While Bostonians are hesitant to host the Olympics, Americans across the country overwhelmingly support the idea of the games on home turf, according to a new Associated Press-GfK poll.

The support decreases when people are asked if they would want the Olympics in their local area. It dips even further when they are asked if public funds should be used to pay for them.

Nearly nine of 10 Americans — 89 percent — support a bid to host the Olympics somewhere in the United States. Yet just 61 percent would support a bid in their local area. Fifty-two percent of respondents would support an Olympics in their local area if it were paid for with a combination of public and private funds, while 46 percent would be opposed to either that proposal or a local Olympics.

“Our own research tells us that the Olympic brand is incredibly strong in the United States and it’s one of the reasons that we decided to bid for the 2024 Games,” U.S. Olympic Committee spokesman Patrick Sandusky said. “This poll confirms that and shows there is a strong desire, from coast to coast, to see the Games return to the U.S.”

But the nationwide numbers do not echo what’s happening in Boston, where the effort to host the 2024 Olympics has hit a number of roadblocks. Approval ratings around the city have been grim — well under 50 percent — and a referendum has been set up for next November. If that vote doesn’t win both in the city and the state, organizers have vowed to pull the bid, even though the official deadline to declare a city’s candidacy is this September.

That’s one of the key issues the U.S. Olympic Committee board will tackle Tuesday at a meeting at which the Boston group will present an update. The USOC recently received poll numbers on a survey it commissioned. The result of those numbers will also factor into whether the board decides to continue with a Boston bid, look elsewhere, or withdraw from the competition completely. A backup plan could be to consider Los Angeles.

In the AP-GfK poll, 56 percent of respondents said hosting the Olympics is worth the cost to the local areas, and 42 percent said it is not.

“This poll shows clearly that Americans are eager to bring the Olympic and Paralympic Games back to the United States,” said Erin Murphy, the chief operating officer of Boston 2024. “Boston 2024 is proud to represent the U.S. as we work to build a bid that is fiscally responsible and provides world-class venues and a lasting legacy for the Games and the city of Boston.”

One main complaint in Boston is that hosting an Olympics leaves the public vulnerable to footing too much of the bill and is not the proper way to pay for much-needed infrastructure improvements. Last winter’s massive snowstorms shined a light on the city’s less-than-ideal transit system and also brought up more general questions about Boston’s ability to handle major projects.

Last month, Celtics co-owner Steve Pagliuca took over as bid leader. Boston has been revamping its bid, spreading some of the venues farther across the state than originally expected, and vowing that there will be more changes between now and September 2017, when the Olympics are awarded. Boston’s newest plans will be unveiled Monday.

Paris, Rome and Hamburg, Germany are the other declared candidates.

The International Olympic Committee normally takes poll numbers very seriously when deciding who the most viable candidates are to host the games.

One poll in Chicago done shortly before the vote for the 2016 Olympics showed less than 50 percent support for hosting, a number that dropped drastically after the mayor changed course and agreed to sign a standard host city contract that left the city to bear financial responsibility for any losses. Those numbers were one of many factors — most of which were related to the USOC’s then-poor relationship with the IOC — that caused Chicago to finish last in the voting.

The AP-GfK Poll of 1,005 adults was conducted online from June 19 to 21 using a sample drawn from GfK’s probability-based KnowledgePanel, which is designed to be representative of the U.S. population. The margin of sampling error for all respondents is plus or minus 3 percentage points.

Respondents were first selected randomly using phone or mail survey methods and later interviewed online. People selected for KnowledgePanel who didn’t otherwise have access to the Internet were provided access at no cost to them.