What we found out
In this study we examined two key questions. First, did perceived neighbourhood social cohesion change during the first few months of the COVID-19 pandemic? And second, do subjective opinions about social relationships have an effect on our subjective mental wellbeing?
Changes in neighbourhood cohesion during the COVID-19 pandemic: whose social cohesion was most affected?
Overall, in the early months of the pandemic, we observed a slight decline in the share of people who felt positive
about the social relationships in their local neighbourhood,
compared to the pre-pandemic years. Importantly, however, not all groups in society experienced an equal decline in
these positive feelings. People from certain ethnic minority groups (i.e. those who identified as South Asian, Black,
and ‘Other’) were more likely to report negative changes in social relationships in their local area, compared to White
British people. Furthermore, ethnic minority people who said that racial attacks in their local area were common, were twice as likely to report a decline in neighbourhood cohesion, compared to those who did not share this view.
We also observed significant geographical variation in changes in social cohesion across communities. People living in deprived neighbourhoods and in areas with fewer formal civil society organisations per head were more likely to report a decline in neighbourhood cohesion.
In sum, the people and places who got hit hardest by the COVID-19 pandemic in different areas of life also felt more pessimistic about the social relationships in their neighbourhoods.
How stable are people’s opinions about neighbourhood social relationships?
We discovered that people’s perceptions of neighbourhood cohesion are highly changeable. Over time, people frequently altered their opinions, transitioning between more positive and more negative views. This volatility is hidden in overall measures of reported social cohesion, as the changes are averaged out over time. Approximately three-quarters of people change their views about the quality social relationships between surveys.
Do perceptions of social relationships in the neighbourhood matter for one’s mental health in times of crisis?
In short, we found that perceptions of social relationships in the neighbourhood and mental health are positively
related, but it is difficult to establish the causal direction of this relationship. In fact, our current results suggest that
perceived neighbourhood cohesion go hand in hand with subjective mental wellbeing but the changes in perceived cohesion are not associated with subsequent changes in subjective mental wellbeing. Instead, changes in both seem to happen simultaneously.
Why is this important
There is no doubt that neighbourhood cohesion is an important characteristic of the social fabric of our communities. Existing literature points out that social cohesion might be one of the important factors that positively contribute to people’s mental wellbeing. This is because social cohesion is believed to promote mutual
support and foster a sense of connection, both of which are especially important in times of crisis.
However, despite theoretical propositions implying a cause-and-effect connection between social cohesion and mental wellbeing, there is almost no longitudinal evidence that empirically tests this hypothesis. From a policy point of view, it is important to understand whether strengthening social cohesion at the neighbourhood level is likely to result in improved mental health among local residents, especially during challenging times.
How did we find it out
In our study, we used Understanding Society data from years: 2011/12 (wave 3), 2014/15 (wave 6), 2017/18 (wave 9), June 2020 (COVID-19 wave 3), November 2020 (COVID-19 wave 6) and March 2021(COVID-19 wave 8). In all of these waves, we were able to observe people’s subjective mental wellbeing as well as their subjective perceptions of neighbourhood cohesion. We restricted our sample to England only because policies related to social movement restrictions that could have affected both perceptions of cohesion as well as mental wellbeing significantly varied across the devolved UK nations.
In the analysis of the direction of the social cohesionmental health relationship, we only used COVID-19 waves and limited the sample to people who did not change their address during the pandemic. To test the direction of the
relationship, we applied a structural equation modelling framework, which allowed us to control for reciprocal
relationships between these two concepts and for autocorrelation (that is to take into account the fact that the levels of, for example, cohesion at one point in time are likely to be correlated with the levels of cohesion at the subsequent measurement point).
Next steps
The unique feature of our current study is the ability to longitudinally assess changes in the perceived
neighbourhood social cohesion and subjective mental wellbeing over several points in time. Thanks to the availability of multiple repeated data points across the same individuals over time, we were able to statistically assess the direct effect of changes in perceived social cohesion on subsequent changes in subjective mental wellbeing using a structural equation modelling framework. However, our study is limited to subjective measures (opinions expressed by individuals) and does not consider behavioural indicators or other, more objective measures of neighbourhood social cohesion. This means that we are only able to make conclusions about the likely effects of subjective perceptions of neighbourhood relationships on subjective mental wellbeing.
Future studies need consider a wider range of indicators, especially those that capture objective behaviours of neighbourhood cohesion, in order to determine a true effect of neighbourhood social cohesion on individual wellbeing.
Figure 1 Neighbourhood cohesion is strongly associated with subjective mental wellbeing
Figure 2 Trends
Figure 3 Changes in perceived cohesion for different ethnic groups between 2019 and June 2020
Figure 4 Geographical variation
© MiSoC August 2023
DOI: 10.5526/misoc-2023-004