Table 5.
Multivariable truncated linear regression of the misinformation beliefs in June (Round 3)a. Higher values of the outcome indicate greater support for these beliefs. Data are presented as estimated marginal mean differences (95% confidence intervals) and P values.
| Explanatory Variable | Value | Estimated marginal mean differences (95% CIs) and P values |
|
|||||||||
|
|
|
Symptom management and prevention (PC1) | P value | Causes and transmission (PC2) | P value | Immunity and cure (PC3) |
P value |
|
|
|||
| Sociodemographic variablesb | ||||||||||||
|
|
Age in years, mean (SD) | 44.6 (16.7) | −0.007 (−0.014, −0.001) | .03 | 0.005 (−0.003, 0.014) | .23 | −0.021 (−0.029, −0.013) | <.001 |
|
|
||
|
|
Female gender (vs male)c, n (%) | 909 (66.5) | −0.397 (−0.610, −0.184) | <.001 | 0.222 (−0.087, 0.530) | .16 | −0.088 (−0.341, 0.165) | .49 |
|
|
||
|
|
Education (vs high school or less), n (%) |
|
|
.51 |
|
<.001 |
|
.25 |
|
|
||
|
|
|
Certificate I-IVd | 140 (10.2) | 0.155 (−0.245, 0.556) | .45 | 0.401 (−0.109, 0.912) | .13 | 0.266 (−0.185, 0.716) | .25 |
|
|
|
|
|
|
University education | 1028 (75.3) | 0.173 (−0.121, 0.467) | .25 | −0.498 (−0.899, −0.096) | .02 | −0.051 (−0.388, 0.285) | .77 |
|
|
|
|
|
Language other than English spoken at home, n (%) | 51 (3.7) | −0.298 (−0.827, 0.230) | .27 | 0.463 (−0.254, 1.18) | .21 | 0.212 (−0.375, 0.799) | .48 |
|
|
||
|
|
Socioeconomic status quintile, mean (SD) | 3.69 (1.39) | −0.032 (−0.104, 0.040) | .01 | −0.212 (−0.313, −0.111) | <.001 | −0.007 (−0.092, 0.079) | .88 |
|
|
||
| Additional explanatory variables | ||||||||||||
|
|
Digital health literacyb,e, mean (SD) | 4.18 (0.67) | 0.105 (−0.046, 0.255) | .17 | −0.304 (−0.512, −0.097) | .004 | −0.444 (−0.618, −0.270) | <.001 |
|
|
||
|
|
Perceived public threat of COVID-19f, mean (SD) | 7.33 (2.44) | −0.027 (−0.069, 0.016) | .22 | −0.074 (−0.133, −0.015) | .01 | −0.057 (−0.107, −0.007) | .03 |
|
|
||
|
|
Not likely to get sick, n (%) | 123 (9.0) | 0.083 (−0.264, 0.429) | .64 | −0.285 (−0.781, 0.211) | .26 | 0.133 (−0.267, 0.535) | .52 |
|
|
||
|
|
Confidence in governmentg, mean (SD) | 5.52 (0.94) | 0.028 (−0.094, 0.149) | .67 | 0.117 (−0.054, 0.288) | .18 | 0.051 (−0.093, 0.194) | .49 |
|
|
||
|
|
Institutional trustb,h, mean (SD) | 6.15 (0.95) | −0.229 (−0.339, −0.119) | <.001 | −0.599 (−0.750, −0.448) | <.001 | −0.226 (−0.353, −0.099) | <.001 |
|
|
||
|
|
Social media used as a top-3 information source, n (%) | 680 (49.8) | 0.107 (−0.094, 0.307) | .30 | 0.200 (−0.087, 0.486) | .17 | 0.177 (−0.063, 0.417) | .15 |
|
|
||
|
|
Rejection of official accountsi, mean (SD) | 2.36 (0.83) | 0.172 (0.031, 0.313) | .02 | 0.451 (0.245, 0.657) | <.001 | 0.337 (0.169, 0.506) | <.001 |
|
|
||
aSample for analysis (n=1366); occasional cases of missing data for explanatory variables were not imputed due to a small proportion of missingness (3/1369, 0.2%).
bValues obtained in April (Round 1) and carried forward.
cMarginal mean differences are not reported for gender reported as “not specified” or “other” owing to small sample size but were included in the regression model.
dCertificates I-IV are tertiary qualifications; see the Australian Qualifications Framework [25].
eMean of 8 items, range: 1-5.
fLikert scale, range: 1-10.
gMean of 4 items, range: 1-7.
hMean of 3 items, range: 1-7.
iMean of 4 items, range: 1-5.