The Social Meanings of the Third Sector: How Action and Purpose Shape Everyday Understandings of “Nonprofit”

Abstract

How do nonspecialists of nonprofit practice, law, and scholarship conceptualize the third sector? This article explores the everyday meanings of nonprofit organization and action empirically by reporting on a survey-based exercise in which research participants coded statements describing qualitatively different interactions between various types of entities. The survey, drawing on Crawford and Ostrom’s grammar of institutions, allows for an examination of how lay observers make sense of the sectoral boundaries that occupy specialists’ attention. We find that research participants are less prone to code interactions consistently with the nominal sectors of the organizations presented to them and more inclined to code the interactions based on the types of actions organizations take and their rationale for those actions. We argue that understanding the everyday meaning of nonprofit has important implications for theory and practice.

Keywords

sector theory blurring boundaries nonprofitness social meaning

Since the origin of the field, nonprofit scholars have been trying to identify what constitutes the third sector and makes nonprofit organizations unique. In an era when scholars challenge whether public, nonprofit, and for-profit organizations differ from each other in any meaningful respect (Bromley & Meyer, 2017), this study contributes in two ways to this long-standing existential question. First, we develop an empirical approach for examining differences between institutional characteristics of organizational interactions, drawing on Crawford and Ostrom’s (1995) grammar of institutions. Second, we deploy the approach to examine the perspectives of ordinary people about what types of interactions should (or should not) be classified as nonprofit/charitable.

However taken for granted in nonprofit scholarship, it is not actually obvious what organizations and activities should (or should not) be classified as being part of the nonprofit sector. Some scholars view nonprofits merely as a special case of private corporate activity, with their own peculiar tax benefits and limitations (Jackson, 2006; Louis, 2021). Others describe nonprofit organizations variously as extensions of government (Salamon, 1987, 2002) or as mixed entities exhibiting elements of both public and private institutions (Brandsen et al., 2005; Fischer et al., 2011; Moulton & Eckerd, 2012; Skelcher & Smith, 2015; Zolin & Kropp, 2008). And some believe not only in the presence of a third sector, but of a fourth as well. The rise of social entrepreneurship, including its new corporate forms such as benefit corporations and L3Cs, begs the question of not only the boundaries between public–private distinctions and a third sector but also the boundaries between the third sector and a potential fourth (Sabeti, 2009; Sánchez-Hernández et al., 2021).

Much of this discussion of what does or does not constitute nonprofit structure and function, or what might constitute the nonprofit sector as a whole, is based in scholars’ interpretations of law, theory, and empirical study. Without intention, it can ignore the perspectives of ordinary people whose time, donations, tax dollars, and social behaviors all inform and are affected by the ways in which they perceive and interact with nonprofit organizations and other charitable activity. The engagement of ordinary people in civil society may include nonregistered charitable activity (Smith, 1997) and/or native organizing (R. K. Mitchell et al., 2016), prosocial corporate activity (C. E. Mitchell, 2016), or even simply discussions about what charitable organizations should or should not be doing and how this affects their communities, public policy, and their place in society (Putnam, 2000).

This article continues the process of articulating the meanings of nonprofit as perceived by ordinary people. We ask, “How do everyday people—by which we mean nonspecialists of nonprofit practice, law, and scholarship—conceptualize the third sector?” Nonprofit organizations and the notion of economic “sectors” are embedded in complex social settings, so if we are to understand their place in modern society, we must investigate how the great majority of society (i.e., the nonspecialists) regards them. Implied in the research question, of course, is that there are potentially multiple meanings of the nonprofit sector and that the everyday meanings might be different from technical, legal, or scholarly meanings. Empirically, we show that there is an identifiable everyday meaning of nonprofit and that it is unique from the typical constructions in the nonprofit literatures.

The Need for an Everyday View

Although the word nonprofit has a specific technical meaning, its everyday meaning could in fact be quite different. There is some evidence that nonspecialists have a hard time differentiating nonprofit from other types of organizations (Handy et al., 2010; Permut, 1981). In Handy et al.’s (2010) study, for example, most participants did a poor job identifying nonprofit organizations on a list that included government and for-profit organizations too. Although these results can be interpreted in many ways, for our purposes they at least provide empirical justification for further investigating how nonspecialists understand the meaning of the term ‘nonprofit’.

Also relevant to our purposes here, scholars working at the intersection of law and society have become increasingly interested in legal consciousness, or how everyday people regard the legal constructions that touch their lives (e.g., Cowan, 2004; Ewick & Silbey, 1991; Sarat, 1990). They observe that laws and societies have a reciprocal, mutually constitutive relationship, so understanding any field that comes under the purview of law requires looking not only at the law but also at how law is received and interpreted by people (Silbey, 2005). The benefit of doing so is not just academic. People understand and act based on their interpretations of law, and those interpretations (and therefore actions) are not always synchronized with laws “on the books.” The difference between formal legal rules and “rules in use” is an important consideration in institutional analysis and design (Ostrom, 2009).

To the point of this article, what would it matter if there is an everyday meaning of nonprofit that is different than the technical or legal meaning of the word? There are several implications. First, good social science and social policy depends on having an accurate understanding of the phenomenon in question. Although we are not the first to look beyond law and regulation to understand how people perceive nonprofits (Farwell et al., 2019; Handy et al., 2010; O’Neill, 2009), any holistic account must include social meanings as well. Second, for those with an interest—normative or otherwise—in tracking the robustness of the nonprofit sector, it is worth remembering that the durability of institutions relies not only on their legal grounding but on their sociocultural embeddedness as well (Child et al., 2016; Scott, 2014).

Third, we cannot understand how and why people organize to address social problems (and with what consequences) in the way that they do if we do not comprehend how people perceive what nonprofits are and what they view as the legitimate scope of nonprofit action. Nonprofit organizations are not simply passive tools of government (Salamon, 2002). Rather, multidimensional people establish and inhabit nonprofit organizations and sectors (Child et al., 2015; Young, 1983) and they bring with them their own ideas about what it means to form nonprofit organizations and do nonprofit work, which may or may not accord with the interests and ideas of governments and policymakers. Fourth, we cannot fully understand practical and theoretical developments, such as blurring boundaries (Bromley & Meyer, 2017; Dees & Anderson, 2003), including social enterprise and other hybrid forms (Battilana & Lee, 2014; Reiser, 2010), as well as new notions of nonprofitness (Knutsen & Brock, 2014), if we do not understand the everyday meanings of nonprofit beyond its legal definition.

Applying a Grammar of Institutions

Our empirical approach is informed by Crawford and Ostrom’s grammar of institutions (Crawford & Ostrom, 1995; McGinnis, 2011), a framework for identifying and differentiating strategies, norms, and rules as unique types of institutional statements. Especially important for our purposes, the grammar assumes broadly that institutions are represented to some degree in language and that dissecting this language provides a way of making sense of the governing institution(s).

Elements of the grammar (Crawford & Ostrom, 1995) include the attributes of the actors to whom the institutional statement applies; the deontic content of the statement, specifying which actions may, must, or must not be undertaken by the relevant actor; the aim or target of the action; the conditions under which this particular statement is deemed appropriate or relevant for application; and an “or else,” which specifies the consequences on those who fail to implement the statement as intended. Siddiki et al. (2011) propose adding an additional component, the object, which is the “inanimate or animate part of a statement that is the receiver of the action” executed by the actor (p. 85).

Applying the grammar to the case at hand suggests three ways of conceptualizing what might shape notions of what it means to be and do nonprofit. These are a participant-centric perspective, focused on the players in a given interaction (both the actor and the object of the action); an action-centric perspective that gives primacy to what is actually being done (such as engaging in selling or donative behavior); and an intention-centric perspective that depends primarily on the aim of the interaction (whether to benefit the actor, the recipient, or something or someone external to the interacting pair). We examine all three of these perspectives.

Data and Methods

To answer our research question, we developed a methodology in which nonspecialist survey respondents engaged in a coding exercise. As we designed the methodology outlined below, our primary concerns were the purpose of the study, the knowledge and vernacular of the respondents, and the need to manage respondent cognitive load and response fatigue (Deck & Jahedi, 2015; Egleston et al., 2011). We refined the instrumentation through pilot testing and informal cognitive interviews (Desimone & Le Floch, 2004).

By using the exercise, we sought to explore alternative constructions of nonprofits (e.g., as part of the private sector, part of the public sector, a combination of the two, or something altogether different). Such constructions are found in various scholarly literatures (Brandsen et al., 2005; Fischer et al., 2011; Jackson, 2006; Louis, 2021; Moulton & Eckerd, 2012; Salamon, 1987, 2002; Skelcher & Smith, 2015; Zolin & Kropp, 2008) and we wanted to test whether and how academic constructions hold up in the face of public perception. We thus shaped the coding exercise consciously to narrow the results to those that would help us distinguish, first, a three-sector model from a two-sector model and, second, a three-sector model from a four-sector model.

Procedure

Drawing inspiration from Crawford and Ostrom’s (1995) grammar of institutions, we identify key elements that distinguish different types of exchanges from one another. These elements include the participants in the exchange, comprising the people or organizations that initiate the interaction (i.e., a government entity, a business firm, a nonprofit organization, or an individual) and the receiving party in the exchange (i.e., a government entity, a business firm, a nonprofit organization, or an individual); the nature of the exchange action (i.e., sale vs. donation); and the intended purpose of the exchange (i.e., to make the initiator better off, to make the recipient better off, or to make the world a better place).

Each of these elements of an exchange is used as a manipulated variable in our study. Table 1 identifies how each component of the framework is operationalized in our study to create a set of statements. It summarizes the slightly more intuitive terminology (actor, action, recipient, and aim) that we use in the remainder of the article. By exhausting all possible combinations of the elements, we created a total of 96 statements (four actor conditions × two action conditions × four recipient conditions × three aims = 96 total statements). Each study participant was randomly assigned to one actor condition and therefore saw only 24 of the statements.

Table 1.

Elements of the Grammar.

Actor	Action	Recipient	Aim
A business. . .	. . . sells services to. . . . . . donates services to. . .	. . . a business. . . . . . a nonprofit organization. . . . . . a government agency. . . . . . an individual. . .	. . . in order to make the [actor] better off.
A nonprofit organization. . .			. . . in order to make the [recipient] better off.
A government agency. . .			. . . in order to make the world a better place.
An individual. . .

Note. If the probability of categorizing the interaction as nonprofit/charity is highly sensitive to manipulations to the actor and/or recipient variables, then participants’ views are considered participant-centric. If the probability is highly sensitive to manipulations to the action variable, then their views are considered action-centric. And if the probability is highly sensitive to manipulations to the aim variable, then their views are considered intention-centric.

Viewing the interaction statements one at a time and in random order (using Qualtrics’s built-in randomizer), the participant coded each into one of four categories: (a) “This interaction describes what we mean when we call something ‘business’”; (b) “This interaction describes what we mean when we call something ‘government’”; (c) “This interaction describes a mix between what we mean by ‘business’ and ‘government’”; and (d) “This interaction describes something different from ‘business’ or ‘government.’” Importantly, research participants were not requested to identify the actors involved but, rather, to describe the interaction. This procedure likely produced a conservative estimate of the participants’ willingness to code an interaction as nonprofit or charity.

The answer choices allow us to examine how participants distinguish nonprofits from government and business, acknowledge the possibility of mixed institutions, and consider the existence of a third sector (Salamon, 1987; Steinberg, 2006; Witesman, 2016). The first three answer choices represent a two-sector model, with the first two options identifying the market and government sectors, respectively. The third answer choice represents a mixed-sector interaction in which the blurring occurs between markets and government. And the fourth answer option provides an opportunity for coders to indicate that an interaction statement does not fit the two-sector (market/government) paradigm. Only by selecting the fourth option would participants be given an opportunity to identify the interaction as nonprofit/charity, ensuring that they were purposefully identifying it as such.

To ensure that participants focused on characterizing the interaction and did not mistake the exercise for one in which they were meant to simply identify the principal actor, each participant received only one actor condition. The language of the answer choices was designed to invoke a definitional, shared sense of meaning from respondents. By using the ambiguous but collective “we,” participants were prompted to consider that the category should have meaning not just to themselves but to others as well. The phrasing “what we mean when we call something. . .” further emphasizes shared meaning in a definitional sense (Bodle, 2019).

After coding the 24 statements, we identified those statements that participants marked as “different” from business and government. These statements then reappeared one by one and participants were asked to further classify them according to the following two options: (a) “This interaction describes what we mean when we call something ‘nonprofit/charity,’” and (b) “This interaction describes something different from ‘nonprofit/charity.’” It is the first of these options that we focus on as our dependent variable in this study.

Finally, we identified those statements classified by respondents as “different” from nonprofit/charity (and thus also previously classified as different from business and government) and presented them to participants one final time, all at once, asking two open-ended questions about the remaining, nonconforming statements: (a) “How are the above statements different from business, government, and nonprofit/charitable types of interactions?” and (b) “What would you call the types of interactions listed here?” These open-ended questions showed us how our nonspecialist coders conceived of activities that do not fit easily into a three-sector framework. They also served as a validation check to verify that respondents were understanding and thoughtfully engaging the coding exercises.

The procedure outlined above demonstrates that we placed the initial burden of proof on nonprofit status, as an interaction would only be coded as such if the coder rejected the idea that an interaction could be described using a two-sector model framework: as a business or government interaction, or a combination of the two. For “third sector” interactions, we then placed the burden of proof on the fourth sector, as a coder must have identified an interaction as explicitly different from nonprofit/charity to qualify for a category outside the three-sector model.

This approach stands in contrast to other ways of uncovering everyday meaning. Ethnographers, for instance, use qualitative methods and an interpretive style of analysis to understand how people make sense of their worlds. Ethnomethodologists employ a variety of strategies, including fine-grained conversation analysis, to inspect the co-creation of meaning and order, typically without imposing a theoretical model such as what we have in the institutional grammar (Garfinkel, 1988; Maynard & Clayman, 1991). We view our approach as complementary to these alternatives. Although more inductive approaches benefit from their open-ended nature, the method we rely on has the advantage of providing a highly systematic, structured read of the data, which renders patterns discernible and intuitive. Both styles, we argue, are needful as ways of understanding the meanings people attach to the words they use.

Participants

For data, we employed workers from an online labor market, Amazon Mechanical Turk (AMT; Berinsky et al., 2012; Paolacci & Chandler, 2014). A considerable amount of scholarship focuses on the utility of AMT samples for social science research (Buhrmester et al., 2018). The broad consensus is that, with the appropriate caveats and precautions, AMT samples perform quite well in comparison to other samples (Shank, 2016), even nationally representative ones (Coppock & McClellan, 2019).

We posted a link to the survey at AMT in November 2016. Participants were limited to adults who had completed at least 1,000 tasks previously and had high approval ratings (>95%). They received US$2.75 for completing the survey, which, based on the average time to completion, was comparable to a US$9.17 hourly wage.

Major concerns about AMT samples include inattention, dishonesty, and attrition. We addressed these by using practice and attention questions, including a reminder and encouragement to pay attention when the study was at its mid-point, allowing participants to participate in the study only once, and paying attention to time-to-completion. We used a Captcha validation challenge at the outset of the survey to ensure that the coding tasks were being completed by people and not software. We limited the number of initial statements to 24 to maintain a reasonable cognitive load, and we presented the statements one at a time and in random order to reduce respondent fatigue and primacy effects. We also examined the qualitative responses for evidence that respondents understood and actively engaged the task.

We dropped 65 participants from the sample who missed at least one attention question and/or completed the survey in less than 3 min. Reading the statements could scarcely be completed in 3 min (let alone coding them and answering follow-up questions), so we concluded that participants who provided answers in less than that amount of time were not paying sufficient attention to the questions to warrant including their answers in the data set. The median time to complete our survey instrument was 9.8 min, with some respondents taking much longer. Table 2 shows the sample characteristics. The study was approved by our university institutional review board.

Table 2.

Sample Characteristics.

Variable	Description	n	M	SD	Minimum	Maximum
Age	Age of participant in years	481	36.74	11.59	18	73
Female	1 if participant is female; else 0	481	0.50	0.50	0	1
White	1 if participant is white; else 0	481	0.77	0.42	0	1
Partner	1 if participant is married or living with a life partner; else 0	481	0.45	0.50	0	1
College	1 if participant holds a bachelor’s degree or higher; else 0	481	0.49	0.50	0	1
Income US$60k	1 if participant earns US$60,000 or more per year; else 0	481	0.16	0.37	0	1
Full-time	1 if participant is employed full-time; else 0	481	0.60	0.49	0	1
Management	Number of years experience paid work in management	481	3.06	5.50	0	44
Nonprofit	Number of years experience working for nonprofit org.	481	1.56	4.32	0	40

Analytic Strategy

The dependent variable of interest for this study is binary, where 1 indicates that the participant categorized the statement as describing a nonprofit/charitable interaction and 0 indicates that the participant did not categorize the interaction as nonprofit/charitable. To be coded as a 1, the participant would have first declared that the interaction “describes something different from ‘business’ or ‘government.’” With the statement presented again, the research participant would have then selected, “This interaction describes what we mean when we call something ‘nonprofit/charity’” instead of “This interaction describes something different from ‘nonprofit/charity.’” The variable would be coded as a 0, then, if the research participant coded the interaction as business, government, a mix between business and government, or something different from nonprofit/charity (see Online Appendix A).

We regress this variable on the characteristics of the interaction (the actor, the action, the recipient, and the aim of the interaction) using three different types of logistic regression models (see Online Appendix B). Then, for ease in interpretation, we present the results as predicted probabilities, estimated using Stata’s margins command.

The first two models are mixed effects logistic regression models, with errors clustered by research participant ID. The shortcoming of these models is that they do not take account of the fact that research participants were also nested in the assignment condition (e.g., a participant assigned to the government condition received only statements with a government as the actor).

The third model, also a mixed effects logistic regression, improves on the first by making adjustments for the nested nature of the data—where repeated observations are nested within research participants, who are themselves nested within the assignment conditions. However, the assignment condition (i.e., actor) is also a variable of interest and using the assignment condition as one of the levels in the model prevents it from also being estimated as a coefficient in the model.

To be sensitive to the nested nature of the data and also estimate the effect of actor on the dependent variable, we ran a series of models similar to the first model but restricted to participants who shared the assignment condition (and thus with the actor variable omitted). Although Models 4 to 7 do not provide a coefficient for the actor, we are able to generate predicted probabilities for each assignment. Using the same predicted probability parameters for each model allows us to compare the predicted probabilities for each actor given those conditions.

We consider the participants’ views to be participant-centric if the probability of categorizing the interaction as nonprofit/charity is highly sensitive to manipulations to the actor and recipient variables. We consider them action-centric if the probability is highly sensitive to manipulations to the action variable, and intention-centric if the probability is highly sensitive to manipulations to the aim variable. These are not statistical distinctions. Rather, we use the labels as orienting concepts to describe general patterns in the data.

Results

Figure 1 shows the probabilities associated with each variation in the statements coded by research participants. These probabilities are based on the regression models in Online Appendix B. For instance, the first row shows the probability of a statement being coded as nonprofit/charity if the actor was a nonprofit, the action is to donate, the recipient is an individual, and the aim is to make the world a better place. This interaction had a .63 probability of being coded as nonprofit or charity. Each of the remaining rows in Figure 1 represents one (and only one) variation on those four conditions, with the baseline condition repeated three more times for ease in comparison.

Figure 1.

Probability of an interaction being coded as nonprofit/charity.

The insight that Figure 1 provides is less about any individual predicted value and more in how they compare to the baseline condition. This is the condition against which the others should be compared. It is important to remember here that research participants were asked to characterize the interaction, not to categorize the organizations involved in the interaction. These probabilities, then, identify what elements of the institutional grammar (actor, action, recipient, or aim) matter most in shaping the research participants’ perceptions. For instance, it is clear that participants see government action—when government is mentioned as a recipient, but especially when it is described as the instigating actor—as something quite different from nonprofit or charitable action. The probability of a statement being coded as nonprofit or charity is only .16 when government is the focal actor. More surprising, perhaps, is that the difference to research participants between a nonprofit versus a business donating to an individual to make the world a better place is rather modest (.63 vs. .53). This supports the notion that the boundaries separating the business and nonprofit sectors are to some degree blurred.

Other aspects of the grammar also have a noteworthy impact. For example, participants were largely unwilling to associate selling behavior with nonprofit or charitable action. Thus, a statement describing a nonprofit intending to make the world a better place by selling something to an individual had only a .22 probability of being coded as nonprofit or charitable action. This is approximately a .40 change in probability when compared with a nonprofit intending to benefit the world by donating to an individual (.63 − .22), although the ultimate aim is the same. Besides the penalty associated with a government initiating the interaction, the switch to selling behavior is the most penalized variation. The penalty to self-focused behavior is nearly as high, however. Even if a statement describes a nonprofit donating to an individual, the probability of a coder describing that interaction as nonprofit or charitable falls by .38 if the interaction is seen as benefiting the nonprofit rather than “the world” (.63 − .25).

Once government as an actor is removed from consideration, one of the key differences in how participants view these interactions is that the type of organization undertaking the action is less consequential than the nature of the action and its aim. Yes, government as an actor changes the interaction for research participants. But when it comes to variation based on whether the actor is a nonprofit, business, or individual, the variation in participants’ predictions is rather small in comparison with the effect of changing the type of action or its aim. The maximum change of probability when the actor and recipient variables are manipulated (excluding government) is .10 and .17, respectively. The maximum change in probability when the action variable is manipulated, however, is .41. It is .38 when the aim variable is manipulated.

Importantly, and counterintuitively given the focus on sector boundaries, notions of what constitutes nonprofit does not fall simply along sector lines. This is more evident if we examine the combinations of factors that reflect real-world interactions. In Figure 2, we introduce nomenclature for 12 different types of interactions common in today’s economy, which are based on different combinations of the focal elements in our institutional framework. For instance, one example of corporate philanthropy is a business donating to a nonprofit to benefit the world, which has a .63 probability of being coded as nonprofit or charity. This is perceived very differently from enlightened self-interest, operationalized in Figure 2 as a business donating to a nonprofit to benefit the business (.16). Increasingly popular, sector-blurring types of interactions—such as nonprofit enterprise, social enterprise, and nonprofit commercial income—are seen to have a relatively low likelihood of being coded as nonprofit or charity (respectively, .22, .09, and .05).

Figure 2.

Probability of common interactions being coded as nonprofit/charity.

We suspect these findings are at odds with how specialists tend to think of nonprofit or business activity. Many interactions that they would likely see as falling squarely within the nonprofit sector, such as nonprofit collaborations, nonprofit enterprise, and nonprofit contracting, are often judged by coders as not being nonprofit or charity. On the contrary, some business-centered activities are as likely to be coded as nonprofit/charity as are more conventional nonprofit activities. For instance, businesses donating to charities to benefit themselves (.16) are nearly as likely to be seen as nonprofit or charity as are nonprofit collaborations (.21).

Why is this the case? For our research participants, it is the action, with the intended outcome an important consideration, that makes an interaction nonprofit or charitable. Plotting examples of real-world cases next to each other highlights in sharp relief the finding suggested in Figure 1 that our research participants are not inclined to base their judgments on the instigator of an interaction, especially if the organization in question is not a government.

The models in Online Appendix B, which are the basis for the predicted probabilities in Figures 1 and 2, have as independent variables the actor, action, recipient, and aim of the interaction. In supplemental models, we included controls for gender, marital status, education, race, experience working in a nonprofit organization, income, and age. Adding these variables does not change the predicted probabilities reported above in any material manner, although they reduce the working sample size, nor do they indicate that any of these demographic characteristics consistently shaped assessments of what it means to be or do nonprofit. (Results available on request.)

For participants who indicated that the statement described an interaction as both “something different from business or government” and “something different from nonprofit/charity,” we asked how the interaction statements were different from business, government, and nonprofit/charitable types of interactions. Their written answers do not suggest a cohesive understanding of activity occurring outside of a three-sector framework—a so-called fourth sector. They were, instead, inconsistent and equivocal. Some participants resorted to three-sector terminology although they had declared already that the interaction was different from business, government, and nonprofit/charity. Others, rather than articulating a distinctly different type of activity, simply observed how difficult it was to code the interaction statements given the expectations that prevail in the three-sector economy.

Discussion

This study proposes a methodology for examining institutions and institutional boundaries based on the work of Crawford and Ostrom (1995). We applied the methodology to obtain a view from ordinary people about what they see as the boundaries and contours of the third sector. Overall, we find that the methodology has promise for application in the empirical, quantitative study of institutions. Our research does in fact support the idea of a third sector, one that maps to general identification as a space for nonprofit and charitable work. This is remarkable given that the coders in our study had to explicitly view such interactions as something fundamentally different from government and business activity.

It is also clear from our research that everyday people view the nonprofit sector somewhat differently from scholars who study the sector. For example, the participant-centric view, which would suggest that nonprofit or charitable action is identifiable primarily by the type of organization engaging in the action, does not correspond well to our research participants’ perceptions of the nonprofit sector, except when government is included as the instigator of the activity. The action-centric perspective describes our participants’ responses quite well, however, as does—perhaps to a lesser degree—the intention-centric perspective.

There are several implications of understanding the everyday view of nonprofits. Nonprofit practitioners operate in a social environment, and their ability to work effectively alongside different constituencies is no doubt shaped by their understanding of those constituencies. Nonspecialist constituencies—including beneficiaries, donors, and other members of the communities in which nonprofits work—may have a different understanding or expectation of nonprofit status than practitioners do (or others who have a technical-legal way of conceptualizing nonprofit). These understandings likely shape the expectations that people have of nonprofits, what they see as the proper role of nonprofits, how they believe nonprofits should interact with other entities (such as business and governments), and the value they place on nonprofit action.

Everyday understandings inform not only the lived experience of those who work in nonprofits and are served by them, but they are also an important part of the sector’s institutionalization in modern society. How deeply an institution is embedded in society is, after all, partly a question of how everyday people understand it (Scott, 2014). If specialists understand nonprofit to mean clearly distinguishable entities with a unique role to play, while nonspecialists view nonprofit as a fuzzy category that can be applied broadly to a wide range of actions, then that difference might suggest that nonprofit is not as deeply institutionalized as specialists think.

These findings about the flexible nature of the everyday meaning of nonprofit resonate with research that suggests the benefits of replacing our use of nonprofit with nonprofitness (Knutsen & Brock, 2014; Robichau et al., 2015). We thus offer suggestive empirical evidence that accords with others’ notions that adding nonprofitness to our vocabulary solves some of the problems that sector-based, categorical thinking has introduced. Expanding the purview of nonprofitness to any type of organization, rather than to nonprofit organizations only, could permit describing the organizational landscape in new ways. This approach that seems especially warranted given the state of our so-called sector-bending economy (Dees & Anderson, 2003). Thus, moving past efforts to create static organizational classifications (Salamon et al., 2004; United Nations, 2003), scholars could consider a wide range of organizational phenomena—such as social enterprise and social entrepreneurship, organizational hybridity, and corporate social responsibility—that do not fit neatly into rigid sector frameworks. Using the methodology we have developed, it is possible not only to articulate the potential institutional boundaries of nonprofitness and also to create an intentional and formalized language around nonprofitness and, eventually, the nuances within nonprofitness.

Our findings have implications for practice as well. Many people engage in charitable and donative activity that operates outside of formal corporate structures. Scholars have lamented that such activity is difficult to observe and classify (Smith, 1997). This also means that these less corporate efforts likely have decreased access to private donations and volunteers. Such activities may include informal volunteering and civic participation, group social activity and community building, and more formal efforts such as social justice movements, native organizing, and religious practice (C. E. Mitchell, 2016; Putnam, 2000). The methodology outlined in our article suggests an approach that could help identify and classify such activities in a way that can inform both policy and practice. Even nonprofit-type activities that are not part of a formal 501(c) corporate structure could be recognized through a framework for legitimizing the “nonprofitness” of such activities, promoting increased access to private donations and other support.

Limitations

Our specific methodological design choices shaped this study and its results in ways that may limit any claim that the work is complete. Although we believe the overall structure of the methodology is both sound and instructive, it is clear that the method is sensitive to specific design choices. Indeed, we hope that other scholars identify potential improvements to our method and build on the work we have presented here.

First, we necessarily limited the scope of analysis by providing only two to four variations per grammatical category (actor, action, recipient, and aim). Certainly, there are other possible variations to explore, which could shed more light on how lay observers think about what constitutes nonprofit or charitable activity. Moreover, our choice to start with the institutional grammar necessarily framed the analysis. A complementary study could usefully examine meanings of nonprofit in a more inductive, open-ended way to see how those results compare with the ones we report here.

Second, word choices likely affected our results in ways that cannot be fully anticipated. For example, in place of our skeletal interaction statements, we could have used vignettes with varying levels of illustrative detail. Future research could explore how specific word choices shape the results. How might the results differ if the actor in each statement was merely an “organization” rather than a nonprofit, business, or government agency? We acknowledge that words carry significant meaning and that our word choices shaped the results of our study.

Third, our coding options may also have affected our results. Although 25% of all statements in our study began with the phrase, “A nonprofit organization. . .,” only 16% of the statements were coded as “nonprofit/charity.” Our interpretation is that despite respondents specifically being told a nonprofit was engaged in some action, they still considered the action to fit within either a two-sector model (subsumed under either business or government or a mix of these) or in a fourth sector (such as, perhaps, social enterprise). But we acknowledge that an alternative possible explanation exists—that because of the business/government binary presented in the first round of coding, respondents categorized interactions differently than they would have if all three-sector options were present in the first round of coding. Future research could explore this possibility.

The words used to describe the “actor” and the words used in the respondents’ choices are the same. Yet, what participants were asked to describe was the interaction. One possible remedy is dropping the “actor” altogether or calling it something neutral like “organization,” putting the emphasis on the action. Another alternative would be giving respondents answer options that characterize the exchange they are observing (e.g., donating, selling, redistributing) rather than using the type of sector classifications we used.

Finally, further research will need to establish how well these findings represent the larger population and what nuances may exist in coding carried out by different subpopulations or stakeholder groups. Even within the everyday meaning of nonprofit, there could be multiple meanings. Future work could include comparisons between different respondent groups, including lawyers, the media, nonprofit practitioners, nonprofit scholars, or students of nonprofit management.

Directions for Future Research

The institutional analysis methodology used in this study shows promise for the continued exploration of the multifaceted nature of the nonprofit sector. It can be used to explore ways that different groups of people view the sector, thereby creating a multidimensional view of the sector from the perspective of its many stakeholders. This type of work would be a natural extension of the work we have presented here. The methodology could also be deployed in other settings to examine, for example, the boundaries of the fourth sector or other emergent institutional phenomena.

Applications of the method could examine not only nonprofitness but also publicness and privateness, which are parallel and complementary concepts that transcend formal organizational or sectoral boundaries and are currently under study as institutional phenomena. A more generic formulation of this study using less nonprofit-oriented language could be used to examine social-institutional boundaries generally and may contribute to our study of sectors as institutions.

The substantive study of the nature, boundaries, and character of the nonprofit sector also merits additional study. As we have demonstrated, even the various scholarly views of the role of the sector—which differ from one another—do not fully comprehend the ways in which we can usefully conceive of the sector and its function. We urge continued study of the nonprofit phenomenon from multiple disciplines and perspectives.

The study of the nonprofit sector—its meaning, role, and purpose—should also continue to employ a wide variety of methods, not just the institutional analysis we have applied here. Particular effort should be made to broaden our understanding of nonprofitness into those spaces that are more institutionally visible than legally visible. For example, grassroots nonprofit and charitable activity and nonregistered (including religious) nonprofit activity should be at the forefront of scholarly attention. In addition, work on social movements and social justice, insurgencies, women’s and indigenous studies, and so on are of paramount importance. If we are to understand the social contours of nonprofit and charitable activity as a significant influence on modern society, we cannot limit our attention only to formally registered U.S.-based 501(c)3 organizations. Rather, we must think expansively about the social and institutional forces that have led to the formalization of the nonprofit sector.

Supplemental Material

sj-docx-1-nvs-10.1177_08997640221081523 – Supplemental material for The Social Meanings of the Third Sector: How Action and Purpose Shape Everyday Understandings of “Nonprofit”

Supplemental material, sj-docx-1-nvs-10.1177_08997640221081523 for The Social Meanings of the Third Sector: How Action and Purpose Shape Everyday Understandings of “Nonprofit” by Curtis Child and Eva Witesman in Nonprofit and Voluntary Sector Quarterly

Footnotes

Acknowledgements

For useful feedback on earlier drafts, we thank Eric Dahlin, John Hoffmann, Jon Jarvis, Jane Lopez, and Chris Silvia.

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Curtis Child

Eva Witesman

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Curtis Child is an associate professor of sociology at Brigham Young University. He studies nonprofit organizations, for-profit businesses, and the legal, cultural, and moral boundaries that separate the two sectors. He is interested also in economic sociology, and one of his current projects examines how friends and families negotiate relationships in the context of direct selling. In other projects, he has investigated the fair trade and socially responsible investing industries, how entrepreneurs choose between nonprofit and for-profit forms, and social movement activism in the apparel industry during the 1990s. His research can be found in Nonprofit and Voluntary Sector Quarterly, Organization Studies, Journal of Business Ethics, Social Forces, Sociological Science, American Sociological Review, and Voluntas.

Eva Witesman is an associate professor of public management at Brigham Young University. She is interested in macro-institutional arrangements and how they can facilitate good decision-making and, ultimately, the flourishing of humankind. Her specific research interests include public service values, anticorruption efforts, and the evolving institutional characteristics of the public, nonprofit, and private sectors. She is the author of “An Institutional Theory of the Nonprofit,” published in Nonprofit and Voluntary Sector Quarterly, as well as other publications in that journal, Public Administration Review, Public Administration, and Voluntas.

References

Battilana

Lee

(2014). Advancing research on hybrid organizing–insights from the study of social enterprises. The Academy of Management Annals, 8(1), 397–441.

Berinsky

A. J.

Huber

G. A.

Lenz

G. S.

(2012). Evaluating online labor markets for experimental research: Amazon.com’s Mechanical Turk. Political Analysis, 20(3), 351–368.

Bodle

(2019). What we mean when we say “we.” Medium. https://medium.com/@andybodle/what-we-mean-when-we-say-we-e65d65a5fcc7

Brandsen

Van de Donk

Putters

(2005). Griffins or chameleons? Hybridity as a permanent and inevitable characteristic of the third sector. International Journal of Public Administration, 28(9–10), 749–765.

Bromley

Meyer

J. W.

(2017). “They are all organizations”: The cultural roots of blurring between the nonprofit, business, and government sectors. Administration & Society, 49(7), 939–966.

Buhrmester

M. D.

Talaifar

Gosling

S. D.

(2018). An evaluation of Amazon’s Mechanical Turk, its rapid rise, and its effective use. Perspectives on Psychological Science, 13(2), 149–154.

Child

Witesman

E. M.

Braudt

D. B.

(2015). Sector choice: How fair trade entrepreneurs choose between nonprofit and for-profit forms. Nonprofit and Voluntary Sector Quarterly, 44(4), 832–851.

Child

Witesman

E. M.

Spencer

(2016). The blurring hypothesis reconsidered: How sector still matters to practitioners. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 27(4), 1831–1852.

Coppock

McClellan

O. A.

(2019). Validating the demographic, political, psychological, and experimental results obtained from a new source of online survey respondents. Research & Politics, 6(1), 2053168018822174.

10.

Cowan

(2004). Legal consciousness: Some observations. The Modern Law Review, 67(6), 928–958.

11.

Crawford

S. E. S.

Ostrom

(1995). A grammar of institutions. American Political Science Review, 89(3), 582–600.

12.

Deck

Jahedi

(2015). The effect of cognitive load on economic decision making: A survey and new experiments. European Economic Review, 78, 97–119.

13.

Dees

J. G.

Anderson

(2003). Sector-bending: Blurring lines between nonprofit and for-profit. Society, 40(4), 16–27.

14.

Desimone

L. M.

Le Floch

K. C.

(2004). Are we asking the right questions? Using cognitive interviews to improve surveys in education research. Educational Evaluation and Policy Analysis, 26(1), 1–22.

15.

Egleston

B. L.

Miller

S. M.

Meropol

N. J.

(2011). The impact of misclassification due to survey response fatigue on estimation and identifiability of treatment effects. Statistics in Medicine, 30(30), 3560–3572.

16.

Ewick

Silbey

S. S.

(1991). Conformity, contestation, and resistance: An account of legal consciousness. New England Law Review, 26, 731–749.

17.

Farwell

M. M.

Shier

M. L.

Handy

(2019). Explaining trust in Canadian charities: The influence of public perceptions of accountability, transparency, familiarity and institutional trust. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 30(4), 768–782.

18.

Fischer

R. L.

Wilsker

Young

D. R.

(2011). Exploring the revenue mix of nonprofit organizations: Does it relate to publicness? Nonprofit and Voluntary Sector Quarterly, 40(4), 662–681.

19.

Garfinkel

(1988). Evidence for locally produced, naturally accountable phenomena of order, logic, reason, meaning, method, etc. In and as of the essential quiddity of immortal ordinary society, (I of IV): An announcement of studies. Sociological Theory, 6(1), 103–109.

20.

Handy

Seto

Wakaruk

Mersey

Mejia

Copeland

(2010). The discerning consumer: Is nonprofit status a factor? Nonprofit and Voluntary Sector Quarterly, 39(5), 866–883.

21.

Jackson

P. M.

(2006). Nonprofit risk management & contingency planning: Done in a day strategies. John Wiley.

22.

Knutsen

W. L.

Brock

K. L.

(2014). From a closed system to an open system: A parallel critical review of the intellectual trajectories of publicness and nonprofitness. Voluntas: International Journal of Voluntary and Nonprofit Organizations, 25(5), 1113–1131.

23.

Louis

(2021). Unlocking progressive corporate governance: The black and brown HDFC key. Law Review Article. https://ssrn.com/abstract=3855558

24.

Maynard

D. W.

Clayman

S. E.

(1991). The diversity of ethnomethodology. Annual Review of Sociology, 17(1), 385–418.

25.

McGinnis

M. D.

(2011). An introduction to IAD and the language of the Ostrom workshop: A simple guide to a complex framework. Policy Studies Journal, 39(1), 169–183.

26.

Mitchell

C. E.

(2016). When worlds collide: Bringing the native/indigenous transnational nonprofit leader to the conversation (Publication Number 10128131) [Doctoral dissertation, University of San Diego]. ProQuest Dissertations Publishing.

27.

Mitchell

R. K.

Weaver

G. R.

Agle

B. R.

Bailey

A. D.

Carlson

(2016). Stakeholder agency and social welfare: Pluralism and decision making in the multi-objective corporation. Academy of Management Review, 41(2), 252–275.

28.

Moulton

Eckerd

(2012). Preserving the publicness of the nonprofit sector: Resources, roles, and public values. Nonprofit and Voluntary Sector Quarterly, 41(4), 656–685.

29.

O’Neill

(2009). Public confidence in charitable nonprofits. Nonprofit and Voluntary Sector Quarterly, 38(2), 237–269.

30.

Ostrom

(2009). Understanding institutional diversity. Princeton university press.

31.

Paolacci

Chandler

(2014). Inside the Turk: Understanding Mechanical Turk as a participant pool. Current Directions in Psychological Science, 23(3), 184–188.

32.

Permut

S. E.

(1981). Consumer perceptions of nonprofit enterprise: A comment on Hansmann. The Yale Law Journal, 90(7), 1623–1632.

33.

Putnam

(2000). Bowling alone: The collapse and revival of American Community. Simon & Schuster.

34.

Reiser

D. B.

(2010). Blended enterprise and the dual mission dilemma. Vermont Law Review, 35, 105–116.

35.

Robichau

R. W.

Fernandez

Kraeger

(2015). An integrated framework of inter-sectorality: Nonprofitness and its influence on society and public administration programs. Journal of Public Affairs Education, 21(3), 315–336.

36.

Sabeti

(2009). The emerging fourth sector. http://www.aspeninstitute.org/publications/emerging-fourth-sector-executive-summary

37.

Salamon

L. M.

(1987). Of market failure, voluntary failure, and third-party government: Toward a theory of government-nonprofit relationship in the modern welfare state. Journal of Voluntary Action Research, 16, 30–48.

38.

Salamon

L. M.

(Ed.). (2002). The tools of government: A guide to the new governance. Oxford University Press.

39.

Salamon

L. M.

Sokolowski

S. W.

List

(2004). Global civil society: An overview. In Salamon

L. M.

Sokolowski

S. W.

, & Associates (Eds.), Global civil society: Dimensions of the nonprofit sector (Vol. 2, pp. 3–60). Kumarian Press.

40.

Sánchez-Hernández

M. I.

Carvalho

Rego

Lucas

M. R.

Noronha

(2021). Social innovation and entrepreneurship in the fourth sector. In Sánchez-Hernández

M. I.

Carvalho

Rego

Lucas

M. R.

Noronha

(Eds.), Social innovation and entrepreneurship in the fourth sector: Sustainable best-practices from across the world (pp. 1–5). Springer.

41.

Sarat

(1990). The law is all over: Power, resistance and the legal consciousness of the welfare poor. Yale Journal of Law & the Humanities, 2, 343–379.

42.

Scott

W. R.

(2014). Institutions and organizations: Ideas, interests, and identities (4th ed.). SAGE.

43.

Shank

D. B.

(2016). Using crowdsourcing websites for sociological research: The case of Amazon Mechanical Turk. The American Sociologist, 47(1), 47–55.

44.

Siddiki

Weible

C. M.

Basurto

Calanni

(2011). Dissecting policy designs: An application of the institutional grammar tool. Policy Studies Journal, 39(1), 79–103.

45.

Silbey

S. S.

(2005). After legal consciousness. Annual Review of Law and Social Science, 1, 323–368.

46.

Skelcher

Smith

S. R.

(2015). Theorizing hybridity: Institutional logics, complex organizations, and actor identities: The case of nonprofits. Public Administration, 93(2), 433–448.

47.

Smith

D. H.

(1997). The rest of the nonprofit sector: Grassroots associations as the dark matter ignored in prevailing “flat earth” maps of the sector. Nonprofit and Voluntary Sector Quarterly, 26(2), 114–131.

48.

Steinberg

(2006). Economic theories of nonprofit organizations. In Powell

W. W.

Steinberg

(Eds.), The nonprofit sector: A research handbook (2nd ed., pp. 117–139). Yale University Press.

49.

United Nations. (2003). Handbook on non-profit institutions in the system of national accounts (Handbook of National Accounting, Issue Series F., No. 91). https://unstats.un.org/unsd/publication/seriesf/seriesf_91e.pdf

50.

Witesman

E. M.

(2016). An institutional theory of the nonprofit: Toll goods and voluntary action. Nonprofit and Voluntary Sector Quarterly, 45(4), 97S–115S.

51.

Young

D. R.

(1983). If not for profit, for what? A behavioral theory of the nonprofit sector based on entrepreneurship. Lexington Books.

52.

Zolin

Kropp

(2008). Public or private by genetics or design? A case study of organizational design decisions for new public ventures. International Public Management Review, 9(1), 130–147.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB