DETROIT, MI — When Scott Walker campaigned to become the governor of Wisconsin in 2010, there was one promise that stood out among all other others. In the wake of a recession that had hit Wisconsin hard, Walker vowed that he would create 250,000 private sector jobs by the end of his first term in 2015. This wasn’t an offhand remark, but part of a detailed proposal that Walker made, an oft-repeated and high-profile promise to Wisconsin voters.

Not long after the election, James Nelson, deputy business editor at the Milwaukee Journal Sentinel (and a recent contributor to CJR), told Walker that he would be using PolitiFact Wisconsin as a vehicle to track Walker’s progress with jobs. “Great!” replied Walker, according to Nelson. “We look forward to that.”

Now, halfway through Walker’s first term, it looks increasingly unlikely that Wisconsin will hit the governor’s target. According to the most current numbers, Wisconsin’s businesses have added 66,000 jobs since Walker took office, leaving him 184,000 to go. A recent Census report ranked Wisconsin 42nd out of 50 states for private sector job creation. Against that backdrop, Walker’s political allies are beginning to back away from the promise—earlier this month, Wisconsin Assembly Speaker Robin Vos distanced Assembly Republicans from the commitment. And while Walker’s administration, early in his term, celebrated jobs reports with press releases and at least one prominent press conference that took credit for increases, jobs data are now released with qualifications about whether or not the numbers are a useful measure of the state of Wisconsin’s economy.

Meanwhile, Nelson and the PolitiFact Wisconsin team are pushing forward with a rigorous check on the jobs promise. When it comes to bringing accountability to Walker’s most prominent campaign commitment, theirs is the only game in town.

The PolitiFact enterprise is most famous for its “Truth-o-meter,” in which researchers check the relative veracity of claims made by leading politicians. But the operation also tracks campaign promises. The Wisconsin team uses the “Walk-o-meter” to track roughly 50 promises made by Walker when he ran for office, marking them as “Promises Kept,” “Compromise,” “Promise Broken,” “Stalled,” “In the Works,” and “Not Yet Rated.” Walker’s jobs promise is currently rated as “in the works.”

To count job growth in Wisconsin, Nelson and others are using as a baseline the most accurate federal count of state employment in 2011, Walker’s first full year in office. This is the Quarterly Census of Employment and Wages, which draws from reports from more than 90 percent of state employers. When the 2012 QCEW numbers are released in June, the Politifact team will re-set its baseline count of Wisconsin jobs. In the interim, to keep an ongoing count as accurate as possible, PolitiFact adds (or subtracts) monthly jobs data that pertain to Wisconsin: the Current Employment Statistics released by the Bureau of Labor Statistics. The jobs data are subject to revision—as a March 14 news release (PDF) from the Wisconsin Department of Workforce Development highlights—and the PolitiFact team keeps up by updating its online graphics and publishing newsy blog posts that contextualize the numbers.

It’s fairly arcane stuff, and no matter how detailed the PolitiFact team gets in its approach, it’s not a perfect system. In addition to being subject to revision, the monthly numbers from the BLS come from a relatively small sample size, which creates large margins of error. And adjustments to the formula over time can put strict month-to-month comparisons on tenuous ground. “Let’s face it, it’s imprecise,” Nelson said. “It’s a snapshot. We know that.”

But it is the best available evaluation of the best available numbers. And it’s proved useful. In December, Walker gave a speech in which, while discussing right-to-work policies, he said Wisconsin’s businesses had created “just under 100,000 jobs” since he took office. Nelson took a close look at how Walker had arrived at the number and concluded that “he was not even close” to the true figure; the resulting PolitiFact item assigned Walker’s statement a “pants on fire” rating. Because of PolitiFact Wisconsin’s consistent scrutiny, Nelson knew how to spot Walker’s number-fudging—and he had two years of evidence to back him up.

Of course, Walker’s pledge itself raises questions about how much control any elected official—even a governor whose party also controls the state legislature—has over the course of short-term job growth. Fortunately, the Journal Sentinel has weaved the work of its PolitiFact arm into articles and editorials that bring a fuller treatment to the story. An April 2012 editorial, for example, rebuked the governor and GOP legislators for taking their “eyes off the ball on job creation.”

And a smart recent piece by the paper’s Craig Gilbert on Walker’s presidential ambitions cites PolitiFact’s work, building off that foundation of fact to get the governor to address his record on jobs:

Walker got elected in 2010 on a specific, well-publicized promise of 250,000 new jobs. In a national race, job growth is an obvious metric for judging governors on economic performance. (It may not be entirely fair, given all the factors beyond a governor’s control, but that hasn’t stopped governors themselves from using it). It’s hard to sell yourself as “somebody who’s turned a state around” if that state ranks poorly in job creation.

…Using a combination of data, PolitiFact Wisconsin estimates the state is only about 15% of the way toward meeting Walker’s target of 250,000 more first-term jobs, with fewer than 40,000 net new jobs generated halfway through his four-year term.

Gilbert’s post continues:

When I asked the governor if it’s fair to judge him on the state’s slow jobs growth, he said yes. But he went on to cite factors he says people should be taking into account…

He defended his own policies as job-friendly, but argued as he has before that the reaction to them—the protests and recall drives—hurt job growth “not just through June 5 (2012) but probably for some time thereafter.”

He went on:

“You had that, compounded by the national economy not improving at a good pace, the fiscal cliff talks, now the sequester and the other looming issues about concerns over the Affordable Care Act, the (health care) exchanges, who’s in, who’s out - those are all legitimate reasons. Now I’m not … you know, building the groundwork or the foundation for saying this is why we’re not going to get to a fixed number (of jobs). It’s just that those are all things that are legitimate issues that are out there that we have to deal with and try and figure out ways to overcome.”

For all the attention it has attracted, PolitiFact is still something of a novelty, having emerged during the 2008 presidential race at the then-St. Petersburg Times (now the Tampa Bay Times). But Nelson and his team are practicing the most traditional sort of public interest journalism—collecting and reporting evidence that hold elected officials accountable to their promises.

The “innovation” here is simple: sustained, painstaking attention. That translates into a willingness to offer more minute details online than ever would have been allowed into a traditional print story, as well as the construction of a standing platform where a running story—the state’s progress toward Walker’s pledge—can be told as it unfolds. Those steps in turn help reporters and readers alike build up a base of knowledge—which means they’re ready to push back against self-serving or misleading official statements, like Walker’s 100,000 jobs claim.

“What we are doing, as responsibly and thoughtfully as possible, is to monitor [Walker’s] most important campaign promise,” Nelson said. He added: “We know that not only readers are paying attention, but the political establishment is paying attention.”

If you'd like to get email from CJR writers and editors, add your email address to our newsletter roll and we'll be in touch.

Anna Clark is CJR's correspondent for Michigan, Wisconsin, Ohio, and Pennsylvania. A 2011 Fulbright fellow, Clark has written for The Guardian, Grantland, and Salon; blogs at Isak; and can be found on Twitter @annaleighclark. She lives in Detroit.