A survey released Thursday backs up builders’ anecdotal stories that the $8,000 tax credit is boosting home sales now and asserts an extension would continue the stimulus.

Of the first-time home buyers surveyed by Zillow.com, 18% of respondents said an extension of the tax credit through 2010 would be a “primary influence” in their decision to buy while 25% said it would be a “significant influence. Another 25% said it would have “some” influence, and 31% said it wouldn’t make any difference at all.

Zillow estimates 18 million first-time buyers would purchase homes if the tax credit is extended for another year, through November 2010, adding up to $14.86 billion in tax credits.

However, experts acknowledge that a number of those buyers would have purchased a home whether or not there was a federal tax credit. For example, Zillow chief economist Stan Humphries said the survey suggests that about 334,000 buyers would be brought in by the credit who would not buy a house without it. That would mean the true cost to the government for every extra buyer generated by the credit would be $44,491 in tax credits.

Still, many in housing and real estate believe anything that helps the housing market, even in a small way, is worth it.

“While 334,000 may seem like a small number relative to the total number of home buyers who would claim the credit, their addition to the market next year could make the difference between a robust annual increase in home sales next year and a flat or negative change in home sales relative to this year,” Humphries asserted.

“There's little doubt that the tax credit will boost demand at the margin, and that fact will make it easier to work down our current high inventory levels of existing homes on the market,” he continued. “That said, the cost of bringing these additional home buyers into the market is substantial.”

Harris Interactive fielded the study for Zillow, interviewing a nationwide sample of 2,175 adults online. Of those surveyed, 408 would qualify for the tax credit and said they intend to buy a home by the end of 2010. Zillow noted that it can’t estimate sampling error because the survey is not based on a probability sample.

Teresa Burney is a senior editor at BUILDER and BIG BUILDER magazines.