1 Introduction

Multidimensionality is a key feature of social exclusion, that is, exclusion concerns different life domains, and disadvantages in one domain can be interrelated with disadvantages in other domains. Interrelationships across social exclusion domains have implications not only for the experience of the individual, but also for policy, since cutbacks or interventions addressing one domain may have additional effects on one or several other domains.

This section includes four chapters that examine interrelationships across social exclusion domains. These chapters place focus on older adults living in long-term care institutions; the relationship between exclusion from material resources and loneliness; how transport influences the multidimensionality of social exclusion; and homelessness among older adults.

The purpose of this chapter is to introduce interrelationships that can arise between different domains, within the multidimensionality of social exclusion in later life, and to broadly position the chapters presented in this section within this literature. The chapter starts with a brief reminder of some of the domains often considered in research on social exclusion. Thereafter, research on interrelationships in terms of cross-sectional and longitudinal associations across domains is presented. Finally, each of the chapters within the section are briefly introduced.

2 Social Exclusion Domains

While key work by Burchardt et al. (2002), based on an analysis of the whole population, identified four domains of social exclusion – social interaction, political engagement, consumption, and production – researchers on exclusion in later life have critiqued the relevance and moved away from areas focused on labour market participation and economic production (e.g. Scharf et al. 2005). Instead, and as demonstrated by Walsh et al. in Chapter 1 of this volume, research related to ageing has typically emphasised domains that are more instrumental to the daily lives of older adults. This includes emphasis on the domain of neighbourhood and community (Scharf et al. 2005) as being crucial to addressing fundamental needs and staying independent in later life (Scharf et al. 2005; Van Regenmortel et al. 2018). Recent work has demonstrated that this domain can uniquely contribute to explain the variance in well-being of older adults (Dahlberg and McKee 2018).

Research on social exclusion of older adults has commonly considered exclusion from some or several of the following domains: material resources, social relations, civic participation, neighbourhood, services, and socio-cultural aspects of society (for overviews, see: Van Regenmortel et al. 2016; Walsh et al. 2017; Walsh et al. this volume), with a recent review identifying as many as 16 domains within the literature on social exclusion in older-age (Van Regenmortel et al. 2016). While some individuals may experience exclusion from one of these domains, others may be simultaneously excluded from two or more domains. This has been labelled multiple or deep exclusion (Scharf et al. 2005; Barnes et al. 2006; Levitas et al. 2007). As will be considered in the next section, some of the outcomes of these different forms of exclusion can be linked to each other and function as a part of other domain specific exclusionary processes.

3 Previous Research on Interrelationships

This research overview considers quantitative and not qualitative studies, with a focus on studies that have considered the multidimensional social exclusion construct rather than exclusion from, for example, just two domains. Quantitative approaches to researching multidimensionality include examining the number of domains on which older adults are excluded, clusters of different forms of exclusion, and the interrelationships across several domains.

In considering the first approach, exclusion from two or more domains have been found in 36% of people aged 60 years or older in deprived urban areas in England (Scharf et al. 2005), in 20% of a national sample of people aged 50 years or older in England (Barnes et al. 2006), and in 11% of people aged 55 years or older in Australia (Miranti and Yu 2015). This approach is based on a pre-defined threshold at which people are regarded as being socially excluded. The threshold for exclusion can be set in absolute terms by defining a level of activities or resources that can be considered as standard in the society in which people live (as in the studies by Scharf et al. and Miranti and Yu) or in relatives terms, for example, by defining a certain proportion of those least active as excluded (as done by Barnes et al.). Of course, the proportion of individuals identified as excluded is directly determined by the level at which this threshold is set (see Van Regenmortel et al. 2018).

A second approach to the study of multidimensionality of social exclusion is to examine clusters of different forms of exclusion indicators across the domains. Van Regenmortel et al. (2018) identified four clusters with, for example, one cluster that was comprised of people who had a higher probability of digital, civic and financial exclusion. Becker and Boreham (2009) explored social exclusion over time and found three dynamic clusters. Based on nine waves of data collection, they identified that between 44 and 68% of individuals had been in the same cluster at the previous wave of data collection, while 26–39% had been in a different cluster at the previous wave.

These approaches to researching social exclusion consider the coexistence of exclusion from different domains or the links between exclusion indicators, but not the interrelationship between exclusion from different domains. Interrelationships can be examined by analysing bivariate associations between domains. This approach has only been used in a small number of studies, mostly from the United Kingdom. The findings of these analyses are summarised in Table 22.1. Positive associations have been found between the majority of domain combinations, that is, being excluded on one domain increases the risk of exclusion from another domain. For example, a Swedish study with a sample of individuals aged 70 years or older found significant associations between all the exclusion domains analysed (Heap et al. in progress). While these associations were generally quite weak, stronger associations were found between exclusion from social relations and the other domains.

Table 22.1 Research findings on interrelationships across domains
Full size table

Relationships can also be reciprocal between more than two domains. While this specific dynamic is even less studied, interrelationships have been identified between social relations, civic activities and neighbourhood exclusion, and between civic activities, material consumption and financial products (Barnes et al. 2006).

Interrelationships across social exclusion domains can also operate over time, so that exclusion from one domain increases the risk of subsequent exclusions from another domain. This can be a cumulative process (see Silver and Miller 2003), where exclusion in one domain can lead to additional disadvantages in other domains (cf. the cumulative advantage/disadvantage theory, Dannefer 2003). Evidence for cumulative processes was found in a Swedish study of people aged 77–95 years at follow-up, particularly if the initial disadvantage concerned social relations (Heap and Fors 2015).

Longitudinal research on social exclusion in older adults is rare. That said, a study from Australia has shown that individuals experiencing social exclusion at one point in time have a considerably higher risk of experiencing exclusion again later (Miranti and Yu 2015). However, this research considered multiple exclusion over time and not interrelationships between exclusion from different domains. Based on a study of people aged 50 years or older in England, Kneale (2012) has identified to what extent people who were excluded from one domain had a higher risk of exclusion from another domain 6 years later. The analyses showed that exclusion from social relations more than doubled the risk of exclusion from the domains of financial products and from consumer goods, whereas exclusion from local amenities more than doubled the risk of exclusion from consumer goods at follow-up. Furthermore, exclusion from civic activities raised the risk of exclusion from social relations, and exclusion from financial products raised the risk of exclusion from cultural activities.

In summary, although there are individual studies of multidimensionality and the interrelationships between different domains, the amount of research with such focus is limited. In particular, there is a lack of longitudinal research and research covering different country settings.

4 Outline of This Section

This section includes four chapters that contribute to previous research by examining interrelationships across social exclusion domains. Chapter 23 focuses on exclusion among older adults who have lost part of their independence, namely those living in long-term care institutions. Here, social exclusion can be embedded in the culture of institutional care, such as via a task- rather than person-oriented approach, and an imbalance of power between residents and staff. In this chapter, interrelationships across three domains of social exclusion are examined: socio-cultural exclusion, exclusion from civic participation (separating civic exclusion out into the two merged domains – see chapters by Walsh et al. and Torres in this volume) and exclusion from social relations.

Chapter 24 examines the critical relationship between exclusion from material resources and loneliness as an outcome of exclusion from social relations. By isolating the effects of material deprivation on loneliness, the analyses show a strong and significant association between material deprivation between both the level of loneliness and increase in loneliness over time.

In Chap. 25, focus is placed on how transport influences the multidimensionality of social exclusion and cuts across most of the exclusion domains, not just access to services. It is argued that transport is not only a matter of individual choice but also a societal/structural issue and that age-related mobility – just as disability – is a construct of the intersection of demands and lack of support in the environment.

Finally, Chap. 26 focuses on perhaps one of the most extreme forms of social exclusion: homelessness. Through a review of gerontological research journals, the authors show that homelessness among older adults is related to all domains of social exclusion. While research usually considers interrelations across several domains, exclusion from services, amenities and mobility is the domain receiving most attention in research on homelessness.

While these chapters were written prior to the outbreak of the COVID-19 pandemic, the handling and consequences of the pandemic have made the topics examined even more pressing. Those excluded from social relations, material resources and/or neighbourhoods have not always been able to follow recommendations on social distancing, making them particularly vulnerable to the virus, while others face an increased risk of social isolation and loneliness. For example, physical distancing is difficult to achieve for people living in homelessness; older adults in institutions have become more dependent on staff to meet their social needs; and people living in the community have become dependent on paid or unpaid help from others.