Conscientious individuals were more likely to maintain social distance and less likely to contract COVID-19; Agreeable individuals were more likely to comply with social distancing requirements yet more likely to contract COVID-19

Personality and early susceptibility to COVID-19 in the United Kingdom, Satoshi Kanazawa. Journal of Community & Applied Social Psychology, October 27 2021. https://doi.org/10.1002/casp.2578

Abstract: This paper takes advantage of a unique dataset with a prospectively longitudinal, nationally representative sample (n = 5,178) that began in 1958 and has information on COVID-19 health status in 2020 to examine the effect of Big Five personality traits on compliance with social distancing requirements and contraction of COVID-19. The results show some consistency with epidemiological recommendations (Conscientious individuals were more likely to maintain social distance and less likely to contract COVID-19; men were less likely to comply and more likely to contract) but more inconsistency (Agreeable individuals were more likely to comply with social distancing requirements yet more likely to contract COVID-19; Open and Neurotic individuals were no less likely to comply yet more likely to contract COVID-19). The results highlight the importance of Big Five personality factors for behaviour in the global pandemic and may call into question the universal effectiveness of social distancing requirements for all individuals. However, the small number of confirmed cases of COVID-19 during the early months of the pandemic requires caution in interpretation of the results. 

3 DISCUSSION

To the best of my knowledge, this is the first study to examine the effect of Big Five personality factors on the actual contraction of COVID-19, rather than compliance with various government mandates (social distancing, handwashing, etc.) or attitudes towards COVID-19. It is also the first study of personality and COVID-19 conducted in the United Kingdom. Taking advantage of an ongoing, prospectively longitudinal study with a large, nationally representative sample in the United Kingdom, with information on the respondents for their entire lives (62 years), I examined the association between Big Five personality factors (measured 11 years prior) and compliance with the government mandate on social distancing as well as actual COVID-19 health status and symptoms.

The analyses of the National Child Development Study (NCDS) data replicated earlier findings on the positive effect of Conscientiousness and Agreeableness on compliance (Abdelrahman, in press; Blagov, 2021; de F Carvalho et al., 2020; Götz et al., 2021; Zajenkowski et al., 2020), but did not replicate the earlier findings of a positive effect of Neuroticism (Abdelrahman, in press; Blagov, 2021; Götz et al., 2021) or a negative effect of Extraversion (de F Carvalho et al., 2020; Götz et al., 2021). In fact, Extraversion was the only Big Five factor that was not significantly associated with any of the dependent measures examined here.

Just as any study that uses a longitudinal cohort dataset, sample attrition and selection bias are potential problems in the current study. For example, NCDS respondents who participated in the May 2020 COVID-19 survey were healthier at Sweep 9 in 2013 than those who did not. On self-perceived health (1 = poor, 2 = fair, 3 = average, 4 = good, 5 = excellent), 4.4% of those who participated in the 2020 survey rated themselves “excellent” compared with 3.6% of those who did not, and .9% of those who participated in the 2020 survey rated themselves as “poor” compared with 2.9% of those who did not. The mean self-perceived health was significantly higher among the 2020 participants than non-participants (3.51 vs. 3.16, t = −15.848, p < .001). Thus, it is possible that some of the Sweep 9 participants who did not participate in the COVID-19 survey may have suffered (or even died) from COVID-19. However, this particular problem is not specific to the COVID-19 survey. NCDS respondents who participated in Sweep 9 survey in 2013 were healthier at Sweep 8 in 2009 than those who did not, and the difference was greater in 2013, when there was no global pandemic; 18.4% of those who participated in Sweep 9 rated their health “excellent” compared with 1.0% of those who did not, and the mean self-perceived health was significantly higher among the Sweep 9 participants than non-participants (3.53 vs. 3.16, t = −11.537, p < .001). So non-random sample attrition by health appears to be constant and not specific to the times of global pandemics. However, the fact that the NCDS sample consists entirely of a single age group (age 62 in 2020) may potentially limit the generalizability of my findings to younger or older populations.

Perhaps the most striking finding in the analyses presented above was that many of them were seemingly inconsistent with the public health recommendations from epidemiologists and the legal requirements imposed by the government throughout the world (including the United Kingdom). Citizens in all nations were and still are required to maintain social distance in order to prevent the spread of the coronavirus and reduce the risk of infection. Some of the results above were consistent with this recommendation and requirement. More Conscientious individuals were more likely to comply with the requirement to maintain social distance, and they were less likely to contract COVID-19 and suffer from fewer COVID-19 symptoms. Similarly, men were significantly less likely to comply and significantly more likely to contract (albeit with no more symptoms). In sharp contrast, however, more Agreeable individuals were more likely to comply with social distancing requirements yet at the same time more likely to contract COVID-19 and suffer from a larger number of COVID-19 symptoms. In addition, more Open and more Neurotic individuals were no less likely to comply with the social distancing requirements, yet they were more likely to contract COVID-19 and suffer from a larger number of COVID-19 symptoms. The results suggested that the public health recommendations may not be universally effective for all individuals at all times and may instead need to be tailored for different individuals of varied personality types.

However, the very small number of confirmed cases of COVID-19 in the NCDS data (n = 19; .4%) during the early months of the pandemic in May 2020 (although the total of “confirmed” and “medically advised” cases was much larger; n = 296, 5.7%) requires caution in interpreting the results of the current analyses. More research and attempts at replication are clearly necessary to examine the effect of personality factors on COVID-19 contraction further, particularly in later months of the pandemic, to capture a larger number of confirmed cases in a nationally representative sample. Towards this end, CLS began Wave 2 of the COVID-19 survey in September 2020, and Wave 3 in January 2021.