Curb your assumptions about COVID noncompliance

Curb your assumptions about COVID noncompliance

Last week, as COVID-19 case numbers continued to rise in Sydney despite weeks of tight restrictions, a group of psychologists published a paper in PLOS One. Based a survey of 1,575 people across four countries, the paper aimed to gather psychological data on those who complied, and didn’t comply, with COVID-19 rules.

The results were rich, with detailed information on personality traits that were shared by different groups of compliers and non-compliers. The paper was immediately, unexpectedly, popular online. An article describing the study received over 30,000 upvotes on Reddit. Also unexpectedly, to the researchers at least, were the headlines that accompanied it: phrases like “Selfish male extroverts”, “Self-centred men”, “Older males” and “COVID rule breakers”  appeared on news sites everywhere.

Ironically, a lot of this coverage achieved the opposite of what the study aimed to do: rather than characterise the messiness of the COVID non-compliant people, they labelled them as selfish, stupid, and mostly male.

“What our research at the time demonstrated was that non-compliance has a complex face,” says Sabina Kleitman, associate professor of psychology at the University of Sydney, and lead author on the paper.

“There are real differences within each group. For non-compliance, there are huge individual differences and huge differences in motivation.”

The researchers divided responses into four groups using a technique called latent profile analysis. One group was non-compliant, while the other three were compliant, but with different backgrounds, ages, and reasons for following COVID rules.

Ironically, a lot of this coverage achieved the opposite of what the study aimed to do.

While a slight majority of the 10% of non-compliers in the sample were male, and on average they were more likely to be extroverted, and less likely to be cooperative and considerate, each of these signifiers came with large error bars.

The clearest example is age. The non-compliant group had a mean average age of 29.11, making it the second-youngest group. This was highlighted in some headlines as “older”, which isn’t accurate. And as Kleitman points out, “there are quite large standard deviations”. Standard deviation – a statistical measure that shows how much a sample varies – appears all over scientific literature, but rarely makes it into the 24-hour news cycle. It’s the start, rather than the end, of a complex statistical analysis.

It shouldn’t surprise anyone at this point, that headlines benefit from being contentious, especially on topics with as much public interest as COVID-19. A story that labels COVID rule breakers as selfish is likely to get more readers than a more accurate headline.

Dr Merryn McKinnon, a researcher at the Centre for the Public Awareness of Science at the Australian National University, who wasn’t involved with the study, describes this as a “perennial tension”.

“There is a danger in going for the punchy headline over an accurate representation of the story,” she says, “particularly when it comes to things like COVID, where people are tired, and they’re dealing with uncertainty.

“I think the tendency for stronger emotional responses is possibly higher, because we’ve been under these uncertain conditions for so long.”

And emotional responses were high. Another statistic which garnered particular attention in this paper was termed intellect, which in everyday English is usually equivalent to intelligence. The non-compliant had lower levels of intellect as a personality trait, and this became summarised as “stupid people” break COVID restrictions.

In this research, intellect had a much more precise meaning than simply intelligence.

In personality research, intellect has a much more precise meaning: it measures people’s willingness to accept new ideas and experiences as a personality trait.

“It’s a confusing term,” says Kleitman. “If you don’t read the paper, [you could] confuse it with intelligence. Unless you know psychology, it’s very easy [to confuse].”

In fact, the research showed that these non-compliant people had a lower predisposition to exploring new ideas.

“I’m not defending the unacceptable behaviour of the non-compliant group, but they’re not stupid,” says Kleitman.

“It’s a terrible thing, to call a whole group of people stupid.”

It’s difficult not to respond to the behaviour of the non-compliant with anger or alarm. Kleitman and colleagues noticed this when reading about people’s plans for their upcoming week – one of the things they were asked about in the survey.

“Particularly alarming was ‘seeing family and friends’,” she says. “Because you’re putting your loved one at great risk. This is what happened back then. And I’m worried that it’s still happening. Looking at the news, my worry is that doing that, you’re really endangering your loved one.”

So, why do these people break COVID rules, endangering lives in the process?

“When it comes to people not complying, it’s rarely as simple as they simply don’t want to, or they don’t understand.”

“Obviously there are different reasons for compliance, and there might be different reasons for non-compliance,” says Kleitman.

McKinnon agrees. “When it comes to people not complying, it’s rarely as simple as they simply don’t want to, or they don’t understand.”

But seeking to understand resistance to COVID advice further – “is it ideological? Is it because they don’t understand? Is it because they’re afraid?” – is critical to lessening it, both with COVID and with other areas of public health, like vaccines.

“Taking a ‘we’re right and you’re wrong’ stance… isn’t productive and isn’t going to help anybody, anywhere,” says McKinnon.

Kleitman’s study, based on an online survey as it was, adds a small chunk of information to the body of research we have to work with.

“The psychological tests, they’re valid, but they’re still a fairly blunt stick,” says McKinnon. “Humans aren’t simple, we can’t be easily categorised.”

“Obviously in psychological research, you would like to have the most reliable measures,” says Kleitman. “The thing is, we’ve measured something unprecedented and novel.”

It’s also worth noting that the survey ran very early in the pandemic – April and May 2020 – something usually reported in articles describing it, but naturally left out of headlines. There are advantages and disadvantages to looking at data from so early on. One advantage is that the study was done before there were months of political rhetoric associated with following or breaking the rules.

“We hoped that at the time, there was no shame associated with different behaviours, because it was just too early,” says Kleitman.

“The thing is, we’ve measured something unprecedented and novel.”

“I was surprised that people did not have a habit of just checking the legitimacy of the source, so I hope by now that’s changed,” she adds.

Non-compliant people were less likely to use official sources of information, but the chaos of the early pandemic upended lots of people.

“Even…incredibly smart people, well-educated people, at the time, were so disoriented with this bombardment of information,” says Kleitman.

She hasn’t done the research to confirm, but she thinks people may have become more discerning in their information intake over the past year – having seen the consequences of operating on the wrong facts. “I think people are much better educated now than they were at the time. So hopefully that’s one good side effect from the pandemic.”

“Most journalists aren’t looking to sensationalise, or twist words, or misrepresent findings,” says McKinnon.

But the sensationalised stuff often travels further. Is there a way to stop your research from being taken out of context? According to McKinnon, it helps if the researcher starts on the front foot: making sure the abstract of the paper, any media releases associated with it, media staff at their institutions, and any journalists they’re in touch with have the same idea.

“Researchers, first and foremost, have that responsibility to ensure that they are crystal clear in what the key findings and what the takeaway messages are,” she says.

And if all that fails, “the researchers can also use their platforms or outlets like The Conversation to put forward their actual point of view”.

Can you stay in control of your message that way? Probably not entirely, on a contentious and ubiquitous subject like COVID regulations.

But with good collaboration, we have a much higher chance that the right ears – like those of COVID policy makers – are hearing the nuance.

Please login to favourite this article.