Table 3.
Correlation between factors (demographic, knowledge, socio-cognitive) and social distancing behaviour
| Factor | N | Mean (SD) | Median (IQR) a | Correlation Social Distancing Behaviour | |
|---|---|---|---|---|---|
| Spearman’s rs (2-tailed) [95% BCa CI]b | p | ||||
| Social distancing behaviour score | 281 | 2.60 (0.63) | 2.75 (0.5) | – | – |
| Age | 281 | 23.53 (5.75) | 22 (5) | 0.03 [−0.09, 0.15] | 0.658 |
| Gender (F/M) | 278 | N/A | N/A | −0.13 [− 0.26, − 0.01] | 0.026 |
| Ethnicity (Minorities/White) | 281 | N/A | N/A | −0.12 [− 0.23, 0.01] | 0.026 |
| Student home/Intl status | 278 | N/A | N/A | −0.03 [− 0.16, 0.09] | 0.573 |
| Health course | 278 | N/A | N/A | 0.02 [−0.10, 0.14] | 0.732 |
| Exposure | 279 | N/A | N/A | 0.014 [−0.10, 0.11] | 0.814 |
| Knowledge – COVID-19 | 275 | 2.53 (0.75) | 3.0 (1.0) | 0.06 [−0.06, 0.19] | 0.293 |
| Risk perception | 281 | 4.31 (1.56) | 4.33 (2.0) | 0.07 [−0.04, 0.19] | 0.217 |
| Self-efficacy – avoidance of infection | 281 | 6.14 (1.86) | 6.0 (2.0) | 0.12 [0.01, 0.24] | 0.047 |
| Social distancing effectiveness knowledge | 277 | 9.66 (0.72) | 10.0 (0) | 0.11 [−0.02, 0.22] | 0.075 |
| Attitudec | 277 | 0.87 (0.95) | 1.0 (1.0) | 0.16 [0.05, 0.27] | 0.009 |
| Advantages | 280 | 4.61 (0.52) | 5.00 (0.92) | 0.26 [0.13, 0.38] | < 0.001 |
| Disadvantages (reversed) c | 278 | 2.26 (0.81) | 2.33 (1.33) | 0.02 [−0.11, 0.16] | 0.746 |
| Social support | 276 | 4.40 (0.65) | 4.67 (1.0) | 0.28 [0.15, 0.39] | < 0.001 |
| Trustd | 280 | 3.89 (0.71) | 4.0 (1.0) | 0.28 [0.16, 0.39] | < 0.001 |
| Self-efficacy – social distance | 280 | 4.39 (0.66) | 4.67 (1.0) | 0.43 [0.32, 0.52] | < 0.001 |
Abbreviations: N/A not applicable, BCa CI Bias corrected accelerated confidence intervals ,SD standard deviation; IQR, interquartile range Bold: significance p < 0.05
a Median (IQR) reported as based on the Kolmogorov-Smirnov test for normality, variables were not normally distributed
b 95% BCa CI from 1000 bootstrap samples
c Cronbach α for Attitude < 0.5 so subscales (advantages and disadvantages) analysed. Cronbach α for Disadvantages subscale < 0.5 so key representative item analysed “I think I will lose my job (or have lost my job and will not be able to find other work)” – reversed - correlation with social distancing behaviour: rs = 0.09 [− 0.04, 0.21], p = 0.153
d Cronbach α for Trust < 0.5, so two key items analysed “I think the decisions that have been made to reduce the spread of the novel coronavirus are fair”- correlation with social distancing behaviour: rs = 0.21 [0.10, 0.33], p = 0.001; “I think the authorities should relax the restrictions even if there are still many new cases of COVID-19 emerging” – reversed - correlation with social distancing behaviour: rs = 0.24 [0.12, 0.36], p < 0.001