Numeracy and COVID-19: examining interrelationships between numeracy, health numeracy and behaviour

Abstract

During the COVID-19 pandemic, people across the globe have been exposed to large amounts of statistical data. Previous studies have shown that individuals' mathematical understanding of health-related information affects their attitudes and behaviours. Here, we investigate the relation between (i) basic numeracy, (ii) COVID-19 health numeracy, and (iii) COVID-19 health-related attitudes and behaviours. An online survey measuring these three variables was distributed in Canada, the United States (US) and the United Kingdom (UK) ( n = 2032). In line with predictions, basic numeracy was positively related to COVID-19 health numeracy. However, predictions, neither basic numeracy nor COVID-19 health numeracy was related to COVID-19 health-related attitudes and behaviours (e.g. follow experts’ recommendations on social distancing, wearing masks etc.). Multi-group analysis was used to investigate mean differences and differences in the strength of the correlation across countries. Results indicate there were no between-country differences in the correlations between the main constructs but there were between-country differences in latent means. Overall, results suggest that while basic numeracy is related to one's understanding of data about COVID-19, better numeracy alone is not enough to influence a population's health-related attitudes about disease severity and to increase the likelihood of following public health advice.

Task features change the relation between math anxiety and number line estimation performance with rational numbers: Two large-scale online studies.

“…A large dataset of adults ( N = 2,124), recruited through Qualtrics panels, was used in the current secondary data analyses (Lau, Wilkey, et al, 2022). Participants from Canada, the United States, and the United Kingdom were recruited in December 2020 and were stratified by age, gender, and educational attainment (see Table A1 for demographic characteristics).…”

“…Participants from Canada, the United States, and the United Kingdom were recruited in December 2020 and were stratified by age, gender, and educational attainment (see Table A1 for demographic characteristics). After data cleaning according to preregistered exclusion criteria (for full details see “Exclusion Criteria and Outlier Identification” and “Missing Data” from Lau, Wilkey, et al, 2022), the final analytic sample was 2,032 participants. Participants were excluded when they (a) had multiple submissions, (b) indicated that they were not serious when completing the survey, (c) incorrectly answered more than one attention check question (e.g., “This is an attention check, please select answer 3 for this item;” Barends & de Vries, 2019), (d) had more than 25% missing data from selecting “prefer not to answer,” or (e) indicated that they were vaccinated or participated in a COVID-19 vaccination trial.…”

“…The online Qualtrics survey administered in Lau, Wilkey, et al (2022) took approximately 20 min to complete and included four sections: demographics, basic numeracy, COVID-19 health numeracy, and COVID-19 health-related attitudes and behaviors. First, participants completed the consent form followed by a section about demographics and other cultural variables.…”

“…Finally, participants were excluded if they provided nonsensical responses to healthrelated math problem-solving items. Because both parent studies were conducted with a Qualtrics panel of participants at similar points in time, we excluded overlapping participants (n = 30) who took part in both Lau, Wilkey, et al (2022) and Thompson et al (2021) from our analyses in Study 2. Thus, the final sample used for the analyses reported below was N = 1,790 (see Table A1 for demographic characteristics).…”

“…We conducted preregistered secondary data analyses on two large samples of adults (Study 1: Lau, Wilkey, et al, 2022; Study 2: Thompson et al, 2021). In Studies 1 and 2, we investigated the following research questions: (RQ1) Does the relation between MA and performance vary by number type (e.g., fractions, whole numbers, percentages, etc.)?…”