The public has been overwhelmed by a surge in misleading and false information during the COVID-19 pandemic. The World Health Organisation has decried this “infodemic”, which can lead to mistrust in health authorities and undermine the public health response. People often do not know which information to trust, making them vulnerable to disinformation.
New research suggests that South Africans are more likely to trust scientific sources, such as doctors and the World Health Organisation, than their own government. Most disapprove of the government’s handling of the pandemic.
These are the findings of an online survey we conducted to find out where people were getting their information about COVID-19 from and which sources they trusted most. We also conducted a small experiment to test people’s views on vaccinations. Both were done online, which means that the views represent only those South Africans with access to the web.
Our study, conducted with support from the World Health Organisation’s (WHO’s) Africa Infodemic Response Alliance, showed that when it came to getting information about the pandemic, South Africans appeared to rely mostly on “traditional” media sources. On average, 74% said they got information about COVID-19 via media such as television, radio and newspapers.
The results also showed that approval of the South African government’s response to the pandemic had declined from a year ago, when we conducted a similar study.
The current survey showed a high level of disapproval: 61% of respondents said they “strongly” or “somewhat” disapproved of the way the government was handling the pandemic, while only 21.1% said they “strongly approved”. This has an impact on the effectiveness of messages promoting vaccination. If receivers of pro-vaccination messages disapprove of the sender of the message, they are less likely to trust the content of the message or share such messages with others.
The deteriorating level of trust in the government may be related to the stuttering vaccine rollout in the country, which was high on the news agenda at the time of the study. The rollout plan suffered several setbacks and the government was widely criticised for not meeting its targets. The survey was also fielded at the time when the country’s health minister, Zweli Mkhize, was put on special leave while an investigation against allegations of corruption was under way.
Our study consisted of two parts.
First, we conducted an online survey in which we asked 1,585 South African social media users what media they consumed, which sources of information they trusted most, and their attitudes towards COVID-19. We also asked them how they would evaluate the government’s response to the pandemic.
The second part of the study involved an online experiment with 1,180 social media users. We sought to determine how effective social media messaging strategies were in promoting vaccination, and what role the sender of the message played in how users responded to it.
The online survey showed that, overall, medical doctors and the World Health Organisation were the most trusted sources of information, followed by radio and television. News websites, family and the South African government were less trusted. But they were still more trusted than social media, friends, community leaders, celebrities and faith leaders.
Respondents who intended voting for the governing African National Congress (ANC) or opposition Economic Freedom Fighters (EFF) tended to trust the government’s communication more than supporters of other parties, such as the official opposition party, the Democratic Alliance (DA).
Overall, most said they consulted established news media sources like television (85.6%), radio (79.2%) and newspapers (online 58.3%, print 73.4%) more than they did social media. The exception was Facebook, which had a high usage (85.1%), followed by WhatsApp (67.5%).
Google was also a popular platform to obtain information from (85.3%), but other social media platforms like TikTok (19.6%), Twitter (29.2%), Instagram (26.6%) and YouTube (45.6%) were much less popular sources of information.
In the Facebook experiment, participants saw one of four versions of a Facebook post that included a video encouraging citizens to get vaccinated. Each of the four versions was made to look like it had been posted from a different account. Two of these accounts were from political parties in South Africa (the ANC and DA), and two were institutional accounts (WHO and the National Department of Health).
All four posts included the same video, which was designed to look like a #ViralFact message such as the ones distributed by the WHO’s Africa Infodemic Response Alliance. The video combined two common health communication messaging strategies, “humour” and “fear”.
We were interested in comparing how users would react to the same information coming from different messengers. Specifically, we looked at whether different messengers would result in people being more or less likely to get vaccinated. We also looked at whether users would be more or less likely to share the social media posts depending on where they came from.
We found that media users’ intentions to get vaccinated weren’t particularly swayed by which political party did the posting. In all cases, after seeing the Facebook ad, their intention to get the COVID-19 shot remained very high, confirming findings by other researchers.
But when it came to sharing social media posts, users were less likely to say they would share the Facebook post when they thought it came from an ANC account. Users who were told the post came from the WHO, the National Department of Health or the DA were significantly more likely to share the post.
The study supports others showing a relatively high rate of vaccine acceptance among South Africans. It also suggests that the content of pro-vaccination messages is important for promoting vaccine acceptance. So is the sender.
The strong disapproval of the government’s handling of the pandemic, as well as the overall low levels of trust in the ANC, should be a warning to government communicators that crafting persuasive pro-vaccine messages is not enough. The trust deficit in the messenger also has a negative impact on people’s trust in the message itself, and people’s likelihood to share those messages.