Executive Summary
This report examines attitudes as they relate to COVID-19 and vaccinations. It is noted that while there
is significantly less concern among Australian respondents about the likelihood of catching COVID-19
and the impact on health if caught, the sample indicates a significantly high willingness to receive a
COVID-19 vaccination. It is possible that a campaign that emphasis the role of the role of vaccination
in helping to protect the more vulnerable in the community would have a particularly stronger impact
on promoting uptake, as opposed as focusing on the benefits to the individual, especially as there is
strong agreement among all respondents that people have a duty to protect themselves and others.
There is less trust in government, business, and pharmaceutical companies in Australia, so campaigns
should be as transparent as possible, and a clear and honest reporting of vaccine efficacy and side
effects is required. Overall, this report concludes that Australians are willing to be vaccinated and
given the increased use of the Pfizer vaccine, the biggest hurdle will likely be logistical issues.
Contents
1. Introduction: ....................................................................................................................................... 3
2. A Note on the Sample ........................................................................................................................ 3
3. Results................................................................................................................................................. 4
3.1. Health and COVID ........................................................................................................................ 4
3.2. Impact of COVID on Movement and Work ................................................................................ 5
3.3. Attitudes toward Government and COVID-19 Measures .......................................................... 6
3.4. Attitudes Towards Vaccination ................................................................................................... 7
4. Willingness to get Vaccinated ............................................................................................................ 7
5. Encouraging Vaccination .................................................................................................................. 10
6. Conclusion and Recommendations ................................................................................................. 11
Appendices ........................................................................................................................................... 14
A1. Notes on Data Cleaning ............................................................................................................. 14
A2. Attitudes towards Government Measures ............................................................................... 16
A3. Attitudes towards Vaccination .................................................................................................. 17
A4. Assessing Linearity for Regression ............................................................................................ 18
A5. Output for Initial Regression Model.......................................................................................... 22
A6. Output for Final Regression Model ........................................................................................... 25
A7. Correlation of Attitudes and Incentives for Vaccination .......................................................... 27
3
1. Introduction:
The impact of COVID-19 has been seismic, virtually eliminating international travel, placing restrictions
on more general travel and activity patterns, with the economic impact being an approximate 3.5%
contraction in global GDP – the largest fall since the Second World War1. More recently, the discovery
of viable COVID-19 vaccinations has led to hope that some degree of normality might resume. Several
countries are approaching or have exceeded half of their population being vaccinated, but Australia
currently lags significantly behind with approximately 1.5% of the population fully vaccinated2.
A recent study discovered that about one-third of Australians said they were unlikely to get
vaccinated, with caution mostly expressed due to potential side effects and a lack of urgency given
Australia's low infection rates3. A separate survey found that while most people would get a safe and
effective vaccine, eight in 10 Australians were worried about possible side effects4. Indeed, prior to
the COVID-19 pandemic, the World Health Organization declared vaccine hesitancy to be one of the
top 10 threats to public health5.
However, the recent lockdowns in Victoria as a result of another chain of community transmission
emanating from a hotel quarantine breach, may have sharpened focus on the role of vaccinations.
Additionally, community concerns may also be heightened due to the fact that 75% of the country's
deaths have occurred in aged-care facilities6 resulting in Federal Government calls for the mandating
of COVID-19 vaccination for all staff working in residential aged care7.With the role of vaccination in
protecting at risk members of the community, along with the potential of avoiding future lockdowns
as a strategy for blunting chains of community transmission, this report examines attitudes towards
COVID-19 vaccination to discover what issues might exist in the choice to be vaccinated or not, and
how willingness to consider vaccination and/or be vaccinated can be increased.
2. A Note on the Sample
The survey was completed predominantly by students in MMGT6012, with some responses coming
from family and friends. The average age of respondents is 29.1 (σ = 12.1), 73% are in Australia (the
rest overseas, mainly China), 16% felt they were from a household of below average income, 37%
average and 47% from a household with an above average relative income. 67% of respondents are
female. Given the convenience sample that was used for recruitment, it is not surprising that the
sample is biased toward younger, tertiary educated respondents, and is disproportionately female in
composition. Caution should be used in extrapolating these results to a wider population, but for the
purposes of this project we will assume that recommendations can be made (with particular reference
to encouraging younger members of society to strongly consider vaccination for COVID-19)8.
1 https://www.brookings.edu/research/social-and-economic-impact-of-covid-19/
2 https://www.smh.com.au/national/covid-19-global-vaccine-tracker-and-data-centre-20210128-p56xht.html
3 https://www.bbc.com/news/world-australia-57181038
4 https://www.abc.net.au/news/2021-05-05/covid-vaccine-hesitancy-willingness-survey/100116166
5 https://www.theguardian.com/commentisfree/2021/may/25/being-open-about-why-australias-vaccination-take-up-is-low-is-the-first-
step-to-improve-it
6 https://www.theguardian.com/australia-news/2021/may/27/operator-of-unvaccinated-melbourne-nursing-home-flabbergasted-by-
ministers-claim-residents-opted-out
7 https://www.theage.com.au/national/victoria/no-jab-no-work-on-covid-front-line-for-aged-care-staff-20210531-p57wtl.html
8 All questions were analysed for significant differences in responses based on gender, age, relative income, whether they are in Australia
or overseas, the number of restrictions experienced, and the duration of restriction experienced. Only those significant at the 5% level are
reported. See Appendix A1 for summary of data cleaning.
4

3. Results
3.1. Health and COVID
As shown in Figure 1(a) & 1(b), respondents in the sample tend to be rank themselves as above
average health (though older respondents are significantly more likely to rank themselves as below
average9) and believe themselves to be well-informed about health and medical issues.
Figure: (1A): Reported Health and Level of Knowledge about (1B)Health and Medical Issues
When asked about the risk of COVID-19, perhaps again representing the predominance of younger
respondents in the sample who are located in Australia, on average the risk of contracting COVID-19
is perceived to be unlikely10, see Figure 2. Figure 3 shows the perceived risk of COVID-19 to health if
caught. Over the sample, the risk of COVID-19 on a respondent’s own health along with the impact on
the health of an average person is thought to be low as well (not significantly different from the
neutral point of the scale), however the risk of COVID-19 to the health of a family member or a loved
one is thought o be significant (the average of 6.4 on the ten-point scale is significantly larger than the
mid-point value of 5.511). Interestingly, the average impact of health varies significantly based on
location, with those overseas perceiving COVID-19 to be a significantly greater risk across all three
scales12.
Figure 2: Likelihood of Contracting COVID-19
9 Spearman’s Correlation = 0.385, sig = 0.000. Note that higher scores on scale equate to lower levels of health.
10 One sample t-test against fixed value of 5.5 (midpoint of scale): t = -7.85, sig. = 0.000.
11 One sample t-test against fixed value of 5.5 (midpoint of scale): t = 3.765, sig. = 0.000.
12 Independent samples t-test: Own (t = 4.08, sig = 0.000); Average (t = 4.91, sig = 0.000), Family (t = 2.39, sig = 0.000).
19%
45%
32%
4%
Well
above
average
Above
average
AverageBelow
average
26%
51%
24%
0%
Extremely
well
informed
Well
informed
Informed as
next person
Not well
informed
12%
18%
28%
11%
17%
2%
5%
2% 1%
4%
Extremely
Unlikely
2 3 4 5 6 7 8 9 Extremely
LikelyAverage = 3.7
5
Figure 3: Risk of COVID-19 to Health if Contracted
Figure 4 provides a small insight into the impact of COVID-19 on wellbeing. Relative to a normal year,
respondents feel significantly more socially isolated, anxious, and stressed, but report no change to
physical activity on average13. Those who have experienced a greater number of periods of restrictions
on movement report significantly higher average levels of anxiousness and stress14.
Figure 4: Subjective Wellbeing Relative to a Normal Year
3.2. Impact of COVID on Movement and Work
Figure 5 displays the average level of concern towards beyond the direct health impact alone. Once
again there exists significant differences in the perceptions of those in Australia versus those overseas.
With the exception of the impact of COVID-19 on the economy where attitudes are the same,
Australians are significantly less concerned about the impact on the health system, their
job/education, and potential social isolation15. Indeed, while those overseas expressed significant
concern across all statements16, Australians are only concerned about the economy and social
isolation17. Females expressed significantly higher levels of concern about social isolation18, and
younger respondents expressed more concern about the impact on jobs/education and social
isolation19.
With regards to restrictions on movement, Figure 6a shows that the vast majority of respondents
experienced a government enforced lockdown, and Figure 6b displays that most have only has to
restrict movements once, but a approximately half of all respondents having to do so two or more
times. The average total duration of restricted movement is reported at 5.7 months since the start of
the pandemic. A third of respondents (33%) reported losing income because of COVID-19 measures.
13 One sample t-test against fixed value of 3 (mid-point of the scale): Isolate (t = 8.438, sig. = 0.000); Anxious (t = 9.439, sig. = 0.000);
Stressed (t = 7.157, sig. = 0.000).
14 ANOVA: Anxious (F = 4.388, sig. = 0.015); Stressed (F = 3.499, sig. = 0.034).
15 Independent samples t-test: Health (t = 2.18, sig = 0.000); Job (t = 4.27, sig = 0.000); Isolation (t = 2.20, sig = 0.000).
16 One sample t-test against fixed value of 3 (mid-point of the scale): Health (t = 3.72, sig = 0.000); Economy (t = 3.38, sig = 0.000); Job (t =
6.84, sig = 0.000); Isolation (t = 4.82, sig = 0.000).
17 One sample t-test against fixed value of 3 (mid-point of the scale): Economy (t = 4.12, sig = 0.000); Isolation (t = 3.75, sig = 0.000).
18 Independent samples t-test: t = 2.57, sig = 0.000.
19 Spearman’s Correlation: Job (-0.258, sig = 0.000); Isolation (-0.223, sig = 0.000).
4.7
4.9
6.1
7.0
7.2
7.4
1 2 3 4 5 6 7 8 9 10
Own health
Average person
Family member / Loved one
Overseas
Australia
3.0
3.7
3.8
3.9
1.0 2.0 3.0 4.0 5.0
Physically active
Stressed
Anxious
Socially isolated
Extremely
low risk
Extremely
high risk
A lot less A little less About the same A little more A lot more
6
Figure 5: Wider Impacts of COVID
Figure: (6a) Had Restrictions on Movement and (6b) Number of Periods of Restriction
3.3. Attitudes toward Government and COVID-19 Measures
Governments have played an important role in the response to COVID-19, and Figure 7 shows the
responses. As with previously discussed results, there are differences in attitudes between Australian
based respondents and those living overseas. Those overseas exhibit significant levels of agreement
with all statements20, and while Australians are on average neutral attitude with respect to
governments caring more about people than the economy, and that pharmaceutical companies are
appropriately regulated; and there is significant disagreement with the statement that business can
be trusted to do the right thing for society and that their own health is more important than the health
of people I don’t know21.
Figure 7: Attitudes to Government and COVID-19 Measures
20 One sample t-test against fixed value of 4 (mid-point of the scale) – see Appendix A2 for results.
21 One sample t-test against fixed value of 4 (mid-point of the scale) – see Appendix A2 for results.
3.2
3.3
3.5
3.5
3.7
4.0
3.5
3.9
1.0 2.0 3.0 4.0 5.0
Capacity of health system
My job / education
Health of the economy
Social isolation
Overseas
Australia
Government
81%
Voluntary
10%
None
8%
One
48%
Two
25%
Three or
more
27%
3.5
3.6
4.0
4.1
4.7
4.8
5.5
5.8
5.1
4.5
4.8
5.2
4.7
5.6
5.5
5.8
1.0 2.0 3.0 4.0 5.0 6.0 7.0
My health is more important than health of people I don't know
Business can be trusted to do the right thing for society
The govt. cares about people more than the economy
Pharmaceutical companies are appropriately regulated
The govt. makes decisions that are in our best interest
The govt. should incentivise people to get the COVID-19 vaccination
People should use a COVID-19 tracing app on their phone
Should be criminal offence to spread false info. about contagious…
Overseas Australia
Strongly
Disagree
Strongly
Agree
Not at all
concerned
Extremely
Concerned
A little
concerned
Somewhat
concerned
Very
concerned
7
Those overseas agree more strongly on average, that governments care more about more than the
economy, business can be trusted to do the right thing for society, their own health is more important
than the health of people they do not know, pharmaceutical companies are appropriately regulated,
and that the government should incentivise people to get the COVID-19 vaccination22. Those on
average to higher incomes have higher average agreement to the statements that governments make
decisions that are in our best interests, and that their own health is more important than the health
of others they don’t know23.
3.4. Attitudes Towards Vaccination
A series of questions were also asked to gauge the respondent’s attitudes towards vaccinations. Figure
8 shows that, on average, there is significant level of agreement that respondents want life to return
to normal as fast as normal, belief in vaccinations and science, that people have a duty to protect
themselves and others, and that vaccinations help stop viruses from spreading. Importantly, there is
significant disagreement with negative outcomes of vaccinations, indicating that on average there is
low concern about safety of the vaccine, and about potential side effects24.
Figure 8: Attitudes Towards Vaccinations
In terms of socio-demographic differences, Australians have significantly higher average belief in
vaccinations and science, and significantly lower concerns about potential side effects25. A significant,
albeit weak, negative correlation between age and the life returning to normal statement indicates
that younger respondents are wanting normality to return more quickly26, and the length of time spent
in restrictions is positively correlated with the belief that vaccinations help stop the spread of the
virus27.
4. Willingness to get Vaccinated
As seen in Figure 9a, just under a quarter of respondents have had a COVID-19 vaccination, though
those overseas are more likely to have been vaccinated (39%) than those in Australia (15%)28. With
regards to the type of vaccine preferred (shown in Figure 9b) Pfizer is the one that most respondents
have had (followed by AstraZeneca), but the Pfizer vaccine is particularly more preferred among those
22 Independent samples t-test – see Appendix A2 for results.
23 ANOVA: Govt. act in best interest (F = 4.388, sig. = 0.015); My health more important (F = 3.430, sig. = 0.037).
24 One sample t-test against fixed value of 4 (mid-point of the scale) – see Appendix A3 for results.
25 Independent samples t-test: belief in science (t = 2.807, sig. = 0.008); side-effects (t = 2.369, sig. = 0.021).
26 Spearman’s Correlation: -0.285, sig = 0.005.
27 Spearman’s Correlation: 0.275, sig = 0.007.
28 Crosstab: Chi-Square = 9.719, sig. = 0.021.
3.4
3.5
4.5
4.6
4.6
4.7
1.0 2.0 3.0 4.0 5.0 6.0 7.0
I am concerned about the potential side-effects of vaccines
Concerned about the safety of the vaccine in its development
Vaccinations help stop the spread of the virus
People have a duty to protect themselves and others
I believe in vaccinations and science
I just want life to return to normal as quickly as possible
Strongly
Disagree
Strongly
Agree
8
yet to be vaccinated29. The large number of respondents who have received the “other” type of
vaccination are located overseas, particularly in China. And have received the Sinopharm vaccination.
Figure: (9a) Received Vaccination and (9b) Vaccination Received/Preferred
Respondents were also asked to consider how willing they would be to receive a COVID-19 vaccination
on a scale from 1 (I do not want one) to 10 (I will definitely get one)30. Similarly, they were also asked
how likely they would be to receive a vaccination, if they had the ability to choose which one they
would receive. The distribution of responses is shown in Figure 10, and average of both measure of
willingness indicate a significant willingness to be vaccinated31. This is true even among the subset of
respondents who are yet to be vaccinated (7.8 for willingness, 8.1 for willingness to receive vaccine of
choice)32. Further, though the difference in small, there is a significantly higher willingness to be
vaccinated with the vaccine of choice among all respondents33, and among those yet to be
29 Crosstab: Chi-Square = 6.527, sig. = 0.011
30 Those respondents who have already received a COVID-19 vaccine were assumed to be 10 on this scale.
31 One sample t-tests against fixed value of 5 (mid-point of the scale): willingness to have (t = 11.367, sig. = 0.000); willingness to have
preferred (t = 12.626, sig. = 0.000).
32 One sample t-tests against fixed value of 5 (mid-point of the scale): willingness to have (t = 8.013, sig. = 0.000); willingness to have
preferred (t = 9.110, sig. = 0.000).
33 Paired samples t-test: average difference = 0.278, t = 2.227, sig. = 0.028.
34 Paired samples t-test: average difference = 0.355, t = 2.240, sig. = 0.028.
Yes
22%
No
78%
47%
38%
14%
0%
65%
10%
20%
5%
Pfizer AstraZeneca Other Moderna
Had vaccination
Yet to be vaccinated
3% 4% 1% 0% 2%
6% 4% 5%
9% 7%
61%
2%
5%
1% 0% 2%
4% 3% 1%
8% 7%
67%
Do not
want
1 2 3 4 Unsure 6 7 8 9 Definitely
get
Willingness to receive COVID-19 vaccine Willingness to have vaccine of choice
Ave. = 8.3 Ave. = 8.5
9
To better understand why some respondents may be more or less willing to receive a vaccination than
others, regression modelling was performed. Prior to modelling the dataset was cleaned and variables
were coded appropriately for regression35. Following this, scatterplots were used to examine the
relationship that each independent variable had with the dependent variable (Willingness to receive
a COVID-19 Vaccine), checking for problems with non-linearity and heteroskedasticity (it should be
noted that some scatter plots revealed a potential non-linear pattern and resulted in a quadratic
transformation of each problematic variable)36. Initially, all available independent variables were
regressed against the willingness to receive a vaccine. Examination of the VIF statistics revealed that
multicollinearity was not present in this initial model, and examination of further outputs revealed the
error term to be approximately normally distributed in this initial model37. Given that all assumptions
of linear regression are satisfied, this initial model was further refined via the stepwise process.

The final regression model is presented in Table 1. This regression model is significant38 , and 70.4%
of the variation in willingness to receive a COVID-19 vaccination can be explained by the independent
variables listed in the table. Note that the independent variables are ordered by size of relative impact
(as measured by the standardised beta coefficients) starting with the variable with the biggest impact.
Table 1: Explaining Differences in Willingness to Receive COVID-19 Vaccine
Variable Coefficient t-value
(Constant) -5.494 -2.437
(Government cares more about people than money)2 -0.171 -2.841
Government cares more about people than money 1.385 2.732
I believe in vaccinations and science 2.184 6.050
Risk of COVID-19 to health of an average person 0.491 3.584
Unrestricted domestic travel within country 1.179 5.293
Risk of COVID-19 to your own health -0.359 -3.072
Willingness (vaccine of choice) – Willingness (get vaccine) 0.653 4.669
Should be a criminal offense to spread false information about contagious diseases -0.371 -3.262
I am concerned about the safety of the vaccine in its development -0.422 -2.646
I am of about average health 0.797 2.130
The variable with the largest impact on willingness to be vaccinated is the attitude that “government
cares more about people than money”, the positive coefficient indicating that as people agree more
strongly with this statement, the willingness to be vaccinated increases. Note that the increase is non-
linear, in particular the negative coefficient on the squared term indicates that an increase in the belief
that the government care more about people will increase a willingness by a larger amount when
moving from low to moderate levels of agreement, compared to moving to from moderate to high
levels of agreement.
35 Refer to Appendix A1 for details on data cleaning and coding.
36 Refer to Appendix A4for scatterplots of potential non-linear variables.
37 See Appendix A5 for results of initial regression model.
38 ANOVA: F = 20.261, sig. = 0.000 – see Appendix A6 for results of final regression model.
10
The positive coefficient for belief in vaccination and science indicates that more trust equates to
greater willingness to be vaccinated, likewise the more of a risk a person perceives COVID-19 to be to
the health of an average person the more likely someone is to get vaccinated; and those who reate
themselves of being of average health are also more likely to get vaccinated than others. Strategies to
encourage people to get vaccinated will be discussed in the next section, but the regression model
also reveals that the ability to engage in unrestricted domestic travel also results in significantly higher
willingness to be vaccinated.
To understand how being able to choose a preferred vaccine might impact on willingness, the
difference between a person’s willingness to be vaccinated if they could choose the vaccine they
received and their overall willingness to be vaccinated was calculated. This would produce a positive
number when the willingness to be vaccinated increases if the respondent can get their vaccine of
choice. The positive parameter indicates that respondents are thus more willing to be vaccinated
overall if they can get their vaccination of choice.
The negative coefficient for concern about the safety of the vaccine indicates that those who agree
with this statement more are less likely they are to be vaccinated, which is to be expected. However,
the negative coefficient for risk of COVID-19 to a respondents own health indicates that those who
perceive COVID-19 to be of greater risk to themselves are less likely to get vaccinated. This may seem
counter-intuitive, but potentially those that are concerned about contracting COVID-19 are equally
apprehensive because of the potential side effects (and the fact that most vaccines work by
introducing a small amount of the virus being vaccinated against). Likewise, the negative coefficient
indicating that those who agree more strongly that spreading false information should be a criminal
offense are less willing to be vaccinated also seems counter-intuitive, but maybe those people who
are very willing to be vaccinated are not deterred by false information or alternatively those who are
hesitant to get vaccinated and are looking to research the vaccination process may want to avoid false
information wherever possible while trying to determine their course of action.
5. Encouraging Vaccination
A series of potential strategies to encourage vaccination were presented to respondents and they
were asked to rate how much more (or less) likely that such an incentive would make them be to get
vaccinated. The results are shown in Figure 12, and while the majority of respondents state that all
four strategies would make them more likely to get vaccinated, the two that provide the seemingly
largest incentives are the ability to travel internationally and domestically without restriction (71%
and 61% of respondents respectively, stated this would make them a lot more likely to get vaccinated).
Higher numbers of females said that unrestricted domestic travel would make them a lot more likely
to get vaccinated39, whereas a higher number of those located in Australia stated that a tax rebate or
stimulus payment would make them a lot more likely40.
Significantly positive correlations indicate that41:
• Those who believe the government makes decisions in our best interest, believe business can be
trusted to do the right thing by society, think that pharmaceutical companies are appropriately
regulated, believe vaccines stop the spread of the virus, who want life to return to normal as
39 Crosstab: Chi-square = 9.809, sig. = 0.028.
40 Crosstab: Chi-square = 8.779, sig. = 0.032.
41 Spearman’s correlation – see Appendix A7 for correlation matrix.
11
quickly as possible, and who feel more socially isolated, state that this incentive increases their
likelihood of getting vaccinated. Those who are concerned about vaccine side-effects are not likely
to be motivated to get vaccinated by being able to attend mass gatherings.
• Those who are believe business can be trusted to do the right thing by society, believe in
vaccinations and science, believe vaccinations stop the spread of the virus, who want life to return
to normal as quickly as possible, state that unrestricted domestic travel would increase their
likelihood to get vaccinated. Those who are more concerned about vaccine side-effects and about
the safety of the vaccine and its development, are not likely to be motivated to get vaccinated by
this incentive.
• Unrestricted international travel is more likely to be a motivator among those who believe
vaccinations stop the spread of the virus.
• Those who believe in vaccinations and science, who believe vaccinations stop the spread of the
virus, and who feel more socially isolated all state that a tax rebate or stimulus payment will make
them more likely to get vaccinated.
Figure 12: Incentives to Increase Willingness to be Vaccinated
6. Conclusion and Recommendations
This report has provided an analysis of attitudes of a convenience sample of respondents, as they
relate to COVID-19 and vaccinations, with particular reference to the willingness of people to receive
a COVID-19 vaccination and how different incentives might increase that willingness. While the sample
may be biased given the nature of how the data was collected (analysis shows the sample to be
younger, of above average health, who consider themselves to be well informed about health and
medical issues), the sample does give relevant insight into how this segment of society might respond
to vaccination. It could be argued that it is this group that need to be most convinced in order to
approach “herd immunity”, given the relatively smaller impact health impacts of COVID-19 on the
majority of the sample.
2%
27% 27%
44%
3%
13%
23%
61%
1%
11%
17%
71%
2%
35%
23%
40%
Less likely No change More likely A lot more likely
Mass gathering without social distancing
Unrestricted domestic travel
Unrestricted international travel
Tax rebate or stimulas payment
12
It is worth noting that across all analysis there is very little variation in response based on socio-
demographics potentially reflecting a consistency of opinion across the population (though it might
also be a result of too little variation in the sample outside of younger tertiary students). However, of
particular importance to the vaccination efforts in Australian, the likelihood of catching COVID-19 is
thought to be low on average, and the risk to health of getting the virus is perceived to be significantly
lower than those who are located in other countries. Additionally, Australian respondents are
significantly less concerned about the impact of COVID-19 on social isolation, the job or education,
and the health system itself. These results point towards Australian respondents as having a relatively
lower perceived risk of COVID-19 overall, perhaps do to the limited health impacts of COVID-19 within
Australia, which may result in a deleterious impact on vaccination rates.
There is a significant and high likelihood of being vaccinated which is a positive result of the vaccine
roll-out in Australia. However, the relatively small impact of COVID-19 compared to other parts of the
world may mean that getting vaccinated may not be seen as urgent. Furthermore, there is significantly
less trust among Australians in the regulation of pharmaceutical companies, in business and in the
government caring more about people than money. These attitudes may also result in slow uptake of
the vaccine. It is interesting to note that Australians are significantly less likely to find their own heath
to be more important than the health of others: it is thus possible that a campaign that emphasis the
role of the role of vaccination in helping to protect the more vulnerable in the community would have
a particularly stronger impact on promoting uptake, as opposed as focusing on the benefits to the
individual. This is further supported by the finding that there is strong agreement among all
respondents that people have a duty to protect themselves and others.
While there is significant agreement in the sample that vaccinations and science can be trusted, and
that vaccinations help stop the spread of the virus, that level of agreement is still far below “strong”,
indicating that there is scope for authorities to develop educational campaigns that clearly and concise
communicate the safety and efficacy of various vaccinations. This is important as the regression
modelling indicates that greater belief in vaccinations and science, and lower concerns about the
safety of vaccines, leads to significantly higher willingness to be vaccinated. The regression model
further indicates that campaigns should emphasise that despite the relative success Australia has had
in handling the pandemic, the risk of COVID-19 to the health of the average person is not gone, and
the only way that we will be allowed to travel safely without fear of potential lockdowns to curb
transmission, is to achieve nationwide immunity through the vaccine process.
With regards to incentives to encourage uptake, unrestricted travel is the predominant mechanism
that would result in an individual being “a lot more likely” to receive a vaccine. That said, respondents
emphasised the importance of being able to access honest communication/research about vaccine
efficacy and side-effects, understanding that the relatively low chance of a negative outcome
compared to the very high impact of successful vaccine program should be strongly reinforced. While
making it a criminal offense to provide false information would actually reduce willingness to be
vaccinated (perhaps removing dissenting views may lead to distrust), effort should be made to
comprehensively refute false information in a way that is accessible to a wide audience (both with
respect to being able to access that debunking and understand the argument).
People are more willingness to be vaccinated if they can receive the vaccine of their choice, which
may be problematic for Australia as the majority prefer the Pfizer vaccine, of which Australia currently
has limited supply. Further supporting the notion that there are logistical barriers to vaccination rather
than attitudinal, are the statements among respondents that vaccines should be free, the process
should be more convenient, and the rollout could be better managed. Others have also noted that
supply and distribution issues, confusion about how to get vaccinated, difficulties in accessing the
13
online booking system, and being unable to find time off work all play a role in slowing vaccine
uptake42. With regards to vaccine programs overall (not just specific to COVID-19), behavioural
scientists have concluded that effect strategies involve making vaccines free and services convenient,
reminders when a vaccine is due, default appointments, performance monitoring and feedback, on-
site vaccination, standing orders, incentives and requirements43.
Overall, this report concludes that Australians are willing to be vaccinated and given the increased use
of the Pfizer vaccine, the biggest hurdle will likely be logistical issues. To further convince those
experiencing hesitancy, several strategies have been suggested above to convince those individuals
to become vaccinated. It should be noted though, that this sample does represent a relatively more
educated person, who is perhaps more trusting of science. As such, further research should be
undertaken to examine attitudes and behaviours in a more diverse sample, with emphasis on those
groups who are perhaps distrustful of science and vaccinations overall.
42 https://www.theguardian.com/world/2021/may/31/have-you-convinced-your-australian-family-or-friends-
to-get-the-covid-vaccine
43 https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)32206-6/fulltext
14
Appendices
A1. Notes on Data Cleaning
Deleted observation numbers 95, 96, 97, and 100 to 106 due to large amounts of incomplete /
missing data.
Note that the following respondents are older respondents who are in a relative minority compared
to ages of other respondents, in particular respondents 48 and 49 who are both 73:
• 3, 4, 5, 10, 11, 13, 18, 19, 29, 38, 46, 48, 49, 52, 56, 68, 69, 78, 81
Question 10 was cleaned to ensure all response were in numeric form and were converted to be in
monthly units:
• #58 changed from + 2years to 18 months.
• #63 “idk” changed to black cell / missing.
• #99 “yes” changed to black cell / missing.
Question 12 and Question 14 recoded to a value of 10, if respondent has already received a COVID-
19 vaccination (i.e. if Q11 = 1)
Question 25_2 cleaned to be a number between 1 and 100:
• #12 changed from “I have no work” to 100.
• #32 “0” change to blank and this respondent answer to Q25 was changed to “1” (no impact).
Respondent number 93 had a string of strongly disagrees (1) for Q17 questions about science, but
said 10/10 for vaccination likelihood. Seems counter intuitive. Changed that respondent’s values to 5
(i.e. assumed that respondent did not pay attention to the scale numbering and answer at the wrong
end accidently).
Dummy code the following variables to prepare for regression:
• Q2: Mode = 2
• Q3: Mode = 2
• Q8: Mode = 1; small sample outside of (1) Govt enforced
• Q9: Mode = 1
• Q11 to 0/1
• Q13: Mode = 2
• Q21: Mode = 2, only 1 who said non-binary so delete gender for #62
• Q23: to 0/1
• Q24: Mode = 3
• Q25: to 0/1
15
The normality of the continuous variables were examined, this in bold in the following table are not
normally distributed:
Question Skewness Kurtosis
Q4_1 -0.248 -0.721
Q4_2 -0.746 -0.130
Q4_3 -0.651 0.042
Q4_4 -1.329 0.021
Q4_5 2.500 -1.656
Q4_6 0.283 -0.927
Q4_7 -0.704 -0.200
Q4_8 -0.812 -0.500
Q5 1.154 1.151
Q6_1 -0.457 0.153
Q6_2 -0.538 -0.374
Q6_3 -0.982 0.230
Q7_1 -0.786 -0.237
Q7_2 -0.518 -0.017
Q7_3 -1.012 -0.033
Q7_4 -0.789 -0.534
Q10 0.264 1.071
Q12 1.930 -1.697
Q14_1 2.777 -1.972
Q17_1 8.373 -2.628
Q17_2 8.970 -2.404
Q17_3 6.221 -2.115
Q17_4 -0.594 -0.582
Q17_5 -0.200 -0.801
Q17_6 8.585 -2.817
Q18_1 -1.145 -0.435
Q18_2 0.701 -1.284
Q18_3 1.298 -1.551
Q18_4 -1.453 -0.167
Q22 3.856 2.146
16
A2. Attitudes towards Government Measures
Respondents Overseas
Respondents in Australia
17
Overseas vs Australia
A3. Attitudes towards Vaccination
18
A4. Assessing Linearity for Regression
Q4_2
Q4_8
R² = 0.0036
R² = 0.063
0
2
4
6
8
10
12
0 1 2 3 4 5 6 7 8
Q12
R² = 0.0536
R² = 0.1017
0
2
4
6
8
10
12
0 1 2 3 4 5 6 7 8
Q12
19
Q5
Q6_1
R² = 0.0131
R² = 0.0807
0
2
4
6
8
10
12
0 2 4 6 8 10 12
Q12
R² = 0.0097
R² = 0.0666
0
2
4
6
8
10
12
0 2 4 6 8 10 12
Q12
20
Q6_3
Q7_2
R² = 0.0054
R² = 0.0431
0
2
4
6
8
10
12
0 2 4 6 8 10 12
Q12
R² = 0.0097
R² = 0.0454
0
2
4
6
8
10
12
0 1 2 3 4 5 6
Q12
21
Q17_5
Q26_3
R² = 0.1726
R² = 0.2579
0
2
4
6
8
10
12
0 1 2 3 4 5 6
Q12
R² = 0.0884
R² = 0.1541
0
2
4
6
8
10
12
0 1 2 3 4 5 6
Q12
22
A5. Output for Initial Regression Model
23
24
25
A6. Output for Final Regression Model
Note that the VIF statistic for Q4_2 and Q4_2-squared are both high (> 10).
However, this is logical as (Q4_2)2 is a mathematical derivation directly from
Q4_2. It does not matter is these two variables are correlated in this way, as they
are related to the same effect (the scale measuring the strength of agreement
that the government cares more about people than money).
26
27
A7. Correlation of Attitudes and Incentives for Vaccination
Q18_1 Q18_2 Q18_3 Q18_4
Q4_1 Correlation Coefficient .216* 0.198 0.146 0.082
Sig. (2-tailed) 0.035 0.053 0.155 0.428
N 96 96 96 96
Q4_3 Correlation Coefficient .248* .215* 0.14 -0.018
Sig. (2-tailed) 0.015 0.036 0.173 0.859
N 96 96 96 96
Q4_7 Correlation Coefficient .212* 0.197 0.088 0.048
Sig. (2-tailed) 0.038 0.055 0.396 0.643
N 96 96 96 96
Q4_8 Correlation Coefficient .404** .405** .342** .324**
Sig. (2-tailed) 0 0 0.001 0.001
N 96 96 96 96
Q17_2 Correlation Coefficient 0.195 .346** 0.145 .289**
Sig. (2-tailed) 0.057 0.001 0.159 0.004
N 96 96 96 96
Q17_3 Correlation Coefficient .332** .351** .307** .312**
Sig. (2-tailed) 0.001 0 0.002 0.002
N 96 96 96 96
Q17_4 Correlation Coefficient -0.197 -.217* -0.13 -0.123
Sig. (2-tailed) 0.054 0.034 0.206 0.233
N 96 96 96 96
Q17_5 Correlation Coefficient -.284** -.219* -0.115 -0.15
Sig. (2-tailed) 0.005 0.032 0.265 0.145
N 96 96 96 96
Q17_6 Correlation Coefficient .328** .301** 0.157 0.185
Sig. (2-tailed) 0.001 0.003 0.127 0.071
N 96 96 96 96
Q26_3 Correlation Coefficient .216* 0.163 0.198 .278**
Sig. (2-tailed) 0.035 0.114 0.054 0.006
N 95 95 95 95
Q26_4 Correlation Coefficient 0.083 0.099 0.056 0.183
Sig. (2-tailed) 0.424 0.34 0.589 0.076
N 95 95 95 95