On solidarity and volunteering during the COVID-19 crisis in Denmark: the impact of social networks and social media groups on the distribution of support

As an immediate reaction to the pandemic, many states issued far-reaching lockdown measures and closed their borders. Both government and non-government welfare providers had to downscale or completely put on hold social and health support and services in order to contribute to the prevention of the spread of the virus. Consequently, a large group of citizens did not receive the support they depended on. The present paper explores how informal civil society quickly mobilised to provide support and fill the void that emerged when state, market, and formal civil society welfare providers locked down. This includes satisfying existing support needs as well as catering to new needs that arose during the lockdown such as, among other things, support in relation to homeschooling, shopping, transport, and comfort for people who experienced mental stress while being in isolation (Andersen and Kirkegaard 2020; Christensen et al. 2020; Toubøl and Carlsen 2020). In so doing, the paper offers insights into the organisation and distribution of informally organised support during the COVID-19 pandemic and contributes to the growing literature that engages with this topic by offering a perspective from Denmark to existing research into the German (Koos and Bertogg 2020) and Chinese cases (Woodman 2020).

Throughout the paper, we distinguish between, on the one hand, support provided by state and formal civil society actors, including formally organised voluntary work (Red Cross etc.) and, on the other hand, support and help organised in informal civil society. In doing so, we build upon existing research on volunteering that stresses the importance of mapping both informal and formal help (e.g. Boje 2017; Grubb and Henriksen 2019). In this context, the COVID-19 pandemic may be considered a paradigmatic case that offers unique insights into the dynamics between (the distribution of) welfare provisions by formal state and non-state actors, on the one hand, and (the distribution of) welfare support offered by informal voluntary civil society, on the other hand. Thereby, the paper contributes to the scholarly debate about right-based access to welfare goods and services (von Essen et al.2019; Kirkegaard and Andersen 2018) versus a welfare system that relies on volunteers for the provision of goods and services (Henriksen et al.2015; Sivesind and Saglie 2017). This includes considerations about what role governments may attribute to informally organised civil society actors. This is a particularly pressing issue for generous welfare states like the Scandinavian ones that face the challenge of dealing with the economic and social consequences of COVID-19.

In this paper, we focus on two types of informal society and their role in the mobilisation of support in response to the COVID-19 crisis: social networks and social media groups. Research on social support has found that social networks act as support networks and are vital in facilitating support for those in need (House et al.1988; Smith and Christakis 2008). Barry Wellman and Scot Wortley (1990) argue that social networks supply stable and adaptive support, especially important in crises like the COVID-19 pandemic where needs of citizens are hard to predict and require continuous attention and adaptation (Marston et al.2020). Research has also found that following Mark S. Granovetter's distinction between strong and weak ties (1973), the type of tie is important in determining whether and which kind of support is provided (Wellman and Wortley 1990; Smith and Christakis 2008). In a situation where state and formal NGO welfare provision is closed down, we would expect that social networks play a vital role in facilitating help, hence leaving those with limited or no social networks without the help they need. This raises the question if other parts of informal civil society share the capacity of informal social networks to act quickly and with flexibility in response to a crisis. Some scholars have argued that social media groups are effective in organising social support during emergencies (Albris 2018). Yet, others have argued that social media participation is often characterised by being a cheap and insincere form of activism (Lewis el al. 2014) referred to as ‘slacktivism’ (Morozov 2011) or ‘keyboard activism’ (Laer and Aelst 2010). To us, this is an open question that our results allow us to address. Initially, we provide an overview of our methods and design. We then unfold our empirical findings. First, we map the mobilisation, repertoire, and organisation of informally organised volunteering, engaging with the importance of social networks in determining whether one gets the help needed. Second, we turn to social media as a platform for organising support, exploring the amount of help coming from social media groups, the commitment from the average helper, and the type of help social media groups facilitate. Our conclusion presents our main findings and discusses how they contribute to the debate about voluntarism, welfare, and social media.

This paper reports on data drawn from a comprehensive survey on solidarity, volunteering, and support during the COVID-19 crisis conducted between 3 April and 11 May 2020 in Denmark. The survey section of Statistics Denmark (DST Survey) collected the responses from a random sample of the Danish population in the age range of 16–99 years old. The online questionnaire was distributed electronically to the inbox of e-Boks, which the vast majority of the population have and use to communicate with public authorities and institutions as well as private companies such as banks and insurance companies. The small group that has opted out of e-Boks received an invitation by post. The invitations were followed by telephone reminders offering the opportunity of being interviewed by phone.

Out of a total sample of 7964 individuals, 42.6 per cent (3389 respondents) completed the survey. The response rate compares to the general level of response rates in these types of studies (e.g. Toubøl and Frederiksen 2019). When testing for representativity, our data set displays the expected patterns of under-representation of ethnic minorities and over-representation of the elderly, people with higher education, and people with high incomes. To account for this, we weighted the reported results in terms of gender, age, and occupation.

To investigate the question of the consequences of network embeddedness for the distribution of help, we have calculated three indices. The first two indices measure strong and weak network ties, respectively. They were based on five variables measuring how often the respondents met with different categories of relations on a scale from 0 to 4 where 0 reflects not having any relations with the given class of relations and 4 means the respondents report they meet on a weekly basis. Strong ties include relations with (1) family outside the household and (2) friends. Weak ties include relations to (3) colleagues outside the workplace, (4) in religious communities, and (5) from clubs and voluntary associations. The indices were calculated as the normalised average score across the included variables (Tables A1–A3 in the appendix report the survey question and the respondents’ distributions as well as the resulting indices). In the same manner, we measure the need for help as a normalised index of the expressions of lack of seven individually measured kinds of support on a scale ranging 0–2 where zero is the value of respondents whose need for help is fully satisfied and two is the value of respondents expressing that they need help but do not receive any (see Tables A4 and A5 in the appendix for more details). In the analysis reported below, we estimate their relationship in an OLS regression model. To exclude potential confounders, we control for the sociodemographic factors of occupation, income, educational attainment, urbanisation, gender, age, family type, and whether the respondent has children (for the full list of model estimates, see Table A6 in the appendix).

We also analysed data collected from Facebook. Employing an inclusive search strategy used jointly with researcher validation, we carried out what comes close to a complete and comprehensive mapping of all Danish Facebook groups that offer support specifically related to the COVID-19 pandemic. Those groups that have COVID-19-related support as their sole purpose constitute an ethnographic field in which we have carried out online fieldwork with the acceptance of the administrators who were informed of our purpose as researchers upon our applying for membership of the groups. Also, we collected qualitative interviews with group administrators and participants. The ethnographic and qualitative part of the study allowed us to enquire into the issues raised in the survey. Furthermore, they offered comparative insights into support offered via social media and support offered through other means.

These numbers pertain only to the online mobilisation, which is by no means the largest organisational setting of support. Previous research indicates that approximately half of the population regularly provides help on an informal basis (Boje 2017). Hence, the degree of informally (as well as formally) organised support was quite extensive before the COVID-19 pandemic. According to our survey data, roughly 53 per cent of the population provided support. Given that COVID-19-specific needs for support constitute new additional needs, the total sum of support provided is likely to have increased quite substantially during the crisis. In support of that, results from a German survey study by Sebastian Koos and Ariane Bertogg (2020) suggest that not only has the total sum of support activities increased but also the number of helpers. They find that around a quarter of the COVID-19 helpers are newcomers who did not provide support before the crisis (Koos and Bertogg 2020: 3). It appears the COVID-19 lockdown instigated a substantial release of additional support capacity.

The repertoire of the support offered varied greatly. The provision of information was the most dominant type of support with c. 28 per cent of the population participating. We detected a significant stream of information about the virus itself, on behavioural aspects, for instance, about what precautions to take to avoid infection, on the lockdown and gradual reopening of society, economic aid packages, and more. Also, one in five provided help and assistance to individuals and groups who went into isolation during the COVID-19 crisis. Finally, one in 10 provided economic or other forms of material support, supported families or individuals who suffered during isolation, or assisted families with children who needed homeschooling. In total, the repertoire of types of support cover (1) information sharing; (2) caretaking; (3) practical, material, and economic support; and (4) donations.

In sum, our findings indicate that, during the COVID-19 crisis, a group of citizens did not receive the support that they felt they needed. This is very likely the case because the group depended on the support offered by organisations that went into lockdown during the COVID-19 pandemic. Lacking integration in informal social support networks, this group of individuals had few if any places to turn to for help. In relation to this particular group, our research suggests that the welfare provisions, health services, and social support usually provided by formal state, market, and civil society actors is not immediately replaceable by informally organised social support networks because not everyone is connected and has access.

The COVID-19 crisis revealed an innovation in informal civil society providing an organisational setting where access to support does not depend on a personal network, namely social media support communities. We offer two important insights into the role of social media as an organisational setting for support. First, our findings show that social media such as Facebook played a non-trivial role in the mobilisation of informally organised support during the COVID-19 crisis. Figure 2 revealed that 10 per cent of those who offered support did so via social media, rendering it the third-largest organisational setting. Before the COVID-19 pandemic, we expected 2–5 per cent of volunteers in Scandinavia to use social media (Eimhjellen 2019). From this perspective, 10 per cent is a relatively high share. This underlines the point that social media constitute a useful mobilisation technology in times of emergency and crisis (Birkbak 2012; Albris 2018; Carlsen, Ralund et al. 2020; Carlsen, Toubøl et al. 2020).

These findings bear some similarities to Ivar Eimhjellen's recent study (2019) that found information sharing to be the most common activity online. However, the fact that in most regards the repertoire of online volunteers does not seem to differ significantly from that of other organisational settings suggests that online volunteering, at least during times of crisis, can facilitate a much more varied set of activities. This is in line with findings on mobilisation of the Danish refugee solidarity movement in 2015 (Gundelach and Toubøl 2019; Toubøl 2019).

While these results demonstrate that social support facilitated through digital networks such as social media groups is non-trivial, they do not clearly establish that social-media-organised support works as a bridge facilitating help to those without a network. The affordances of social media, especially Facebook groups, clearly allow citizens to join a group quickly and post a request for help. Furthermore, most of the 247 Facebook help groups that we found had the goal of helping all those in need and not only family and friends. They all had the intention of matching those who wanted to help with those in need of help, thereby overcoming the constraints of prior social networks. This interpretation receives further support across the 12 interviews conducted so far with administrators. For instance, one of the administrators explained the purpose of the group as follows.

In line with prior research, our results point to the importance of social networks in providing social support (House et al.1988; Wellman and Wortley 1990; Smith and Christakis 2008), especially in crisis situations like the COVID-19 pandemic. Our data indicate that social networks under the pandemic are effective, attentive, and adaptive suppliers of help to the evolving needs of vulnerable citizens. Importantly, we found that not all ties matter equally. Citizens who lack strong ties are most likely to have an unsatisfied need for help.

To investigate alternative ways of facilitating help that are fast and adaptive, like social networks, but differ from these networks by not leaving behind those without strong ties, we turned to social support organised in social media groups. Here we find evidence that goes against the slacktivism theory of digital participation (Morozov 2011; Laer and Aelst 2010). Our data indicate that, in total amount as well as kind of social support, social media facilitated support played a non-trivial role in Danish civil society's response to the COVID-19 crisis, and the level of social media facilitated activity was notably higher than what has been observed in non-crisis times (Eimhjellen 2019). One possible explanation for this divergence is that social media groups are used effectively primarily in crises or emergencies where citizens feel an urge to respond quickly and anticipate too many bureaucratic obstacles in joining established NGO's.

Our research shows that it was clearly the intention of most social media groups to facilitate help to those without strong support networks. Despite these organisational intentions and the technological affordance, there can be several psychological, organisational, and social barriers that prevent the vulnerable and isolated persons from seeking help in a social media forum: fear of stigmatisation, lack of trust, excluding codes of conduct, perceived (il)legitimacy of support need. Further research is needed into the barriers and enablers of social media social support.

Our findings give rise to the question of whether the changes that we have witnessed during the COVID-19 crisis are contingent upon and limited to the crisis or if do they extend beyond it. The COVID-19 pandemic has led to an increased awareness of the substantial capacity for support in informal civil society as well as the role of social media therein. This awareness is not confined to civil society actors but also extends to state actors. They are likely to take a pronounced interest in these informally organised ways of providing support in a period where they are under pressure to respond sensibly to the economic and societal consequences of the COVID-19 crisis. However, it also gives rise to caution. Our research shows clearly that the bulk of informally organised support flows through existing social networks and, therefore, is not available to the unconnected. It is thus likely that this particular group of citizens will continue to depend upon welfare provisions from formal state, market, and civil society organisations in the years to come.

We are grateful to our research assistant Mia Lunding Christensen as well as student research assistants Esben Brøgger Lemminger, Marie Haarmark Nielsen and Jonathan Holm Salka for their skillful work on collecting and preparing data for this study.

No potential conflict of interest was reported by the author(s).

Benedikte Brincker, Ph.D., is the Head of Department at Department of Sociology, University of Copenhagen. She has published widely in the field of political and historical sociology, especially in the area of state-nation relations. She has applied this body of theory to European nation states and to the Arctic Region and has combined it with research in the field of sociology of education and public administration.

Jonas Toubøl, Ph.D., is a political sociologist at Department of Sociology, University of Copenhagen. His area of research encompasses social movements, civil society, unionization and social class. In addition, Jonas explores new approaches to mixed methods and data triangulation of both quantitative and qualitative data combining interview, ethnography, register, survey and social media data.

Hjalmar Bang Carlsen, Ph.D., is a postdoctoral researcher at SODAS and Department of Sociology, University of Copenhagen. His research interests are political sociology, especially social movements and issue politics, and digital social research methodologies that combine quantitative and qualitative approaches in the study of social interaction and social change.

The tables below report background information for the variables and regression model of the relationship between individual level of weak and strong network ties and the level of unsatisfied needs for support during the COVID-19 crisis, depicted in Figure 3. Tables A1–A3 concern the survey question formulation, the distribution of the responses and the distribution of the indices of network ties. Tables A4–A5 concern the survey question formulation, the distribution of the responses, and the index the respondents’ perception of whether they received the support they needed. Finally, Table A6 reports the estimates of the OLS regression model of the relationship between the index of unsatisfied support needs as the dependent variable and the indices of weak and strong network ties as the focal independent variables.