Thread by @EricMerkley

🚨New pre-print 🚨by @PeejLoewen and I out of the @MediaEcosystem project on how prospective economic cost reduces social distancing expectations. We think this is an important one. Bear with me for a long-ish thread 👇1/

#SocSciResearch #Covid_19

osf.io/yht9v/

Citizens have been asked to take a variety of costly actions to protect themselves and others (i.e. social/physical distancing). This behaviour is essential in the absence of #TestAndTrace and a mass produced vaccine. How sustainable is this? We need more research 2/

We see public health during a pandemic as a public good to which citizens can make a costly contribution by socially distancing themselves. Participation will influenced, in part, by its marginal cost and benefit, and by expectations of other people's behaviour 3/

This is a problem because social distancing is imposing catastrophic economic costs on society. Costs/benefits of this behaviour are not evenly distributed. As marginal costs ⬆️, we expect compliance to ⬇️, especially among those who suffer higher costs relative to benefits 4/

We also expect these costs will undermine expectations that other people will participate in social distancing, which in turn may reduce compliance as well 5/

We conducted two experiments on two large, nationally representative samples of Canadians (N~2,500) at the early stages of the pandemic before the economic collapse came fully into view 6/

We find that randomly assigned information on high prospective economic costs of social distancing ⬇️our respondents' expectations of 1) social distancing compliance by other citizens (studies 1 and 2); and 2) expectations of their own compliance (study 2) 7/

In study 2, the effects of our treatment are stronger among two groups who face higher costs relative to benefits from social distancing: the young, and those outside of urban areas, after accounting for possible confounds 8/

In study 2 we attempted to test whether expectations of others' compliance mediated part of the relationship between our economic cost treatment and self-expectations of compliance. We we unable to exogenously manipulate the mediator, so we conducted a third experiment 9/

In study 3 we provide indirect evidence of mediation by showing that expectations of others' social distancing compliance ➡️ expectations of one's own behaviour.

We exploit the fact that people tend to underestimate the social distancing compliance of others 10/

We randomly assign respondents polling information (based on a previous survey) showing that large majorities of Canadians are willing to take three precautions for an additional three months or more, and one item where the public was divided 11/

Respondents exposed to the polling information sharply ⬆️⬆️their expectations of other people's social distancing compliance. They also ⬆️expectations of their own. But this was only true for the three items that indicated future widespread public compliance would be likely 12/

To sum up, prospective economic cost reduces expectations of social distancing compliance by others and oneself. This latter effect is strongest for those who suffer higher costs relative to benefits, and may be partially mediated by changing expectations in others' behaviour 13/

One key limitation should be acknowledged: we examine *expectations* of behaviour rather than behaviour itself. They are likely linked, but they are not the same thing. This is an unavoidable limitation with survey experiments. More research is needed 14/

To be clear, the implication of this study is not that "social distancing/lock downs are bad." But we can't take sustained citizen compliance for granted given the enormous costs. We *really, really* need alternative models to reduce the burden on citizens, like #TestAndTrace 15/

Also, misleading news coverage of small, unrepresentative protests or of bozos refusing to wear masks is not helpful. The expectations people have of others' behaviour is surely shaped by the media, and these expectations may, in turn, influence their own behaviour 16/

This was an ambitious project. Not everything went as planned. But we think the findings are super important moving forward, especially if we have to deal with a resurgence in the fall or when we inevitably have to manage gov't + societal response to pandemics in the future 17/

Shout out to the rest of the @MediaEcosystem team: @AengusBridgman, @taylor_owen, @derekruths, @ozhilin1. This paper is under review, but comments are welcome. More to come from our team! END/