02 Jul The Darkest Hour for Behavioural Science?
According to John Hopkins University as at June 29, 2020, the United Kingdom has the world’s highest COVID-19 mortality rate.
In mid-March 2020, despite there being conclusive evidence that social distancing was required to slow the spread of COVID-19, the UK Government delayed taking such measures. According to Dr Richard Horton of the Lancet Medical Journal, ‘The handling of the COVID-19 crisis in the UK is the most serious science policy failure in a generation.’
Influential in the decision to delay a lockdown, was behavioural scientists citing ‘behavioural fatigue’. The advice to the UK Government was that people would tire of social distancing and would cease to comply possibly before the actual peak of the pandemic. Despite there being no evidence that behavioural fatigue would undermine social distancing interventions, and the horrific consequences, scientists applying hard evidence from the fields of epidemiology and public health, had their pleas for an immediate lockdown ignored.
As one of the greatest proponents of behavioural science, Rory Sutherland at Nudgestock in 2018 said, ‘the great worry of behavioural scientists is the replication crisis; a lot of experiments don’t faithfully replicate.’ In a study by Nosek, B., et al, Science, Vol 349, as few as one-third of the findings from behavioural experiments published in high-ranking journals were able to be replicated.
A poster-child from behavioural science is the paradox of choice. Barry Schwartz explains that an overwhelming abundance of choice leads to choice-overload which adversely effects purchase behaviour. Contrary to the theory posited by Schwartz, in a meta-analysis of 50 choice-overload experiments, the authors (Scheibehenne, B., et al. Journal of Consumer Research, Vol. 37) found a mean effect size of virtually zero. That is, there was no convergent evidence of the effects of choice-overload.
Perhaps no one has said it better than columnist Sonia Sodha, ‘In matters of life or death, nudge theory is a poor substitute for hard science.’ The delay in the UK lockdown has reiterated what many have already known; that is, behavioural science and its collection of interesting nudges are context specific and are most often neither replicable, nor generalisable.