Introduction
Amidst a host of pressing global challenges ranging from war to climate change, the concept of social inclusion serves as a promising effort to ameliorate exclusion and marginalisation on the part of nations and international organisations. While many nations and the United Nations already employ the concept of social inclusion to motivate their social policy, there remains a lack of clarity on what is precisely meant by social inclusion. Social policy and efforts for inclusion can benefit from a revisitation of social inclusion and social exclusion, concepts that may risk losing their value by meaning everything to everyone.
Social exclusion has been associated with negative outcomes related to health, education, economic inequality, poverty, violence, and well-being for a myriad of marginalised groups (Ben Brik and Brown, Reference Ben Brik and Brown2024; Brown and Ben Brik, Reference Brown and Ben Brik2025; Khan et al., Reference Khan, Combaz and McAslan Fraser2015). Social exclusion not only occurs from individual vulnerabilities but also from structural conditions and discrimination (UNECE, 2022). In response, national governments across the globe have developed social policies to promote social inclusion. The emphasis of the United Nations’ aspirational 2030 Agenda has even been described, in large part, as embodying the concept of social inclusion (UNDESA, 2016; UNECE, 2022). As governments attempt to reduce social exclusion, questions arise about what exactly is meant by the concept, how the concept is measured, and what levers of social policy have been utilised to address it.
This study seeks to understand the theorisation, conceptualisation, and measurement of social inclusion and exclusion from a comparative perspective, as the concepts relate to social policy, to revitalise the theoretical development of the concepts. Further, we aim to understand how the concepts have developed and been operationalised through social policy throughout their theoretical history. We review theory and measurement together in a single study to facilitate connections between conceptualisation and operationalisation. In addition to addressing a scholarly lacuna, this study aims to be practically beneficial for scholars and practitioners involved with reforming and evaluating social inclusion policy.
Specifically, this study answers the following questions as they relate to social policy: What are the theoretical, conceptual, and definitional frameworks for social inclusion and exclusion, and how have they developed over time? How have social inclusion and exclusion been measured? How are these conceptual and measurement approaches understood in a comparative context? And what developments in these conceptual and measurement frameworks would benefit scholars and policymakers in developing social inclusion policy for our era?
The concepts of social inclusion and social exclusion are expansive and often quite nebulous. Social exclusion has been used interchangeably and in overlapping ways, such as other concepts like social inclusion, multi-dimensional poverty, multiple inequalities, and well-being (UNECE, 2022). Moreover, the literature has swung from emphasising social exclusion to social inclusion because of its more positive framing. For these reasons, we refer to social inclusion and social exclusion together, as a dynamic spectrum, using the phrase ‘social in/exclusion’.
Methods
With a concept as malleable as social in/exclusion, Cobigo et al. (Reference Cobigo, Ouellette-Kuntz, Lysaght and Martin2012) recommend conducting a synthesis review instead of a systematic review. This involves collating peer-reviewed academic articles and grey literature from diverse databases while building safeguards to limit bias. While systematic reviews may be appropriate and feasible for some subsections of the literature on social in/exclusion, we believe our review of social in/exclusion is too broad for a strict review that wishes to include conceptual, empirical, and grey literature, which is often difficult to capture through a traditional systematic review.
This methodological approach to answering our research questions is also warranted due to the vastness of the literature on social in/exclusion, which spans at least five decades and numerous countries. Our aim of contributing to potential conceptual developments that inform social policy for our era necessitated a methodology that allowed us to grasp the essence of the concepts, despite their spanning spatial and temporal dimensions. While many sources were identified prior to the turn of the century, we have prioritised reporting studies conducted since the latetwentieth century because they often summarise the previous literature and for the sake of limiting the scope of our discussion.
Search strategy and inclusion/exclusion criteria
We began by searching for reviews of the concept and measurement of ‘social inclusion’ and ‘social exclusion’ in electronic databases between May and July 2023, closely reading these works and synthesising their definitional, conceptual, and measurement approaches to social in/exclusion. While analysing these works narratively, we also reviewed key texts referenced in them and scraped their bibliographies for additional references.
After this unguided review, we followed Cobigo et al.’s (Reference Cobigo, Ouellette-Kuntz, Lysaght and Martin2012) methodology of narrowing our search methods and sources. We built on our previous conceptual review by searching for peer-reviewed studies in ERIC, PsycINFO, and Social Services Abstracts, using the keywords ‘social inclusion’ OR ‘social exclusion’ AND ‘social policy’. Papers were excluded if they were written in a language other than English. We did not limit the search by date, though the previously listed databases do have inherent date restrictions.
Study selection, screening, and quality assessment
Following the examples of previous related reviews (e.g., Levasseur et al., Reference Levasseur, Richard, Gauvin and Raymond2010; Cobigo et al., Reference Cobigo, Ouellette-Kuntz, Lysaght and Martin2012;Wright and Stickley, Reference Wright and Stickley2013; Pinkert et al., Reference Pinkert, Köhler, von Kutzleben, Hochgräber, Cavazzini, Völz, Palm and Holle2019), we reviewed titles and abstracts to determine the relevance of identified papers and identified additional sources by reviewing select studies’ references. We excluded papers if they focused on narrow concepts unrelated to our research questions and limited our inclusion criteria to papers that reported an empirical, review, or conceptual study and provided a definition of social inclusion or social exclusion in the context of social policy.
This search initially yielded 7,442 studies – 890 from ERIC, 2,778 from PsycINFO, and 3,774 from Social Services Abstracts. After reviewing titles and abstracts with our inclusion criteria in mind and removing duplicates, we were left with 117 studies. We then skimmed these papers with the previously mentioned inclusion criteria as guideposts for inclusion, removing another six articles, leaving 111 studies. Figure 1 illustrates this review process.
Flow chart of reviewing process
Flow chart of reviewing process.
Data extraction and analysis
We analysed the key principles, elements, and modes of conceptualisation and measurement narratively to add to the summarisation of the main themes in the literature. This involved closely reading each article, noting definitional, conceptual, and measurement features, and synthesising multiple articles together into common themes. We also reviewed key references for these papers to deepen our review, which greatly expanded the total number of studies. For instance, Panda et al.’s (Reference Panda, Pattanayak, Pandey, Panda, Pandey and Pattanayak2022) illustrative edited volume, Social Exclusion and Policies of Inclusion, was not found in the formal review but offered valuable insight into social in/exclusion around the world. Our aim was not to follow a strict method of systematic review but rather to cast a wide net, deeply review that literature, and synthesise our findings with a saturated understanding of the concepts.
Conceptualising social in/exclusion
While the concepts made considerable evolutions into the early 2000s, we find relative consistency in its use in recent decades. We also find relative consistency in how definitions of the concepts conceptualise social in/exclusion. Most of the definitions align with Oxoby’s (2009) principles: ‘definitions of inclusion (and exclusion) are general, focusing on (i) the access to rights and resources and (ii) the presence of obstacles to social institutions without directly defining the rights or institutions in question’, (p. 5). Following other key texts in the literature, we understand social in/exclusion mainly as a process that contributes to states of social cohesion, states of social exclusion, poverty, and other related outcomes, but also secondarily as a unique outcome in itself.
The conceptual development of social in/exclusion in comparative context
The concept of social in/exclusion evolved from restricted labor market involvement and material deprivation to fuller participation in society, including economic, social, political, and cultural activities (Ben Brik and Brown, Reference Ben Brik and Brown2024). The concept of social in/exclusion is generally attributed to René Lenoir’s (Reference Lenoir1974) discussion of the ‘social misfit’ in Les Exclus (Rawal, Reference Rawal2008; UNDECSA, 2016). However, there has been considerable debate in conceptualising and defining social in/exclusion since its inception. This may be because the concepts change as societies develop (Das et al., Reference Das, Fisiy and Kyte2013), and because social in/exclusion is relative – both between individuals, groups, and nations (Allman, Reference Allman2013; Saunders, Reference Saunders2011; Nissi and Rapposelli, Reference Nissi, Rapposelli, Parodi and Sciulli2012). It may also be influenced by changing social and political currents where the concept was adopted in social policy. Driven by its adoption by French policymakers in the 1980s, the concept was expressly adopted by the European Commission in 1988 (Wilson, Reference Wilson2006). Originally only accounting for people who did not adjust to mainstream society, it has recurringly been redefined to include specific marginalised groups (Aasland and Flotten, Reference Aasland and Fløtten2001). The concept seems to have first been appreciated on the global stage after the United Nations’ 1995 First World Summit on Social Development (Rawal, Reference Rawal2008). It was soon taken up by various social policy initiatives, diffusing throughout Commonwealth countries such as the United Kingdom, Australia, New Zealand, Canada, and South Africa (Allman, Reference Allman2013). Wilson (Reference Wilson2006) argues that the concept was also taken up in the United States in the form of the ‘New Right’ of the 1980s in political discussions of the persistence of poverty and government dependence.
Although social in/exclusion was originally informed by French Republicanism, its meaning seems to be influenced by the dominant political philosophies of the societies in which it is conceptualised (Silver, Reference Silver1994). For instance, Silver (Reference Silver2015) argues that the New Labour government influenced its conception in the United Kingdom, Australia, and Canada, combining communitarian and neoliberal notions of social citizenship; influenced by their historical roots in socialism, Central and Eastern European nations have focused on a single excluded group – the Roma; and many nations in Latin America have focused on the class of informal workers. Chenaut (Reference Chenaut, Panda, Pandey and Pattanayak2022) offers another case study of the Emiliano Zapata village in Mexico, where neoliberal policies incentivised extractive, large-scale oil installations that contributed to community social exclusion.
Silver (Reference Silver1994) argued that the meaning of social in/exclusion depends on the nature of a given society and that conceptualisations vary between nations and cultures, which Francis (Reference Francis, Minogue and Kothari2001) identifies as a strength because it incorporates facets of culture and institution, unlike typical conceptions of poverty. Silver (Reference Silver1994) identified three paradigms of social exclusion reflected in social policy regimes: solidarity, specialisation, and monopoly. The solidarity paradigm is rooted in French Republican thought and links exclusion to the erosion of social solidarity. The specialisation paradigm, drawing on Anglo-Saxon ideas of social differentiation, attributes exclusion to individual specialisation within the labour market. The monopoly paradigm interprets exclusion as stemming from hierarchical group monopolies that wield power through class and status, with remedies found in social citizenship (Rawal, Reference Rawal2008). Later, Silver (Reference Silver2015) argued that Liberal thought associates social inclusion with guaranteed rights to individual freedoms; Republican thought, with social solidarity; Social Democratic thought, with rights to a minimum standard of living; Conservative thought, with natural hierarchy; and Confucian thought, with social harmony over individual liberty. Gidley et al. (Reference Gidley, Hampson, Wheeler and Bereded-Samuel2010) propose an alternative framing that centres on neoliberalism, social justice, and human potential.
While frameworks based on political ideology are insightful, Lyons and Huegler (Reference Lyons, Huegler, Healy and Link2012) described a more universal framework – the Moral Underclass (MUD), Social Integrationist (SID), Redistributive (RED) – and scholars like Chau et al. (Reference Chau, Yu and Boxall2018) have added Collective Production (COP) as a fourth approach. MUD underscores the moral and cultural shortcomings of the individual rather than structural conditions in society; SID frames social exclusion as exclusion from paid employment; RED centres on the redistribution of wealth and power; and COP emphasises co-ownership of the means of production, the agenda, and the production process. Panda et al. (Reference Panda, Pattanayak, Pandey, Panda, Pandey and Pattanayak2022) also point to the political economy of social exclusion, whereby exclusion is a byproduct of capitalism and development, relegating through slavery, colonisation, working-class oppression, and periphery states in the Global South. They offer case studies of witch hunts in India (Kelkar et al., Reference Kelkar, Sinha, Nathan, Panda, Pandey and Pattanayak2022), anti-immigrant-and-refugee policy in the United States (Ahmed-Ghosh, Reference Ahmed-Ghosh, Panda, Pandey and Pattanayak2022), the interplay between global care networks and gender oppression in Singapore (Khan, Reference Khan, Panda, Pandey and Pattanayak2022).
Although social inclusion is often regarded with its logical opposite, social exclusion, the difference between the two is often described as a spectrum, not a binary contrast. For example, Popay et al. (Reference Popay, Escorel, Hernández, Johnston, Mathieson and Rispel2008) define exclusion as a ‘… dynamic, multi-dimensional process driven by unequal power relationships … [that] results in a continuum … characterised by unequal access to resources, capabilities and rights, (p. 2). In this conception, social in/exclusion is understood as a spectrum of lack of resources, rights, and participation. Dichotomous notions of inclusion and exclusion bring up an either/or when there are often actually both/and situations (Jackson, Reference Jackson1999). Universal, dichotomous notions of inclusion and exclusion also resist an intersectional perspective where some groups may have varying intensities of exclusion (Francis, Reference Francis, Minogue and Kothari2001). Moreover, the inclusion of some groups may come with the exclusion of other groups, exclusion can be accompanied by strength and resistance, and some groups may not wish to be included in the organisation of the state (Rawal, Reference Rawal2008).
Social in/exclusion is often conceptualised as multidimensional, a process rather than an event, constantly evolving, and consisting of intersecting disadvantages (Millar, Reference Millar, Abrams, Christian and Gordon2007; Saunders, Reference Saunders2011; Huxley et al., Reference Huxley, Evans, Madge, Webber, Burchardt, McDaid and Knapp2012; Nissi and Rapposelli, Reference Nissi, Rapposelli, Parodi and Sciulli2012). Moreover, individuals are excluded because of the social status associated with their intersecting identities (Das et al., Reference Das, Fisiy and Kyte2013). This understanding has led many scholars to describe various domains of the concept. Among these domains, there is a key distinction between rights-based and participatory-based conceptualisations of social in/exclusion (Huxley et al., Reference Huxley, Evans, Madge, Webber, Burchardt, McDaid and Knapp2012). In attempts to bridge the two perspectives, Huxley et al. (Reference Huxley, Evans, Madge, Webber, Burchardt, McDaid and Knapp2012) suggest that social in/exclusion should be defined as a failure of a system of multimodal integration. Many other scholars have also suggested various domains of the concept (e.g., Aasland and Flotten, Reference Aasland and Fløtten2001; Reimer, Reference Reimer2004; Levitas et al., Reference Levitas, Pantazis, Fahmy, Gordon, Lloyd-Reichling and Patsios2007; Silver, Reference Silver2015).
Despite ever-growing boundaries around the concept, scholars have generally concluded that social in/exclusion is also multileveled, meaning that it manifests at different levels from the individual to the nation. Byrne (Reference Byrne2005) emphasised spatial separation as a driver of social in/exclusion, Reimer (2004) posited that social in/exclusion should be understood at the levels of the individual, family and close networks, local community or employer, and government policies and programmes, and Silver (Reference Silver2015) argued that social in/exclusion is even context-dependent on a particular neighbourhood.
Social in/exclusion has been defined as both distinct from and subsuming of the concepts of poverty and inequality. Some have even suggested that because of the concept’s nebulousness, social in/exclusion has extended notions of material poverty to other social issues due to the lack of a rigorous theoretical framework of what is deemed inclusion and exclusion (Geddes and Benington, Reference Geddes and Benington2001), while others even posit that social in/exclusion replaced the concept of poverty in Europe because it is inclusive of numerous dimensions of poverty (Aasland and Flotten, Reference Aasland and Fløtten2001). Social in/exclusion not only includes material deprivation but also feelings of lack of agency and inferiority often associated with poverty (UNDECSA, 2016). Sen (Reference Sen2000) associated social in/exclusion with his capabilities approach to poverty, and Rawal (Reference Rawal2008) notes that discourse around social exclusion is often concerning poverty and capabilities, particularly for development studies. UNECE (2022) concluded, however, that despite their conflation, poverty and social in/exclusion are distinct concepts, where poverty represents a potential outcome of social exclusion.
Our review of studies in the last decade revealed little conceptual development or clarification, which follows Filia et al.’s (Reference Filia, Jackson, Cotton, Gardner and Killackey2018) conclusion that definitions and conceptualisations of social in/exclusion have been relatively consistent in the literature, generally grouping around the domains of social support and participation, employment or education, and housing and neighbourhood. However, we found a recent stream of studies that extend the concepts outside of the Western context to China (e.g., Cheung, Reference Cheung2013; Zhu and Walker, Reference Zhu and Walker2019). Applications of social in/exclusion in certain Asian contexts may harken to the Confucian perspective advanced by Silver (Reference Silver2015) and the Collective Production proposed by Chau et al. (Reference Chau, Yu and Boxall2018). However, Panda et al.’s (Reference Panda, Pattanayak, Pandey, Panda, Pandey and Pattanayak2022) emphasis on the political economy of social in/exclusion was observed in several case studies in India related to youth (Lund and Panda, Reference Lund, Panda, Panda, Pandey and Pattanayak2022), widows (Pandey and Tripathy, Reference Pandey, Tripathy, Panda, Pandey and Pattanayak2022), and older people (Pandey, Reference Pandey, Panda, Pandey and Pattanayak2022). In this sense, following Nadarajah’s (Reference Nadarajah, Panda, Pandey and Pattanayak2022) discussion of the adoption of social in/exclusion in Papua New Guinea, concepts other than but related to social in/exclusion may be drawn into the umbrella of social in/exclusion (like the concept of poverty in Europe) through policy and idea diffusion in parts of the world outside of Europe, the Commonwealth countries, Latin America, and Asia – such as Africa and the Middle East. Other recent conceptual development came from Adlam (Reference Adlam2020), who suggested a novel way to categorise conceptualisations of social in/exclusion around the notions of metropolitan and cosmopolitan systems and highlighted the concepts’ environmental dimension, and Boutlet (Reference Boutlet, Panda, Pandey and Pattanayak2022) call to broaden inclusion to non-humans.
The conceptual development of the concept points to an evolution from French Republicanism to diffused understandings in various national contexts. Instead of the spread of the novel idea of social in/exclusion, this development may be seen as the incorporation of related concepts under the umbrella of social in/exclusion with policy and idea diffusion and agenda setting by the United Nations and other international organizations. Modifications to the concepts may be driven by shifting social and political currents in their respective national milieus. While the boundaries of social in/exclusion continue to expand to meet contemporary needs, conceptualisations appear to have stabilised at the turn of the twentieth century to describe a multidimensional, multilevel, dynamic, and relational process on a continuum between social inclusion and exclusion. While there has been little conceptual development or clarification, UNECE (2022) recently called for unified conceptualisation and measurement of the concepts around the United Nation’s 2030 Sustainable Development Goals. This call for a unified conceptualisation may be warranted for a high-level understanding of social inclusion, particularly at the national and cross-national levels, but the amalgamation may risk a loss of utility and clarity as the concept may mean everything to everyone as various related concepts from cultures around the planet contribute to its conceptual emergence.
Approaches to measurement
UNECE (2022) argues that measuring social in/exclusion should be a priority because it ‘… enables policymakers to identify individuals and communities suffering from systematic disadvantage and the process by which their exclusion takes place’, (p. 6). Large and detailed datasets have traditionally been necessary to measure social in/exclusion because of its multidimensional nature. While many measures have attempted to capture particular domains of the concept – such as material resources, labor market outcomes, education, health, social support and interactions, community engagement, and personal safety (Scutella et al., Reference Scutella, Wilkins and Horn2009; Scutella and Wilkins, Reference Scutella and Wilkins2010; UNECE, 2022) – Cordier et al. (Reference Cordier, Milbourn, Martin, Buchanan, Chung and Speyer2017) found that measures in the context of social policy have typically focused on employment and housing.
Measures of social in/exclusion can be divided between national-level survey data and individual-level instruments, with extensive literature dedicated to each level of analysis. We present here syntheses of the literature at both levels of analysis to clarify this discussion. We do not find, however, a suitable measure or measurement framework to be uniformly used by policymakers. This is because social in/exclusion may be conceptualised slightly differently in various national and subnational contexts around the planet.
At first brush, national measures of social in/exclusion may be preferable for the development and evaluation of social policy because social programmes tend to seek population-level effects. However, in their multinational review of national measures of social in/exclusion, UNECE (2022) concluded that without a universally agreeable definition, there was no obvious measurement framework to use.
While most studies have leveraged existing data to operationalise social exclusion, the first survey designed to collect data explicitly related to the concept was the 1999 cross-sectional UK ‘Millennium Survey of Poverty and Social Exclusion’, which was organised around the dimensions of income or resources; the labor market; community services; and social relations (Pantazis et al., Reference Pantazis, Gordon and Levitas2006; Scutella and Wilkins, Reference Scutella and Wilkins2010). This study was then emulated by the ‘Community Understandings of Poverty and Social Exclusion’ study in Australia in the following decade. Other examples include the ‘Poverty and Social Exclusion Survey’ in the United Kingdom; the material deprivation, quasi-joblessness, and income poverty components of the annual ‘EU-SILC’ surveys by the European Union; and the ‘Quality of Life, Social Capital, Poverty and Social Exclusion’ in Poland (UNECE, 2022). Though they may exist, we were unable to identify similar surveys attributed to social in/exclusion in other parts of the world.
While many nations have now developed census-like surveys that specifically pertain to the concept of social in/exclusion, it is still common to adapt existing data as proxies. In their seminal book on the topic, Atkinson et al. (Reference Atkinson, Cantillon, Marlier and Nolan2002) suggested measures of poverty, employment, housing, and education as indicators of social inclusion. In a later iteration, Atkinson et al. (Reference Atkinson, Marlier and Nolan2004) suggested ten primary indicators – low-income rate after transfers with low-income threshold set at 60 percent of median income, distribution of income (income quintile ratio), the persistence of low income, median low-income gap, regional cohesion, long term unemployment rate, people living in jobless households, early school leavers not in further education or training, life expectancy at birth, and self-perceived health status – and several related secondary indicators that offered slight variations to the primary indicators. These methods dominated the European approach for at least a decade, though more recent measures have taken slightly different approaches. On the other hand, Chakravarty and D’Ambrosio (Reference Chakravarty and D’Ambrosio2006) developed an axiomatic measure of social exclusion that indexed an individual-level measure of a series of questions with an importance and population size weight testing their index on the European Community Household Panel data using seven categories of indicators: financial difficulties, basic necessities, housing conditions, durables, health, social contact, and dissatisfaction.
While both sets of measures offer pragmatic attempts at measuring social in/exclusion by examining existing indicators, they do not appear to necessarily be grounded in nor adequately represent a nuanced and contextualised conceptual framework, mainly measuring the economic and social dimensions of the concept. A more cogent approach is that of UNDESA (2016), which grouped its indicators around their three conceptual domains – opportunity, employment and income, and participation. They proposed indicators of education, health services, and infrastructure to measure opportunity; labor market participation and income to measure employment and income; and involvement in political, civic, and cultural life to measure participation. Similarly, Wilson’s (Reference Wilson2006) index of social inclusion asks questions that address community togetherness and closeness; differences in characteristics between living in the same community in terms of wealth, income, class, social status, ethnic background, gender, religion, political beliefs, and age; and whether differences are perceived to cause problems. Another, more recent strain of national-level instrument development comes from attempts at measuring social in/exclusion among children at the national level (Gross-Manos, Reference Gross-Manos2015; Crous and Bradshaw, Reference Crous and Bradshaw2017). Although these approaches continue to conflate levels of measurement, they offer examples of indexes that are more conceptually comprehensive.
At least three important considerations of these national-level instruments have been raised in the literature. First, how do we best communicate these aggregate measures? Second, at what threshold do we distinguish between social inclusion and social exclusion? And third, do national measures offer enough granularity to inform social policy?
UNECE (2022) explains that there have been numerous approaches to presenting national measures of social in/exclusion, including composite indices and data dashboards, depending on the purpose of the communication and granularity of data. However, the use of composite indexes is controversial in the literature, with some scholars advocating their communicability, while others disparaging their obfuscation (Ravallion, Reference Ravallion2011; Maitre et al., Reference Maitre, Nolan and Whelan2013). For example, Rogge and Self (2019) advanced the use of the benefit-of-the-doubt weighting methodology when developing a composite index of the European Union’s at-risk-of-poverty or social exclusion index (AROPE), which allows regional contextual data to inform the weighting component of the index. The Europe 2020 AROPE Index negated scores of social exclusion – which actually measured poverty using an at-risk poverty rate, severe material deprivation rate, and low-work intensity household rate – to measure social inclusion (Rogge and Self, Reference Rogge and Self2019). Another interesting example is the use of methods such as nonlinear canonical correlation analysis to reduce the multidimensional measure of social in/exclusion into a single index (van Bergen et al., Reference van Bergen, Hoff, van Ameijden and van Hemert2014).
Another question is how to determine the threshold between social inclusion and exclusion. UNECE (2022) identified three approaches: a single threshold, multiple thresholds, and no thresholds. They posit that the single threshold approach includes when there is an accumulation of deprivations or lacks, when the distribution relates to being more or less excluded, and when a particular threshold of deprivations may be defined, without considering the relative position of exclusion as part of a wider distribution; the multiple threshold approach includes when there is a full distribution ranging from highly included to highly excluded; and the no threshold approach focuses on another concept such as well-being or social cohesion, with social exclusion indicators included but no threshold for exclusion explicitly defined (UNECE, 2022).
Lastly, there is the issue of whether national measures offer enough granularity to inform social policy. Early on in the measurement literature, Atkinson et al. (Reference Atkinson, Cantillon, Marlier and Nolan2002; Reference Atkinson, Marlier and Nolan2004) advanced regional measures of social in/exclusion to make causal inferences, which may be muddied at the national level. More recently, Silver (Reference Silver2015) suggested a neighbourhood-level understanding of the concept. However, with the proliferation of national measures in numerous nations, this question, in our view, has yet to be thoroughly addressed in practice. One notable exception is the work of Prattley et al. (Reference Prattley, Buffel, Marshall and Nazroo2020), who used hierarchical models to tease out causal claims regarding social in/exclusion over the life course. Now, we turn to our review of individual-level measurements.
Considerable scholarship from the last two decades has also been dedicated to developing valid and reliable instruments that measure social in/exclusion at the individual level. While they have some limitations, these instruments, particularly the Social and Community Opportunities Profile (SCOPE), show great promise for being contextualised to new populations and regions.
In their review of individual-level instruments to measure social inclusion in the context of mental health, Coombs et al. (Reference Coombs, Nicholas and Pirkis2013) reviewed ten micro measures of social inclusion, concluding the APQ-6 (Stewart et al., Reference Stewart, Sara, Harris, Waghorn, Hall, Sivarajasingam, Gladman and Mowry2010) and the SCOPE (Huxley et al., Reference Huxley, Evans, Madge, Webber, Burchardt, McDaid and Knapp2012) deserved further testing. In their systematic review of the psychometric properties of measures of social inclusion, Cordier et al. (Reference Cordier, Milbourn, Martin, Buchanan, Chung and Speyer2017) identified an astounding 108 instruments used in the literature. While there exist numerous instruments, most of them measure similar definitions of social in/exclusion (O’Donnell et al., Reference O’Donnell, O’Donovan and Elmusharaf2018). New instruments continue to be developed, such as Diez and Fumanal’s (Reference Diez and Fumanal2023) Social Situation Scale. Despite numerous options, Cordier et al. (Reference Cordier, Milbourn, Martin, Buchanan, Chung and Speyer2017) concluded that the SCOPE was the most promising instrument because it covered the most conceptual territory and did not display negative psychometric properties. Unlike other inventories, these individual-level inventories, particularly the SCOPE, demonstrate good construct validity, internal reliability, and test-retest reliability (Huxley et al., Reference Huxley, Evans, Madge, Webber, Burchardt, McDaid and Knapp2012).
The SCOPE takes a conceptual demarcation between objective and subjective measures of social in/exclusion and measures both social inclusion and social exclusion. SCOPE asked survey questions related to individual demographics, satisfaction with opportunities, perceived opportunities, satisfaction with inclusion in society, objective opportunities, and objective participation along the domains of leisure and participation, housing and accommodation, safety, work, finance, education, health, family and social, and overall inclusion (Huxley et al., Reference Huxley, Evans, Madge, Webber, Burchardt, McDaid and Knapp2012). The SCOPE has also been shown as a viable option to contextualise social in/exclusion in other nations, evidenced by Chan et al.’s (Reference Chan, Chiu, Evans, Huxley and Ng2016) successful effort to validate the SCOPE in Hong Kong among mental health service users. Importantly, before validating the instrument, Chan et al. (Reference Chan, Evans, Ng, Chiu and Huxley2014) conducted a concept mapping exercise with focus groups to understand how the concept of social in/exclusion was understood in the Hong Kong context, which led to slight alterations of the SCOPE because although most of the concept mapped to the new context, respondents emphasised individuals’ civic responsibility over civil rights that guarantee access to resources.
Discussion
While our review of the conceptualisation and measurement of social in/exclusion has yielded a clearer understanding of the conceptualisation and measurement of social in/exclusion, it also highlights the lack of conceptual clarity and suitable measurement framework to be operationalised in social policy. Growing out of a French Republican tradition, the concept of social in/exclusion has diffused to numerous national and cross-national contexts and evolved to become multidimensional, multilevel, dynamic, and relational. Numerous efforts have been undertaken to measure the concept at the national and individual levels. However, we found little evidence of the application of social in/exclusion outside of the Commonwealth countries, Europe, the United States, and Latin America, though the concept may soon be adopted through the subsumption of related topics through policy and idea diffusion and agenda setting by the United Nations and other international organisations.
However, the concept is still not entirely clear to us, possibly because previous scholars have grappled with defining relative concepts understood across a vast literature. To be rigorously applied in the context of social policy, we suggest that what has grown into an all-encompassing concept should be carefully analysed in its component parts and cultural contexts. We outline, here, areas of future study to clarify the concept and its measurement.
While the concept is understood to relate to individuals, institutions, and societies in unique and complex multilateral ways, its operationalisation through social policy can benefit from further exploration of the concept’s multilevel nature. Recent scholarship has begun to emphasise the well-established relational dimension of social in/exclusion and argue that indicators of the concept should account for the relational dynamic between the micro, mezzo, and macro levels (Simplican et al., Reference Simplican, Leader, Kosciulek and Leahy2015; Hill et al., Reference Hill, Thomson, Raven, Wong, Cass, Yeandle and Buckner2016; Hill and Thomson, Reference Hill and Thomson2017). Importantly, MacLeod et al. (Reference MacLeod, Ross, Sacker, Netuveli and Windle2019) demonstrate that the domains of social in/exclusion interact with and mediate each other in different ways, which we believe can also be represented through an ecological model of the concept. In addition to providing conceptual clarity, this focus can benefit both the measurement and intervention of the concept because an intervention at one level may be mismeasured at another, or vice versa. Additionally, the concept can benefit from consideration of the physical environment – both spatial and environmental – alongside economic, political, social, and cultural in/exclusion. Some individuals and groups are marginalised through spatial exclusion, such as urban segregation and rural/urban divides that shape political, economic, and social access (Cameron, A, Reference Cameron2006). Moreover, a burgeoning literature has demonstrated that some individuals and groups are at greater exposure to the effects of climate change, environmental degradation, and pollution (Adlam, Reference Adlam2020).
A closely related issue is that the concept seems to have considerable and sometimes conflating overlap with related concepts such as social capital, social cohesion, and social integration. The concept of social in/exclusion also exists in a network of related concepts that both clarify and obscure each other’s distinct contributions (Oxoby, 2009; Huxley et al., Reference Huxley, Evans, Madge, Webber, Burchardt, McDaid and Knapp2012). In addition to social in/exclusion, the literature uses terms such as social support, social participation, social networks, social mobility, social cohesion, social integration, and social capital, to name a few, with the concepts of social capital and social cohesion as particularly salient in the literature. Despite the ambiguous overlap between the concepts, we have yet to discover a conceptual framework for how these concepts relate to one another. This is further complicated by the fact that these relations may be understood differently in different parts of the world.
We also see a potential disconnect between the conceptualisation and measurement of social in/exclusion in the context of social policy because of these ambiguities and pragmatic data limitations, which have mainly manifested in measures that are conceptually incomplete and fail to account for the multilevel nature of the concept. Another concern garnered from our review is the connection between the concept of social in/exclusion and its measurement, particularly between ecological levels that exist as networks. Any attempt at operationalising measures of social in/exclusion should naturally be grounded in its conceptualisation. Following this maxim will allow for greater validity and reliability of measurements, and, ideally, a greater understanding of the mechanisms that influence and might be used to influence the lived and material experiences of social in/exclusion. Moreover, social in/exclusion may not be either/or, and subsequent measures of social in/exclusion may struggle with dichotomous categorisation. Social inclusion may not necessarily negate social exclusion, so indexes that subtract measures of one from the other to calculate a final measure may have shaky validity.
This review elucidates several streams for future research. Future theoretical work can build on this review by addressing the identified limitations in the current conceptualisation of social in/exclusion. There has been little conceptual or empirical work related to understanding global trends in social in/exclusion and social inclusion policy. Additionally, there is much work to be done in the measurement of social in/exclusion. Indicators can be developed within each dimension of the social in/exclusion that adequately represents that domain and maintains it through factor analysis. These indicators can be contextualised to the particular nation or region, ideally using resident input through surveys or focus groups. Further, in addition to individuals’ information, data should be collected on other ecological levels of interest that relate to the individual. This approach can allow for multilevel or hierarchical modeling (e.g., Prattley et al., Reference Prattley, Buffel, Marshall and Nazroo2020) and network analysis. Moreover, following examples such as the SCOPE, individuals’ perceptions of social in/exclusion can be collected, alongside other psychological and health-related information. To the greatest extent possible, data can also be collected using longitudinal research designs, which can allow for more nuanced understandings of these relationships over time.
Implications for social policy
The multidimensional and relational nature of social in/exclusion highlighted in this review offers a foundation for actionable social policy. Policymakers could prioritise initiatives that bridge economic and social domains, such as subsidised training programmes paired with community integration efforts, to address exclusion arising from both material deprivation and social isolation. Recognising the dynamic interplay of exclusion across ecological levels, policies could also foster participatory governance models where local communities co-design inclusion strategies, ensuring they reflect context-specific needs rather than top-down assumptions. Furthermore, as global transitions such as urbanisation and technological change reshape social landscapes, embedding social inclusion goals within broader sustainability efforts, such as ensuring equitable access to resources, could preempt emerging forms of exclusion.
Conclusion
In this study, we apply a synthesis review to integrate the vast literature on social inclusion and social exclusion. We outline the conceptual development of social in/exclusion and how it has been measured. While there has been relative consistency in the conceptual and definitional use of social in/exclusion since the turn of the twentieth century, its measurement – both at the national and individual levels – has seen recent development. Despite this, we do not find a suitable measure for uniform use in the context of social policy, especially because the concept may vary by time and place. We call for several lines of future study to address these conceptual and measurement limitations. Social inclusion still stands as a promising concept to be used by social policy to ameliorate exclusion and marginalisation. However, significant theoretical work must be undertaken to adequately employ the concept in practice.
Funding statement
This study was supported by funding from the Qatar Foundation.
Competing interests
The authors declare none.
Open access
