What's the best way to persuade the millions of Americans who are still unvaccinated against COVID-19 to get their shots?

Reassuring public service announcements about the vaccine's safety and effectiveness have proliferated. But increasingly, people are realizing that it will take more than just information to sway the hesitant.

In recent randomized survey experiments by the UCLA COVID-19 Health and Politics Project, two seemingly strong incentives have emerged.

Roughly a third of the unvaccinated population said a cash payment would make them more likely to get a shot. This suggests that some governors may be on the right track; West Virginia's governor, Jim Justice, for example, recently announced the state would give young people $100 bonds if they got an inoculation.

Similarly large increases in willingness to take vaccines emerged for those who were asked about getting a vaccine if doing so meant they wouldn't need to wear a mask or social-distance in public, compared with a group that was told it would still have to do those things.

The UCLA project, which is still going on, has interviewed more than 75,000 people over the last 10 months. This collaboration between doctors and social scientists at UCLA and Harvard measures people's pandemic experiences and attitudes along political and economic dimensions, while also charting their physical and mental health and well-being.

To assess the effectiveness of different messages on vaccine uptake, the project randomly assigns unvaccinated respondents to groups that see different information about the benefits of vaccination. Random assignment makes the composition of each group similar. This is important because it allows the researchers to conclude that any differences that emerge across the groups in people's intentions to get vaccinated are a result of the messages each group saw and not of other underlying attributes.

Last October, one group saw messages that framed the benefits of vaccination in a self-interested way — "it will protect you" — while others saw messages that framed benefits in a more social manner: "It will protect you and those around you." The subtle change did little; roughly two-thirds of people in both groups said they intended to get the shots.

Another experiment investigated the persuasive power of certain endorsements. Endorsers included prominent figures, like then-president Donald Trump and Dr. Anthony Fauci, but also included more personal medical sources like "your doctor."

Most of the effects were small. Telling people their doctor, pharmacist or insurer believed the vaccine to be safe and effective had no discernible effect on intentions to vaccinate, though an endorsement by Fauci increased uptake likelihood by about six percentage points.

Endorsements by political figures evoked strong partisan reactions, with Trump's endorsement decreasing uptake among Democrats in 2020 and increasing uptake for Republicans to a smaller degree. President Joe Biden's endorsement decreased uptake among Republicans in 2021. There were hints in 2021 that a Trump endorsement might still increase uptake among Republicans, but the effects were much smaller than when he was in office.

Last month, researchers randomly assigned unvaccinated respondents to see messages about financial incentives. Some people were asked about the chances they would get a vaccine if it came with a $25 cash payment; other people were asked about receiving $50 or $100.

Roughly a third of the unvaccinated population said a cash payment would make them more likely to get a shot. The benefits were largest for those in the group getting $100, which increased willingness (34% said they would get vaccinated) by six points over the $25 group.

The effect was greatest for unvaccinated Democrats, 48% of whom said they would be more likely to get vaccinated if it came with a $100 payment.

Some past research shows that payment for vaccines can backfire, and in the UCLA study about 15% of unvaccinated people report a decrease in willingness to vaccinate because of payments. But at this later stage of a vaccine campaign — when attention has now turned to the hesitant — the net benefit seems to be tilting toward payment.

The incentive to stop wearing a mask and social-distancing in public also had a strong result. On average, relaxing the mask and social distancing guidelines increased vaccine uptake likelihood by 13 points. The largest gains came from Republicans, who reported an 18-point increase in willingness to get vaccinated.

These results show both the difficulty of getting the remaining unvaccinated people to clinics and the promise of efforts aimed at doing so. While most messaging effects were small, monetary payments seem to motivate Democrats, and relaxing cautionary guidelines seems to work for Republicans. (The CDC recently relaxed guidelines on mask wearing outdoors for vaccinated people.)

The movement toward vaccinations among the hesitant may pick up as time passes, and as people observe the consequences of vaccination among those who were first inoculated. When we asked unvaccinated people why they hadn't tried to get a shot, 38% said they were worried about the side effects, and 34% said they didn't think the vaccine was safe. Efforts at persuasion that demonstrate the continued and consistent absence of side effects for most people and the safety of inoculation may allay these fears. Still, a quarter of the unvaccinated say they just don't trust the government's motives, and 14% say COVID-19 is not a threat to them. These people will be harder to convince.

Data from the project shows how eager Americans are to return to normal activities. Among people who work outside their home, 76% of the survey's respondents said they wanted to go back to doing their job the way they were doing it before the pandemic, and 66% said they thought it was safe to do so as of April. These numbers are similar regardless of vaccination status.

The April survey also asked people what kinds of social activities they had done in the last two weeks. Roughly 30% reported eating at a restaurant; 17% reported attending an in-person religious gathering; and 11% met up with a group of more than 10 nonfamily members. Nearly all took place indoors.

The rates of vaccination among people doing these activities largely reflect the rates in the population, which means not everyone who is out and about has gotten the vaccine.

Among those dining out, 32% reported being fully vaccinated (53% reported not being vaccinated at all). The balance among people attending in-person religious gatherings was about equal — 41% said they were fully vaccinated and 41% reported not being vaccinated at all.

Most of the people at social functions with more than 10 nonfamily members were not fully vaccinated, though the share of vaccinated people was higher for indoor gatherings (40%) compared with outdoor functions (27%).

People are venturing out into social spaces, but around them, unvaccinated people still outnumber the inoculated — and rates of vaccination are slowing. Reversing this trend will take more than impassioned pleas from politicians, friends or medical professionals. Delivering real rewards beyond the vaccine's health benefits may be required.