Introduction

This paper presents a novel methodological approach for studying resource stratification in civil society: the multi-dimensional measure of resource stratification in civil society (MMRSC). The sectorial boundaries that are widely used today aim at distinguishing different types of organisations populating different sectors: the private for-profit sector dominated by the business firms, the public sector dominated by public authorities and the civil society sector (or the third sector) dominated by associations and organisations without profit motive (see Ahrne 1996; Billis 2010; Seibel 2015). In contrast to the business and political spheres where concentration of resources and power in a few hands have extensively been studied in the tradition of elite research (for a review over the field see Khan 2012), civil society has been widely understood as a societal sphere representing the plurality of voices, groups, and interests within a liberal-democratic political system, carrying visions from below and expected to counteract the tendency of power concentration rather than to contribute to it (della Porta 2020; Diamond 1994; Keane 2009). This has meant that relatively little attention has been paid to how resources are unevenly distributed and concentrated within the civil society sector (Johansson and Uhlin 2020). At the same time, studies of civil society sector do shed lights on the different organisational resources, capacities, and conditions with which civil society organisations (CSOs) operate (Scaramuzzino and Scaramuzzino 2017), and compete over (e.g., Johansson and Kalm 2015).

We use stratification as the sociological concept of ‘social stratification’ generally defining an unequal distribution of resources and power in modern societies (see for instance Max Weber’s theory of distribution of power in Waters et al. 2010). In other words, our approach is relevant for and can be adopted in studies of CSOs concerned with the structural conditions of the sector, in terms of how different types of resources are (unevenly) distributed among organisations.

The approach is unique in its multi-dimensional understanding of organisational resources and is based on six indicators consisting of economic, political, and media resources. It is designed to be applied both for single case studies where the case is a particular civil society context (e.g., a national context) and for cross-case comparisons (e.g., comparing the civil society sector in different countries). The method allows context-sensitive operationalisation of the indicators, balancing comparability across contexts and the possibility for adaptation to specific civil society sectors.

The method was developed from extensive empirical work which aimed at mapping resource-rich national organisations in civil society in four different European countries (Italy, Poland, Sweden and UK) and at the EU level. This research endeavour was carried out in a research programme about ‘civil society elites’ and more specifically in a study about the composition of such elites. The approach we propose is a development and adaptation of the positional method within the elite research tradition, which is based on identifying a limited number of individuals holding central positions in organisations that are considered powerful (Hoffmann-Lange 2018). Accordingly, in our own study about civil society elites, mapping organisations and characterising them as resource-rich is a precondition for identifying positions of power in the civil society sector.

The methodological approach MMRSC itself, however, has a value beyond the mere identification of resource-rich organisations and elite positions. Therefore, in this paper, we argue for its possible application as a multi-dimensional measure of resource stratification in civil society sector. As will be discussed further the method enables (1) a multi-dimensional understanding of resources in civil society, (2) characterising and mapping organisations based on their access to different types of resources, and (3) comparing resource stratification across different civil society contexts.

The paper is structured in the following way: first, we present the different strands of research that the paper draws on and identify the gap that our approach aims to fill. Second, we present our approach both in general terms and with its concrete application in specific national contexts in Europe. Finally, we discuss the possible applications of the approach as well as its strengths and limitations.

Why we need a multi-dimensional measure of resource stratification in civil society (MMRSC)

Civil society as a neglected field in elite research

Elite studies have mostly neglected civil society as a social sphere where elites, understood as individuals controlling disproportionate amounts of resources, might arise or be reproduced. This is due to CSOs being perceived as not ‘powerful’ enough and too dependent on the support of other organisations and their members (Hartmann 2015). Most previous elite studies have hence not included leaders of CSOs in their definition of elites.

Elected parliamentary politicians and board members of the largest enterprises are clear and straightforward examples of elite populations within political and business spheres. When it comes to the civil society sphere, however, it is less straightforward to talk of a given organisational population whose leaders can be considered as elite(s). The vast diversity in organisational forms (e.g., associations, social enterprises, and foundation) and missions (e.g., voice- and service-oriented organisations) that are present in civil society allows a large variety of resources to be relevant for different organisations (budget, staff, volunteers, reputation, followers, donors etc.). In addition, the segmentation of civil society actors into numerous issue areas makes it difficult to identify a clear hierarchical structure in the civil society sphere.

A few elite studies have included civil society as a separate sphere comparable to the political, business, media, and cultural spheres (e.g., in Göransson 2007 called ‘organisational elite’; Ruostetsaari 2015). These studies tend to have however a narrow and simplified understanding of resources based on the membership size of the organisations they lead (e.g. Göransson 2007). The focus on membership tends, however, to exclude other organisational forms than associations which are not membership based, for instance foundations. Membership is also often associated with interest representation and tend to exclude other goals and missions, for instance service production, for which a large membership base is less important.

Measuring resources in civil society and mapping the civil society sector

Studies of CSOs, non-profit organisations, non-governmental organisations, interest groups, and social movements have employed a variety of methods in their attempt to measure different types of resources. Mapping and exploring the composition of this sector is indeed a major tradition in this research field. The most notable example of the endeavour of mapping CSOs is the ‘Johns Hopkins Comparative Non-profit Sector Project’ (see Salamon et al. 2004). The studies within this project have tended to include a large variety of organisational forms (e.g., associations, social enterprises, and foundations) and missions (e.g., voice- and service-oriented organisations) aimed at defining, mapping, and measuring civil society across different national contexts. Resources have been addressed in terms of both numbers of volunteers, numbers of employees, and different forms of funding (ibid.). These attempts for mapping of the civil society sector have, however, neglected the issue of resource stratification within civil society, as the unit of analysis has been the civil society sector rather than the organisations that populate it. Here, we find both different types of resources that are in focus (budget, staff, volunteers, reputation, followers, donors, access to decision-making, etc.) and different methods of data collection (registry data, interviews, and surveys). These studies are very useful in understanding different forms of resources in civil society. They do not, however, aim to map resource-based stratification and rather consider resources as independent variables that can explain other phenomena, such as advocacy strategies, mission drift, and Europeanisation (e.g., Child and Grønbjerg 2007; Klüver 2010).

Interest group studies focus on interest group organisations and their influence on policy making in terms of different advocacy strategies and actual impact (Beyers 2004; Binderkrantz and Rasmussen 2015). Here, we also find notable example of research mapping the national populations of organised interest such as the Comparative Interest Group-survey project (CIG-survey) (Beyers et al. 2020). This strand of research has mostly a state-centred understanding of power, meaning that the resources considered are in general those that allow influence on policy-making processes. Social movement studies focus on social movements and the organisations that they spawn and often aim to understand and explain mobilisation and success (e.g., Marks and McAdam 2009; McCarthy and Zald 1977). Although the political opportunity structure approach (Kitschelt 1986; Kriesi 2007) includes variables related to the state, much of this research has a society-centred, relational understanding of success as in terms of centrality in networks, visibility in public spaces, etc. (e.g., Diani 2015).

Contribution of the MMRSC approach

The contribution of the MMRSC approach, which will be illustrated in the remaining of the paper, is threefold. Firstly, we propose a multi-dimensional understanding of resources in civil society. As previously discussed, there has been a lack of understanding for the diversity of organisational forms and types of resources that are relevant for defining resources in the civil society sector. This applies both to elite studies and studies of civil society. The approach we are proposing allows a multi-dimensional understanding of resources in civil society in terms of economic, political and media resources. Although all three types of resources have been discussed in studies about CSOs, interest groups and social movement, this is the first attempt of a systematic analysis of all three types of resources at the same time.

Secondly, we propose an approach that allows researchers to characterise and map CSOs based on their access to different types of resources. This is essential for studies of CSOs, interest groups and social movements which are often interested in understanding and explaining organisational behaviour based on access to different types of resources. By acknowledging the multi-dimensionality of resources and providing operationalisation using multiple indicators, the method can provide valuable tools for studies addressing the relationships between specific types of resources and organisational behaviours or outcomes (e.g., advocacy strategies, political influence, societal impact, or trust among the citizens). The approach rests on indicators that can be operationalised in different contexts and across different segments of civil society and types of organisations, allowing comparisons between organisations and across different contexts.

Thirdly, we propose an approach that can guide mapping of resource distribution, concentration, and stratification in the civil society sector. Instead of relying solely on economic resources—the most easily and frequently employed proxy measure of organisational resources—the approach allows us to understand how the distribution of multiple types of resources among CSOs leads to different patterns of stratification. This means that the method paves the way to comparative studies of the power structure of civil society across different contexts in more fine-grained and relevant ways.

To make the approach applicable in a systematic way, we present it in a step-by-step structure. The approach allows researchers to make informed decisions, guided by research questions at hand, about how to draw boundaries of the field, how to choose relevant resources to be considered and how to set the criteria for inclusion of different organisations in a given study.

In the following sections, we present MMRSC approach in general, while illustrating its application by providing examples of the ways in which it was employed in our specific study of civil society elites at national and European level.

Six indicators of organisational resources

Based on previous research, we propose a set of indicators that measure different types of resources in civil society. The approach is underpinned by an understanding of civil society as a sphere in which resources, understood in a broad sense as economic, political, and media resources, are unequally distributed. This unequal distribution of resources produces a stratification of organisations between the haves and have-nots.

Our indicators include six types of resources in civil society at the intersection of two dimensions. One dimension distinguishes resource types internal to civil society from resources external to civil society. Resources can be ‘internal’ to civil society in the sense that they are either disposed at the organisational level or relevant within/among the civil society actors. Resources can be ‘external’ to civil society in the sense that they are indicators of recognition and status from the external actors’ point of view, outside of civil society. The other dimension separates the three qualitatively different types of resources in civil society: economic resources, political resources and media resources:

  • Economic resources: Members, staff, volunteers, budget (internal)/ external funding (external)

  • Political resources: Umbrella organisations and networks and their members (internal)/ Posts in public committees and public consultation (external)

  • Media resources: Followers, likes, group members on social media (internal)/ Traditional media coverage (external)

Economic resources

As economic resources internal to civil society, we consider internal resources that give organisations the capacity to act. Theories of resource mobilisation (see McCarthy and Zald 1977) emphasise the organisational dimensions of social movements and the importance that resources play for collective actors. Large membership basis is considered a resource that can give CSOs representativeness vis-á-vis public authorities (e.g., Dür and Mateo 2012; Scaramuzzino and Wennerhag 2019), while sizeable staff enables the organisation the capacity to engage in many advocacy activities that require specific expertise (Scaramuzzino and Scaramuzzino 2017). For many CSOs, volunteers are also a key resource, frequently utilised as non-paid staff (Salamon et al. 2004). Financial resources in terms of budgets are of course relevant for organisations’ operation.

Concerning economic resources external to civil societies, we consider external funding. Theories of resource dependency (Pfeffer and Salancik 1978) prescribe that CSOs’ strategies are shaped by their dependency on external actors for funding. In general, we argue that public core funding, including funding from international or European institutions, can be seen as a token of privilege and embeddedness in the system (Beyers and Kerremans 2007; Hedling and Meeuwisse 2019). Yet, in some contexts, funding sources from private donations and sponsoring might also be relevant and represent other forms of incumbent status.

Political resources

Our understanding of political resources draws on theories of political opportunity structures (Kitschelt 1986). The extent to which CSOs have access to institutionalised structures where they can influence political decisions can be seen as a resource, as the access can be interpreted as recognition (Hedling and Meeuwisse 2019), giving them the opportunity to influence decision-making (Scaramuzzino and Wennerhag 2019).

When it comes to political resources internal to civil society, we consider being member of umbrella organisations (i.e., federations) or meta-organisations (Ahrne and Brunsson 2005) that aim to represent the sector vis-á-vis the state within specific policy areas (e.g., Federations of disability organisations) or at a more general level (e.g. Federations of CSOs representing the sector). In a multi-level system of governance, these federations are often present at national levels to represent parts of the civil society on different matters (cf. Ahrne and Brunsson 2005; Scaramuzzino and Wennerhag 2019) and pursue influence at the national level vis-a-vis the state.

In terms of political resources external to civil society, we consider having access to policy-making processes through systems of interest representation. These are often arranged by public authorities to allow representation of collective interests in policy-making and to legitimise their policies (Casey 2004; Klüver 2010). These resources relate to what has often been described as inside advocacy strategies (Beyers 2008). When it comes to political resources, we acknowledge, as previously discussed, that these might be bound to specific policy areas as both public governance and civil society mobilisation tend to be structured following the boundaries of specific issues, e.g., disabilities, environment, and gender equality. Many countries have also seen the rise of governance structures that are afferent to the whole civil society sector and attempt to regulate state-civil society relations, with national compacts as the clearest example (Johansson et al. 2011; Reuter et al. 2012).

Media resources

The media resources internal to civil society can be related to what has been called the ‘logic of presence’ and digitalisation (Johansson and Scaramuzzino 2019). The role of internet and social media for CSOs has been highlighted in different ways. Through Internet and social media, CSOs can gain visibility, engage in advocacy activities, mobilise people around specific issues, conduct campaigns, organise volunteers and much more (Enjolras et al. 2012; Scaramuzzino and Scaramuzzino 2017). The capacity to become visible is, of course, among other things dependent on having many followers, group members and likes on the social media channels that are able to share, post, and retweet the message of the organisation (Johansson and Scaramuzzino 2019), all of which we argue to be considered as ‘internal’ in the sense that they are disposed of at the organisational level.

The media resources external to civil society relate to the more traditional media landscape over which CSOs have much less control. Social movement theory has highlighted how dependent many movements are on being visible in and having access to media (cf. Benford and Snow 2000). Being present and having events covered by media becomes very important for getting the message out and to be able to influence public opinion and potentially decision-makers. In interest group research, these media-based strategies have been labelled as outside advocacy strategies (Beyers 2008) although the interplay between lobbying and media strategies has also been explored (Trapp and Laursen 2017).

It is important to note that the distinction between economic, political and media resources as well as between internal and external resources is theoretical. Empirically speaking these resources might be even strongly correlated. An organisation might for instance have a large staff in terms of internal economic resources thanks to public core funding which provides a stable budget as external economic resource, for instance. The point of distinguishing the types of resources that can be considered internal to civil society and external is in other words to capture diverse ways in which CSOs can become resourceful.

The indicators allow us, in fact, to measure access to specific types of resources that could be strategically used by the organisations to gain influence. None of them gives the organisations automatically any power or influence. Most types of resources could even be argued to be attached to specific mechanisms that could inhibit the organisations capacity to wield power or influence. One of the clearest examples is the risk for co-optation when CSOs, due to power asymmetries in relation to public authorities, tend to adapt their goals to be more in line with their counterparts (Najam 2000). This mechanism could be relevant concerning both political and economic resources as organisations might trade access to policy-making and/or public funding with being more compliant with public authorities (cf. Mosley 2012; Verschuere and De Corte 2015; Scaramuzzino and Scaramuzzino 2017).

In the case of followers on social media as an indicator of internal media resources, it captures CSOs’ interaction with adherents in a semi-public space. The fact that the number of followers on social media is potentially vulnerable to manipulation by the companies running the media platforms, should of course be critically taken into account. Even when it comes to traditional media exposure as an external media resource, quantitative measures should be carefully considered considering the possibility that not all media coverage can be considered as a resource for CSOs.

The fact that access to resources might have ambiguous impact on CSOs’ strategic action does not however necessarily mean that their capacity to influence is compromised, they might in fact also function as incentives to act (e.g., Child and Grønbjerg 2007; Neumayr et al. 2015). It is also reasonable to assume that organisations that tend to accumulate different types of resources and hence diversify their resource dependency have an advantage compared to organisation that only rely on one or two types.

Based on these indicators we argue that it is possible to characterise organisations by operationalising relevant indicators of resources. The four steps for employing MMRSC approach are as following:

  • Step 1: setting the boundaries of the civil society sector to be studied

  • Step 2: choosing the indicators considering the relevance of resource types

  • Step 3: operationalising the indicators

  • Step 4: measuring resource stratification

Step 1: drawing boundaries

As in all attempts to map empirical phenomena, the issue of drawing the boundaries of a given research object is essential. The first and most obvious boundary to be drawn in a study of resource stratification in civil society is the one between civil society and other spheres of society. The internationally recognised definition of CSOs (Salamon et al. 2004) is based on five criteria. They should be organised, separated from the state, not having profit-distribution as their primary purpose, self-governing, and voluntary. This definition clearly defines civil society, both in relation to the state and the market, as a separate sphere or sector but also as a sphere populated by organisations (cf. Ahrne 1996).

Even though this definition is well established, it still needs to be operationalised in specific contexts and for specific organisations. For some types of organisations, for instance associations of individuals with not-for-profit purposes, their inclusion might not be problematic. However, concerning other organisations such as trade unions, business organisations, associations of municipalities and regions, and political parties, it is still debatable to what extent they should be considered part of the civil society sector. The associational form would clearly place these organisations in the civil society sector although the interests they represent would for some of these organisations be of the business sector (e.g., for employers’ organisations) or of the public sector (e.g., for associations of municipalities). In a few previous studies of political elites, however, we find some categories of organisations that could be included in a broad definition of CSOs (e.g., Salamon et al. 2004), such as political parties, major interest organisations, professional associations, trade unions, and religious institutions (e.g., Best and Higley 2018). These leaders are, however, understood as being part of the political elite in the mentioned studies, based on a definition of power related to the political sphere.

The choice as to where to draw the boundary in this regard should ultimately be guided by a substantive research problem. In our study of civil society elites, the decision to rather strictly distinguish the civil society sector from the market and public sectors has been guided by the need to exclude the economic and political elites for analytical reasons. From an elite research perspective, it is about drawing the horizontal boundaries of the elites (e.g., Hoffmann-Lange 2018), i.e., the boundaries that separate different types of elites, for instance the economic elite from the political elite.

In our definition of CSOs, we have hence excluded the following types of organisations: political parties, employers’ organisations, and organisations representing industrial sectors dominated by public organisations or private for profit organisations. In our understanding, the leaders of these organisations would rather be considered as parts of the business or the political elites. One exception has been organisations representing employers and producers in the civil society sector (e.g., welfare service producers, cooperatives, schools) which have been included in our study as these interest organisations are representing the interests of the civil society sector with service provision function.

When it comes to other interest organisations such as trade unions, organisations representing professions (e.g., lawyers and doctors) we have chosen to exclude them from our definition. In a classic distinction of interest organisations between ‘producers’ (employers’ and workers’ organisations) and ‘users’, (e.g., pensioners’ and disability organisations) (see Beyers 2004) we have, as often in studies of civil society, excluded the first category and included the second.

Another way of drawing the boundaries can be based on geographic or administrative levels. In many countries, the civil society sector has tended to organise hierarchically following the state’s governance structure (Einarsson 2012; Skocpol 2003). CSOs hence tend to be organised, for instance, locally, regionally, nationally, or at a European or international level (Johansson et al. 2018). Often these structures are based on so-called meta-organisations (Ahrne and Brunsson 2005), i.e., associations of organisations. Our mapping of resource-rich CSOs at the national level in Italy, Poland, Sweden, and UK and also at a European level has mostly led us to meta-organisations, i.e., federations of member-CSOs often themselves organised at a lower geographic or administrative level.

Boundaries can also be drawn based on policy areas or the fields of activity of the CSOs. The International Classification of Non-profit Organisations (ICNPO) offers an example of how CSOs can be divided into different areas of activities (Salamon et al. 2004) and can be used to draw the boundaries for mapping. In our study, we have included CSOs involved in at least one of the following policy areas: Age, Charity, Culture, Disability, Environment, Gender equality, Human Rights and Democracy, Migration and ethnic groups, Religion, Sports and leisure. Given the diversity in the type of CSOs and their sheer number, we deemed choosing a number of specific policy areas as a necessary step to be able to map the resource-rich CSOs.

Drawing these boundaries is a rather difficult task, especially in the civil society sector, which entails a great diversity of organisational forms (e.g., associations and foundations) and aims (e.g., advocacy, service production and community building). Our choices of inclusion and exclusion criteria have been guided by our interest in civil society elites and have provided us with a population of CSOs including many different types such as solidarity organisations, sports associations, cultural organisations, environmental organisations, international aid organisations, users’ and patient organisations, and organisations representing non-profit welfare service providers. Other research questions would have motivated drawing the boundaries in different ways, with more expansive or a narrower definition of CSOs.

While drawing the boundaries of civil society is a task that any study of the civil society sector has to deal with, our contribution lies in developing a method for mapping CSOs based on a set of different types of resources, allowing to study resource stratification. In our method, drawing boundaries is not merely a necessary task for making the mapping of CSOs in a given context manageable. The decision-making process in drawing the boundaries also has consequences for how to operationalise the six resource indicators that we have identified. For instance, many external resources, in terms of funding, access to policy processes, and to media, are bound to specific policy areas. The civil society sector in many countries is organised also according to different policy areas, issues or interests to be represented, which becomes relevant when we discuss resources that are internal to civil society. Once again, a given substantive research question will guide the decisions about setting these boundaries.

Step 2: choosing the indicators

Our approach allows adaptation of the indicators proposed in the previous section for a specific purpose of a given study and for a specific context. In our study of civil society elites, we acknowledge that there are different segments of policy areas in which CSOs’ political resources are of importance. Internally to civil society, these resources are bound to umbrella organisations that aim to coordinate the interests of its members and speak for them. Externally to civil society, these resources are bound to arenas for participation in policy-making processes that are linked to a specific policy area. Furthermore, some umbrella organisations aim at speaking for the whole civil society sector and specific arenas for participation are addressing civil society-state relations. We have hence chosen to operationalise the indicators related to specific policy areas and those related to state-civil society relations (i.e., the civil society policy area) separately. It means that political resources within specific policy areas can be used to gain influence within that specific policy area, while political resources within the civil society policy area can be used to gain influence on the way in which civil society-state relation are regulated. We have thus used two indicators of economic resources and four indicators of political resources.

We have consciously left out the media resources as our study of civil society elites was underpinned by an understanding of CSOs’ embeddedness as proximity to the state (see Hedling and Meeuwisse 2019). This choice has entailed excluding from our sample less formalised organisations that might not have employed staff, membership in umbrella organisations, external funding or access to policy-making but still be considered influential, for example due to visibility in the public debate through their media presence. Given the importance of media’s role in the work of CSOs today (Scaramuzzino and Scaramuzzino 2017), this third type of resources in our approach might be relevant for many other studies of CSOs.

Step 3: operationalising the indicators

Once relevant indicators are selected, they need to be operationalised. The operationalisation of each indicator has been guided by relevant literature on the civil society sector in each context, aided by consulting country experts and organisations representing the civil society sector. The following resources (see Table 1) we have deemed as relevant for measuring each type of resource and hence operationalising each indicator in the contexts studied:

Table 1 Resources chosen for each indicator
Full size table

These indicators are used, in civil society research, in studies of interest groups and social movements as characteristics of specific organisations and actors. The number of employees, the size of the budget, the amount of public funding can be used as measures of resources and possibly independent variables in the analysis of organisational behaviours or outcomes as a dependent variable.

Our operationalisation of the indicators is instead inspired by a set-theoretical perspective. When we refer to concepts as “sets” we imply that there are boundaries that delimit what is included and what is excluded. Cases might fit within the boundaries of the set and hence have membership in it or not (Schneider and Wagemann 2012). Hence, from a set-theoretical perspective, operationalising indicators entails setting boundaries of inclusion and exclusion. For addressing resource concentration, it means that we operationalise each indicator to define whether a particular organisation can be described as having access to disproportionate amount of that particular resource and hence being included in the set of resource-rich organisations or not. Here (see Table 2) we present how the indicators introduced above can be operationalised with certain variations across countries. For each indicator we have defined specific thresholds, e.g. cut-off points for scale variables, that distinguish the haves and the haves-not as two different sets of cases/organisations.Footnote 1

Table 2 Operationalisation of the indicators of economic and political resources in our study
Full size table

As the aim of our study of civil society elites was to compare elites across countries, we have balanced the need for comparability (operationalising indicators as similarly as possible across contexts) and for context-sensitiveness (operationalising indicators in a way that is appropriate for each context). Availability of registry data for resources has also informed our choices in the operationalisation of the indicators.

In our study, the economic resources have been operationalised in terms of employees (internal) and public funding (external). The political resources have been operationalised in terms of participation/membership in umbrella organisations (internal) and in committees/consultation (external). The thresholds, for instance as for how many employees CSOs need to have to be included in our sample, have been chosen to highlight a ‘disproportionate amount’ of a specific type of resource (cf. Khan 2012), taking into consideration the level of professionalisation of the civil society sector in the specific context. Media resources should be operationalised based on context-specific knowledge of what it means to have a disproportionate access to for instance social media channels or traditional media. Setting a threshold for number of followers on Twitter or number of mentions in newspapers would be a viable solution. Analysis of media resources can generate different answers to the question of which organisations have the most resources in civil society and allow including less formalised actors such as social movements, more fluid and contingent forms of networks, and influencers (Santilli and Scaramuzzino 2024).

In the case of our study of civil society elites, we were interested in identifying a set of resource-rich CSOs whose leaders could be considered as part of a civil society elite in accordance with the positional method of elite identification. This aim resounds with an interest in mapping which organisations can be considered being the resource-rich and possibly most powerful in a specific field. The mapping presupposes that we, as researchers, do not a have a before-hand available list of relevant organisations and that we are interested in excluding smaller organisations that do not have access to the resources or that control a very small amount of it. Basically, we were interested in the ‘haves’ rather than the ‘have-nots’.

In mapping a population of resource-rich organisations in each national context, using the indicators introduced in the previous section, we have adopted an ‘inclusive’ approach. We aimed at identifying a broad population of organisations by including all of the organisations that have entered the sample by fulfilling at least one of the conditions for inclusion that stems from each indicator specified in Table 2. A more exclusive approach could include only the CSOs that fulfil at least two or even three of the conditions (Santilli and Scaramuzzino 2024). From the point of view of our aim of studying civil society elites, it made sense to choose an inclusive approach as the correlation between organisational resources and the elite status of the leaders could not be taken for granted. We did not want to risk excluding relevant organisations whose leaders might be considered an elite only because they did not fulfil more than one indicator. Furthermore, some of the national contexts we studied presented a weaker stratification of resources, with very few organisations fulfilling more than one indicator, which would have produced a very small sample of CSOs for these countries.

Step 4: measuring resource stratification

While operationalisation of the conditions of inclusion allows us to identify for each indicator which CSOs can be considered as resource-rich (e.g., as those that have at least 50 employees) and which not, compiling the lists of organisations for different indicators provides us with the opportunity to measure resource accumulation by specific organisations and stratification within a given civil society context. By accumulation we mean CSOs that tend to amass disproportionate amount of resources for more than one type of resource. By resource stratification, we mean the hierarchical structure that emerges within civil society due to accumulation of multiple types of resources.

In all five contexts in which we have applied the approach, we find certain overlaps between the populations of CSOs that are identified as resource-rich with each of the indicators. It means that some CSOs are identified as resource-rich organisations based on more than one indicator. The number of indicators by which a given CSO is identified can be viewed as a measure of the accumulation of different types of resources that the organisation controls. We call this measure the ‘elite score’. This elite score can range between zero and the total number of indicators that are considered and operationalised in each context. If an organisation entered our population of the most resourceful CSOs by fulfilling a condition for inclusion for one indicator, the organisation received a score of 1. If an organisation entered our population by fulfilling the conditions for inclusion for two indicators, the organisation received a score of 2, and so on. In other words, it is an unweighted additive index where each indicator contributes equally to the total elite score a given organisation is assigned with. As presented in Table 2, the conditions for inclusion for our indicators are operationalised in different ways, depending on the data availability as well as on the specific characteristics of different indicators. The elite score thus allows us not only to map out CSOs but also to discern the patterns of resource concentration and stratification. In our case, as we were interested in the civil society elite as leaders of resource-rich organisations, we did not include CSOs that would have scored zero.

In following, we discuss other ways in which the Multi-dimensional Measure of Resource Stratification in Civil society (MMRSC) can be used in empirical studies of the civil society sector as well as some of the results of our study of civil society elites.

Possible applications of the MMRSC approach

One possible application in studies of civil society is to characterise organisations based on their access to different types of resources. It could be interesting to compare the relevance of different types of resources for organisational behaviour. By summing the indicators in an elite score, as we have done, the measure can function as an index that allows researchers to address resource accumulation by specific organisations. In this case, it might be interesting to also include organisations that do not control any type of resource to a large extent. These organisations would score “zero” on our elite score. The elite score can be then used as an independent variable in any study that aims at explaining organisational behaviour or outcome (e.g., advocacy strategies and influence) by means of access to different types of resources. Our civil society elite survey study, for instance see Lee and Scaramuzzino (2024), shows that leaders of CSOs with the elite score higher than three have access to larger social networks and to more arenas for influence than the leaders of organisations with elite score two and one.Footnote 2

The set-theoretical perspective used in the construction of elite score hence allows to create an index-like measure of resource accumulation, counting for the number of resource types that organisations control disproportionately (i.e., above the threshold). Of course, this dichotomous approach on the haves and the have-nots comes with the drawback, in that some nuances can become lost when it comes to scale-based variables (e.g., staff and budget).

In civil society studies, our approach can also serve as a crucial step in studying resource concentration and stratification in the civil society sector. Studies might focus on one particular context or aim at comparative analyses of structures of resource stratification in different contexts. The latter could help us understand different patterns of stratification in relation to different historical, political, and cultural development. For instance, one could explore the role of different civil society regimes or systems of interest representation in structuring resource stratification in the civil society sector.

The following table (Table 3) shows the resource stratification for each of the contexts in our study:

Table 3 Distribution of CSOs across elite score in each national context
Full size table

The distribution of organisations across the elite score indicates patterns of stratification in all of the five contexts forming a pyramid structure, with a few CSOs in each context with higher scores and the majority of the CSOs with lower scores. In some contexts, however, we observe a more clearly hierarchical structure (Italy, Sweden and the EU), compared to the others (Poland and UK). Other ways of addressing accumulation of different types of resources might also be used, for instance assigning different weights to the indicators, based on theoretically informed assessments of their relative importance in specific contexts. Researchers could also probe possible ways in which different types of resources are positively or negatively correlated with each other addressing which types of resource accumulation have a cumulative effect and which types of research show more of a trade-off effect. The same would be relevant for assessing the relevance of different resources that would fit within the same type, such as budget and employed staff. If strongly correlated it would be relevant to include only one of them. Otherwise, a combination would be to prefer.

Systematic mapping of resource-rich organisations also opens up a possibility to compare the resource-rich organisations (i.e., the ‘organisational elite’) identified through the indicators with the rest of organised civil society in a given context. This would be an innovative way to shed a light on the implications and consequences of resource stratification in civil society, for instance related to the role of resources for organisation, participation, and influence in civil society.

While the power structure of the civil society sector is interesting, it is also relevant to look at the organisations that hide behind the numbers. For each context, we have a large set of organisations characterised by an elite score showing how many types of resources they have a disproportionate control of. This allows comparative analyses of landscapes of elite organisations across contexts, for instance looking at the types of organisations, the policy areas they are engaged in, or their ideological orientation.

Our approach contributes to elite studies by allowing research on civil society elites following the same positional method as in political and business elite studies yet adapted to the specificities of the civil society sector. The strategic choices we have described in our application of the approach should be viewed in this light, to identify individual leaders of resource-rich organisations as parts of a civil society elites. Using the identified population of resource-rich CSOs, empirical studies inspired by classical elite studies might then focus on their career trajectories (Lindellee and Scaramuzzino 2020; Santilli and Scaramuzzino 2021), their attitudes (Lee and Scaramuzzino 2024), etc.

Limitations

The MMRSC approach was designed in the process of mapping the most resourceful CSOs in four European countries and at the EU level, in order to identify groups of leaders that could be considered civil society elites in each of these contexts. In the absence of a well-established comparative method of measuring resource accumulation and stratification among CSOs, we were inspired by a set of theoretical perspectives from interest groups, social movement and civil society studies in order to empirically capture different dimensions and types of organisational resources that are of importance for CSOs.

The results of our empirical mapping of CSOs in different national contexts, as well as the method applied as presented in this paper, leave however room for improvement and further development. For instance, the validity and reliability of the method need to be tested in relation to other substantive research questions and to other empirical contexts. The empirical robustness of some of the processes, such as selection of indicators, setting of thresholds for inclusion and exclusion of CSOs, and creating the index of resource accumulation, could be further elaborated and possibly more clearly formalised. Factor analysis could, for instance, provide important insights into the relative importance of different resource indicators in a given context.

While such formalisation would strengthen the method, it is important to acknowledge the fact that the landscape of CSOs is highly complex and takes a variety of forms in different national contexts. In fact, we would argue that the context-specific diversity of civil society sector limits the possibility to propose a priori determined operationalisation of indicators to be employed in the mapping of resourceful organisations. Our approach relies instead on strategic choices based on careful assessments of specific contexts, considering both country-specific adaptation and the need for comparability.

Conclusions and discussion

The contribution of the MMRSC is both theoretical and methodological. Theoretically, the method draws on discussions about what resources are relevant for organisations in the civil society sector. The MMRSC overcomes the challenges of grasping the hierarchical structures between organisations that might characterise civil society and yet be difficult to observe, due to diversity in the types of resources and the multifaceted ways in which power and influence might manifest themselves.

The methodological contribution of the MMRSC as an approach lies in its usefulness both for studying the impact of resources of organisational behaviour in civil society, mapping CSOs based on resource accumulation, addressing resource stratification in civil society and studying elites in civil society. The approach allows comparability across contexts at the same time as it does justice to the diversity of the contexts of civil society and the wide range of aims and forms within civil society. The approach can be adapted to different contexts as the indicators can be operationalised differently based on the contexts. The indicators also allow for different types of resources to be considered important, depending on the type of actors that are considered.

Regarding boundaries, indicators of resources, thresholds and conditions for inclusion and exclusion, our approach leaves much room for choices for researchers. This is in our view one of the strengths of the approach as it allows flexibility and makes it applicable in different contexts. It is also, however, a challenge as all these choices need to be guided by the research question at hand and in-depth knowledge about the context of the study. How to operationalise political resources is for instance dependent on how policy processes look like in that particular context. In a country with strong corporative traditions, we might need to look at formalised committees, while in more liberal systems we might need to look at registries for lobbying activities, or similar. The approach is also developed with an assumption of associational freedom, existence of avenues for participation for civil society in policy processes and a free media. Applying the approach to more authoritarian contexts would likely require other types of resources to be considered.

Following the four steps proposed above means a constant process of exclusion and inclusion of CSOs that should be reflected upon. What part of civil society is left out by drawing the boundaries and what types of organisations are deemed as resource-rich and not based on the indicators chosen should always be carefully considered and commented upon. In our study of civil society elites, we asked experts in each context to look at our sample and, based on our aim, assess to what extent we had missed certain actors they deem part of the organisational elites or if they found some actors that should not have been there. This ‘reputational’ method is often used in elite research alongside the positional method and could be used to validate the results of the mapping of the civil society sector.