Successful vaccination campaigns start with understanding survey bias

The question is no longer will there be enough COVID vaccine for all adult Americans, but whether enough Americans will take it.

(Photo credit: Getty Images).
(Photo credit: Getty Images).

As of April 21, 26% of Americans have been fully vaccinated against COVID-19, according to the Centers for Disease Control and Prevention. Although that’s positive news, herd immunity depends on convincing hesitant populations to get vaccinated. Effective public health messaging starts with understanding survey science. 

All surveys are not created equal

The accuracy of any survey correlates directly with how its questions are phrased. For example, questions may be worded in a leading way. Compare “Would you get a COVID vaccine?” to “Would you get a life-saving COVID vaccine?” The latter phrasing can influence respondents to answer affirmatively.

The way response options are phrased affects outcomes, too. Asking whether someone is willing to receive the vaccine, offering “Yes” or “No” without a “Don’t know” option forces respondents into a category. It also fails to distinguish people who are unsure about receiving the vaccine — the people who are most critical to reach.

The survey size must be large enough to represent the population in question. This is one reason why focus groups are inherently problematic as a means to shape messaging. They also magnify social desirability bias and groupthink in responses.

Messaging mistakes

Survey data alone shouldn’t inform an entire messaging campaign, because people are notoriously bad at identifying which messages will persuade them to take action. Our research repeatedly finds no correlation between what someone says they like and what motivates their actions. 

Public health campaigns can’t just rely on lessons learned in previous campaigns, because what worked on previous vaccination initiatives is very unlikely to drive COVID vaccine acceptance. With flu vaccines, we determined that messages urging Americans to get vaccinated to protect vulnerable, at-risk populations were successful. But we’ve consistently found that espousing the personal benefits of COVID vaccination far outstrips the influence of messaging about community impact. 

Communicators know that messengers matter as much, if not more, than the messages they deliver. People who are already likely to get vaccinated, for example, those working on vaccine campaigns, have the highest degree of trust in conventional messengers like medical professionals and scientists. But our data indicates that skeptics are less convinced by experts, and much more open to hearing vaccine messages from everyday people who have experienced COVID-19.

Boosting vaccine acceptance

Convincing holdouts to embrace the COVID-19 vaccine starts with identifying who is persuadable within the target population. It’s vital to determine any non-attitudinal barriers to vaccination, including financial or logistical.

Next, determine what this persuadable population needs to overcome their reluctance to vaccination, and their psychological barriers.

From there, scientifically validate the correct way to speak to target audiences and the right persons to deliver these messages. Flag any messages that could backfire. Communicate these insights to field managers and outreach coordinators.

Constantly measure campaign success. Regularly look at the data, and be ready to pivot if the campaign isn’t working.

The question is no longer will there be enough COVID vaccine for all adult Americans, but whether enough Americans will take it. The answer depends on messaging that speaks loudly and clearly to the people who most need to hear it, messaging vetted by scientific methodology, not past experience or intuition.

Crystal Son is an epidemiologist turned data scientist and oversees Civis's data and analytics work in healthcare and public health.

Have you registered with us yet?

Register now to enjoy more articles and free email bulletins

Already registered?
Sign in