125 characters. That’s less than the length of this sentence, but is it enough to encourage people to get a vaccine that may help end the deadliest pandemic the world has seen in a century?
As a behavioral scientist, I’m used to boiling down technical information into clear, digestible calls to action. In one of the most profound psychological insights of the 20th century, a psychologist named George Miller noted that rather than being a perfect machine with endless potential, the human brain was severely limited by its short-term memory. Since then, numerous studies have found that, thanks to constraints like cognitive overload and bounded attention, less is often more: Simply using fewer words can make people significantly more likely to understand what you’re trying to get across and, consequently, choose to do it. A former professor of mine tells members of his lab to assume that people only read the first paragraph of the emails they send. Thus, I am used to maneuvering within limited space, to cutting out word play and making sure a summary of the bottom line appears in my first sentence or two. And yet: 125 characters felt tight even for me.
As an academic consultant this summer, I helped the Mind, Behavior, and Development Unit (eMBeD), the World Bank’s behavioral sciences team, investigate behavioral predictors of COVID-19 vaccine compliance and hesitancy in a number of countries. With international vaccine production and distribution accelerating, eMBeD created surveys—which were shared with tens of thousands of people in multiple countries through Facebook—to examine trusted sources of information, social norms, and other factors that influence vaccine uptake.
A number of behavioral scientists have investigated vaccine hesitancy during the pandemic, but the World Bank project was among the first to use social media to disseminate data-gathering tools and messages. Sampling involved stratified clusters using age, gender, and region, based on population weights from the latest available national data. While distributing via social media meant the surveys used convenience samples and were likely skewed towards those who were more technologically savvy, findings were fascinating, and some patterns emerged. For example, in Lebanon and Iraq, which had a combined total of more than 13,000 respondents, those who were hesitant about COVID-19 vaccination were more likely to be women, have strong religious beliefs, and have low confidence that their family or friends would get the vaccine. Interestingly, the strongest predictor of hesitance appeared to be how people felt their country’s government had handled the COVID-19 pandemic in general: People who thought their government had not done a good job were much more likely to be vaccine-hesitant. (Dr. Heidi Larson, director of the Vaccine Confidence Project at the London School of Hygiene and Tropical Medicine, noted similar findings in a talk at the Association for Psychological Science virtual convention in May.)
The surveys also offered hints at deeper nuances: In both Iraq and Lebanon, more than two-thirds of those who said they would not get the COVID-19 vaccine had received other vaccinations for themselves or their children in the past. And in Lebanon, 90% of those who said they would not get the vaccine reported always wearing a mask in public, which was essentially the same level of mask-wearing as those who definitely planned to get vaccinated. In other words, rather than “anti-vax” sentiment, the main concerns in Lebanon and Iraq appeared to be curiosity about the efficacy of the COVID-19 vaccine in particular, as well as worries about its potential side effects.
Based on results from these surveys, we then identified key profiles of those who were vaccine-hesitant for various reasons, and developed messages to address those articulated concerns. The messages were distributed via Facebook ads, with eMBeD randomly assigning people to receive various wording, each approach grounded in a behavioral theory such as peer influence. And that’s where 125 characters comes in: It’s the limit that Facebook places on the primary text in their news feed ads. Any more than that, and your message may be cut off, risking that people won’t be able to see the action you’re encouraging them to take.
During brainstorming sessions with the team, we fussed over every syllable, finessing each sentiment into its most compelling and clear self. Dozens of words were on the chopping block when we finished. We talked through the nuances of the study design, how to balance the desire for multiple measures vs. clarity of what exactly we were asking people to do. Was our outcome a positive indication of their intention to get vaccinated? Or was it whether they clicked through to the Minister of Health site to actually make an appointment?
As noted in our DrPH competencies, the work required communicating public health science to diverse stakeholders, including individuals at all levels of health literacy, for purposes of influencing behavior. I had to integrate knowledge of cultural values and practices into the design after assessing an audience’s knowledge and learning needs. It was like starting with a huge block of limestone and ending with a palm-sized figurine: The process was both dizzying and invigorating. And there, left behind after all the carving, were the messages that would become the ads: spartan and succinct. And, yes, nearly all were well below 125 characters.
