The Practical Democracy Podcast. A podcast about practical change, with guests from public bodies and civil society.

Episode 19: COVID Stressors and Political Participation

Featuring: Dr Luca Bernardi

What stressors did COVID-19 cause, and what was their impact on people's political participation?

Dr Luca Bernardi is a senior lecturer in politics at the University of Liverpool. He and Ian Gotlib, a psychologist and professor at Stanford University, have been working on a piece of research analysing things like fear about one's health, financial worries, and restricted social contact, and their impact on how people feel about politics.

The research so far has thrown up some fascinating findings and a few surprises. In this episode, Dr Bernardi talks about the project, how it was conducted and how COVID-19 has impacted the public's feelings of trust in government and other political behaviour.

Episode 19 transcript

Luca Bernadi: And here what the data is telling us is that it’s more of a story of obviously being worried about yourself and your family and friends, but also a story about restrictions. Which is not surprising, given what we’ve been hearing now in the past few years.

Sabine Groven:  Hello and welcome to the Practical Democracy Podcast by Delib. My name is Sabine Groven, and I will be speaking with some great movers and shakers working to make practical change today. On this episode we’ll be talking about we’ll be talking about studies of political science and psychology, looking at the effects of Covid 19 stresses on political attitudes and support. I’ll let today’s guest introduce himself.

LB: Luca Bernadi, Senior Lecturer in Politics at the Department of Politics at the University of Liverpool.

SG: Luca Bernadi has, together with Ian Gotlib, published a study titled Covid 19 Stresses, Mental, Emotional Distress and Political Support. So firstly Luca will talk you through their starting point and their hypothesis.

LB: So the idea that we started with in mind when we started working on this project together with Professor Ian Gotlib from Stanford university, he’s a psychologist and so we try to combine efforts and join forces between political science and psychology. And so the idea that we started with was that Covid stressors might influence mental health, which in turn influences political attitudes and political support. And so the initial idea was that, we took it from psychological studies about the stressful narrow ability hypothesis, because we know that life stressors are a common risk factor for mental health problems. So that was our starting point, Covid as a life stressor. So this would sort of the idea between, the link between Covid and mental health.

The second part of this model, let’s say working model that we’ve been developing, is actually linking mental health and political attitudes. There’s been a lot of work recently in the past, let’s say at least 5 years or so, linking mental health problems and political engagement in general. So political participation but also political attitudes. Which generally points towards a sort of negative picture, that mental health problems either are negatively associated with political participation and especially voting. And with political attitudes, meaning that for instance people with mental health problems, especially depression, there’s been a lot of work on depression in the past few years, and looking at depression for instance and trust in Government or feelings of representation, and satisfaction with political objects in general. And they all point in a negative direction. So that’s the sort of second link, the one between mental health and political attitudes.

Obviously the idea is also that Covid 19 stressors might have, themselves, an impact on political attitudes and political support. And this was, there was pretty much in the recent literature on Covid, some competing hypothesis if you like. So some scholars were pointing towards these rally-round-the-flag affects. What does it mean? It means that the citizens, they sort of tighten bonds amongst themselves in response to a common external threat, which is Covid. And they abide within approved institutional responses, which in turn should lead to an increase in trust, for instance in Government, or increase trust in political support in general. That would be one hypothesis.

But actually there’s also another hypothesis which comes from political psychology, and more from the research on worry, anxiety and stress. How anxieties and worries can actually facilitate information seeking and information processing. So the idea here would be that Covid 19 stressors in a way would foster blame attributions, by facilitating information seeking and processing. Which in turn would instead point to a different outcome, which is a redaction or a negative effect on political support.

And so putting these three things together, the model that we’ve been working with is actually this one, the Covid 19 stressors through mental distress might then influence political attitudes. So that’s where we’ve been trying to - the puzzle that we’ve been trying to solve in the last sort of couple of years pretty much.

SG: So how did the research work? Who took part and what did they ask people?

LB: Let me tell you something about the data that we’ve been actually collecting since August 2020. The sample is a representative sample of British people. And so we’ve been able to conduct four surveys actually. So one in August 2020, one in March 2021, and these are independent surveys. And then thanks to the University of Liverpool I managed to conduct a longitudinal study from February 2022. And actually the fourth survey is undergoing right now, and people are taking the survey right now. And so there’s going to be a follow-up now in May.

So the evidence that I’m able to use comes from cross-sectional surveys, but also longitudinal surveys. So cross-sectional surveys in this sense can give us a sense of what’s going on, in terms of relationships between these factors that we’re looking at. But also try to see obviously overtime if there are meaningful changes, in terms of for instance an average increase in, for instance symptoms of mental distress for example. And obviously longitudinal studies instead, they would help us more, try to understand the impacts of Covid stressors on mental health and political attitudes.

So having said, so the questions that we asked in the survey, so I can give you a little bit of context about this. By Covid 19 stressors I mean all those worries and anxieties related to Covid. So, you will catch Covid and you will die, or that something like this will happen to your family or your friends. Worries about your financial situation, even worries about losing your job. And worries about the negative impact that Covid has on society in the longer term. So these kind of questions.

And the other set of questions that we’ve been asking are more related to perceptions of anti-pandemic measures. And so we ask people, “How much are you stressed about wearing a mask in public spaces?” Or about restrictions on leaving your home. Or reduction in social contacts.

So that’s two sets of Covid stressors that we’re working with. And it’s interesting because some of them are pointing in different directions when it comes to studying the relationship with, for instance political attitudes. And so in terms of attitudes what we’ve been asking people about is about their trust in Government, their satisfaction with Government performance on Covid 19. And their internal political efficacy. So their own perception of understanding of politics, their self-efficacy in politics. But also their perception of the responsive of the political system, so these feelings of feeling represented by the system. And actually in the last survey we’ve been able to ask as well about satisfaction with democracy.

So what’s the story here? Let’s say that there’s a more coherent story between Covid 19 stressors and mental health. And so we’ve been asking about symptoms of depression, symptoms of anxiety and symptoms of stress mostly, by using validated scales of depression, anxiety and stress.

So the evidence that comes out of the cross-sectional surveys, so when we look at the evidence, in terms of associations between these factors, what’s interesting, I think that it’s worth mentioning here, is that obviously there’s a negative association between Covid 19 stressors and symptoms of mental distress. And we’re seeing this in August 2020, we’re seeing this in 2021, and now in February 2022.

But what was interesting here is that, maybe perhaps people sort of got more accustomed to anti-pandemic measures, and so these effects between Covid 19 stress factors due to anti-pandemic measures, they’re still there on mental health, but they’re sort of effect size is reduced if you like. And it’s not actually the case for Covid, those worry factors that I mentioned before, they actually, this year the effect actually even increased.

And when it comes to relating Covid 19 stressors and political attitudes, so for instance trust in Government or external political efficacy, so these feelings of being represented or satisfaction with the way the Government is handling the pandemic, there was actually something important going on with trust and satisfaction this year compared to last year. Meaning that whereas last year Covid, was probably because of this rally-round-the-flag effect that I mentioned at the beginning, that there was not such a reduction in political trust. Which is something instead that we’re finding here in February. And this is actually in a way makes sense, because if we have a look at the average levels of trust in the survey, trust went down of actually one point between last year and this year. And in a way this is a first set of evidence that we’re finding.

So pretty much Covid 19, especially worries around Covid 19, they are negatively associated with feelings of representation, trust in Government and satisfaction with Government. And so this is in terms of the cross-sectional evidence.

When it comes to trying to understand better the impact of Covid stressors on mental health and political attitudes, so we tried to do this by looking at longitudinal data. So what we did was to interview the same people twice so far. So people that were interviewed in March 2021, they took the survey again in February 2022. So this way we’re able to understand what changes within persons. Because this can tell us much more, in terms of the impact that Covid stressors have been having. And there’s two things here to mention.

So one is related to Covid stressors and mental health, right. And what’s interesting here, when we look at a different set of stressors, being worried about your life and your family and friends, and reduction actually in social contacts and restrictions in leaving your home. These ones are mostly the stressors that have a negative effect on depression, anxiety and stress. And there’s also something going on with finances, worries about your financial situation, which instead is related to an increase in stress. So I think here what the data is telling us is that it’s more of a story of obviously being worried about yourself and your family and friends, but also a story about restrictions. Which is not surprising, given what we’ve been hearing now in the past few years.

But when it comes now to try to understand this impact on political attitudes, and what’s interesting here is that some stressors point in some directions and some other stressors point in some other direction. And so when, for instance when it comes to worries about your own life and your friends and your family, about catching Covid or being seriously ill or may die. Actually the sort of, an increase in these stressors is related to an increase in external political efficacy and trust in Government especially. Meaning the more you’re having these feelings the more you trust the Government, or the more you’re feeling represented in a way. And this actually goes back a little bit to this rally-round-the-flag effects probably that I was mentioning at the beginning.

But this is actually, there’s a counter affect here going on, which is the one related to other kinds of worries, which are more related to financial situations, the financial impact of Covid on your family and your household. But also the negative, the potential negative longstanding effects of Covid on society. So these kind of stressors instead they reduce trust in Government and these feelings of representation. And actually as well the satisfaction with the economy.

So long story short, some Covid factors are more, actually there’s a positive relationship with for instance trust and feelings of representation. But some other Covid stressors point in the opposite direction. What’s interesting here is that when it comes to those other stressors related to anti-pandemic measures, we don’t find any evidence that for instance restrictions on leaving your home or reduction in social contacts reduce for instance trust or feelings of representation again. And so there are some effects of - usually on, that we’re finding on Covid stressors, but some others they are not really there. I think that this is important because then it would help actually policy makers to try and tackle what are the important stressors that might have an impact on political support in this case.

So when it comes to the effect of health on political participation scores have been working with a number of hypotheses this year.

So one is about that obviously if you are - health would be one of the resources for participation. Like many other resources that people have, like time or motivation or money for instance. And usually people who have much more of these resources they tend to participate in politics more, and so health can be seen as a resource as well. And obviously if you’re in poor health the idea here is that you would participate less in politics because you lack - when it comes to mental health especially, the cognitive resources to participate. And the motivation, obviously if you’re depressed for instance, for sure. But also physical strength for participation, because it costs you to go out and participate.

So this kind of hypothesis here has been actually tested and it found support when it came to voting, and also because perhaps people don’t necessarily think that their votes might make a difference because obviously voting is, at least from a rational choice view point, we’ve only ever a very, very tiny chance to influence what’s going on in the political landscapes by voting. And so this can be a reason for why as well people, for instance in poor mental health like depression, they tend to vote less because they lack the self-efficacy and motivation to participate in this sense.

But there are also some other hypothesis that might point in different directions. So colleagues from Finland, like Mikko Mattila and Lauri Rapeli and Konstantinos Papageorgiou.

But even other colleagues working on disability and political participation, like my colleagues Dr Stefanie Reher from Strathclyde. So they’ve been working with also some other hypothesis there which are more about actually being in poor health might even motivate you to participate more in order to change things. Or this can also be because of some group identity, so you belong to a certain group. I knot that with mental health problems this is much more complicated to establish, because obviously when it comes to symptoms of mental distress they come and go and the cycle, for instance of depression, changes so it’s really difficult to make boundaries, in terms of group identities in this sense. But there might be something going on with this anyway, motivation to change things and so participate more.

And when I tried to test these things with the data that I have, the negative effect on voting is there. But essentially some other forms of participation, for some other forms of participation I found actually a positive relationship between symptoms of mental distress like depression, anxiety and stress, and participation. So just to give you an example, for instance people who have more severe symptoms of depression, during the pandemic at least, they’ve been posting more things online about politics, or boycotting products as well and signing petitions.

So there was an increase in these activities during the past year for people who experienced higher symptoms of mental distress, whether this is anxiety, depression or stress. And so it was, it’s actually interesting again that there might be, on the one hand a negative effect on participation when it comes to, especially for voting. And for all, perhaps all those forms of, those physical forms of participation if you like, which is actually something that other colleagues of mine Christophe Jaffrelot for instance and Claudia Lang they actually they found actually depression has a negative impact on those physical forms of participation, like voting for instance. And instead it wasn’t the case for non-physical forms of participation. In a way this echoes also these previous findings from previous research.

SG: What can Luca tell us about the effects of mental health on political participation?

LB: Especially when it comes to people with mental health problems, it’s tough for them to participate in politics on the one hand, at least when it comes to voting and being heard in this sense. But it’s also more difficult for them to participate in surveys, and so it’s much more difficult therefore for policy makers to know what people experiencing mental health problems want, in terms of what are their preferences, what are their priorities. And so this obviously has implications for representation.

But what was also interesting is that with this data I was able to go beyond depression and see also what’s going on with other mental health problems, like anxiety for instance. And again the mental health voting up is there. Just to give you an example, when we look at for instance differences between the top 10 other - on the distribution, the ones that have experienced more severe symptoms of anxiety, stress and depression, and the ones at the bottom, so the ones who experienced very low symptoms or no symptoms, the gap in voting there is actually 7%, which is in line in a way with what other research has been finding about depression. But it’s interesting that this applies to other common mental health problems as well.

SG: So did Covid effect political attitudes?

LB: So from the evidence that so far we’ve been gathering, and in general not only from this project but also from other resources that we will be conducting on depression for instance, and political efficacy and voting, but there might be two things here that it’s worth mentioning.

When it comes to this research on Covid effects right, on political attitudes, then there’s definitely something going on with mental health being a mediator between Covid stressors and political attitudes, and some of those political attitudes that I mentioned earlier. And so obviously the idea is that perhaps if we can improve mental wellbeing and reduce mental distress, then in turn this might have beneficial implications for some of those political support factors that I mentioned before, such as trust in Government or feelings of representation for instance. So that’s the first thing to consider.

And when it comes to the, at least the relationship between mental health and political participation, but especially voting, because especially a consistent negative finding is the one on voting, right. What we’ve been doing and what we’ve been discovering is actually that tackling feelings of representation might actually have beneficial effects on voting. At least that was the evidence that we gathered from our research on depression, because we know that depression, that people with depression have both a lower internal and external efficacy, meaning they both have a lower self-efficacy in politics and a lower - a more negative perception of the responsiveness of the system.

But what was interesting there is that it is this negative perception of representation, or responsiveness, that helps explain why people with depression tend to vote less. And so for instance we can use this evidence to point towards some of those political attitudes they might - so if increase, for instance in this sense feelings of representation, that it might be a similar story with trust in Government. Then you might have potentially positive effects on, eventually on voting. Obviously we need to see whether this is the case also for other common mental health problems than depression. But this is certainly an important starting point.

When it comes to those, for instance when it comes to depression, to those negativity biases associated with depression, what was very interesting in some other research with Rob Jones, that we’ve been finding all the time negative effects of depression on satisfaction with evaluation of a bunch of political objects, like democracy or government or the economy, or health services, education as well. And so the idea here is that if we perhaps tackle all those attitudes, in trying to improve depressed people for instance, feelings of representation but also their assessment and evaluation of political objects, then it might, potentially perhaps might have implications in a positive way on voting.

There we have it. Thank you so much to Dr Luca Bernadi for joining the podcast and sharing his research with us. If you want to follow the work Luca does you can find him on Twitter @Bernadi_L. I have added links to his research in this episode’s description if you want to find out more.

Thank you very much for listening. Should you wish to contact us you can Tweet us @DelibThinks or email

This episode is hosted, edited and produced for Delib by me, Sabine Groven, our Creative Director is Tifanny Maddox. I’ll be back next month with another episode. Until then, you can visit for great content on people making practical change improving democracy. Bye.