement with the specific nonprofit cannot be captured. It is likely that individuals with a direct connection to the nonprofit would react differently than those with no prior connection or involvement. Additionally, we were not able to test the effects of local context, and the actions of peer organizations, on people’s assessment of the specific nonprofit’s actions, beyond providing some basic details about the demographics of the community and the nonprofit’s clientele. Yet, if the nonprofit is operating in a community where all other organizations have taken an action or responded in some way, then the public may view a lack of action more negatively as a result. Given these limitations, future research could build on our findings by testing the salience of these factors through in-depth analysis of a real case example or by surveying participants about the actions of actual nonprofit organizations that are well-known nationally or within a specific study context.