How I Got That Fact: A government database that really works
As part of our efforts to expand and improve fact checking, the American Press Institute regularly presents tips on how media organizations around the country chase down facts. Today, Peter Hancock of the Lawrence Journal-World reveals a valuable resource and offers step-by-step instructions on how to check a candidate’s claim about job growth.
Background: Kansas Gov. Sam Brownback is running for re-election. During his first term he pushed through bold and controversial tax cuts that eliminated taxes on certain kinds of businesses and drastically lowered taxes on others. The rationale was that this would attract businesses to Kansas and stimulate growth. Critics said the tax cuts would mainly benefit the wealthy and would do little to stimulate new job creation.
Fact to check: In his campaign speeches and ads, Brownback asserts that after a decade of losing jobs in the 2000’s while the rest of the nation was growing, the tax cuts have resulted in more than 55,000 new private-sector jobs, and employment in Kansas is now at an all-time high. Meanwhile, several news outlets such as the New York Times and Kansas City Star have editorialized that Brownback’s tax policies were a failure because, from November 2013 to May 2014, Kansas saw a net loss of private-sector jobs while the rest of the nation was gaining them.
How I got the fact: I relied on the Bureau of Labor Statistics to check the facts. BLS has a remarkably user-friendly website – www.bls.gov – that lets you download customized data into a spreadsheet, from which you can easily perform basic calculations. It helps to have a little background or training in how to use Microsoft Excel or Open Office. You can also use those programs to make charts and graphs, or you can have BLS do it for you simply by checking a box at the end of the process. Quick hint: Always use the “seasonally adjusted” figures. Otherwise, when you’re looking at changes across time, your results will be influenced by the choice of starting and ending dates. The first step was to put all of the claims into some sort of context. For that, I downloaded all of the data for both total non-farm employment and private-sector employment, both Kansas and the United States, from January 2000 through the most recent month available.
For this, go to www.bls.gov. Under the “subjects” tab, look under “employment.” You’ll see that the national data is separate from the state and local data. You’ll need to do each one separately, but the process is the same. From the column on the left, choose “SAE Databases.” You have several options about how to retrieve the data. I found the “multi-screen” option the most workable.
It takes you step-by-step through a process to select the data you want. Select the following options: check the box for “seasonally adjusted” numbers and click next; select “All employees, in thousands” and click next; select your state; select “statewide,” or one or more metro areas; select “total nonfarm” and/or “total private” and/or any sub-sectors you want to look at. The next screen asks you to confirm the data you want. You have to click and highlight each one you want, then click next. Another confirmation appears. Just click “retrieve data.”
The results will show you data going back about two years. But from here, you can adjust the time frame you want, even going back as far as 1990. Here, you can also check the box that says “include graphs.” That will give you simple line graphs showing the trend over time. After doing that for both Kansas and U.S. employment, the first thing we noticed was that the Kansas trends almost perfectly mirrored the national trends.
From January 2000 through December 2010, Kansas did indeed see a net loss of private-sector jobs. But so too did the rest of the nation. There was a period from about 2003 through 2007 or mid-2008 when the U.S. economy was growing rapidly. But Kansas employment grew rapidly during that same period too. Both the state and U.S. employment numbers fell off sharply starting in mid-2008 when the Great Recession began.
Brownback took office in January 2011, just as the national economic recovery was beginning. Yes, employment did grow over the next three and a half years. But it grew nationally too. In fact, growth was slower in Kansas than the rest of the nation. We were able to confirm that Kansas added 55,100 private-sector jobs since Brownback took office, and that had been reported earlier. But putting that into context, it amounted to about 5.1 percent growth, while private-sector employment nationally grew by more than 7.8 percent.
We also checked the claim that Kansas employment is now at an “all-time high.” The data from these tables show that’s not true. There were several months from late 2007 through mid-2008 when both total employment and private-sector employment was higher. So we called the Brownback campaign to verify the data and ask about the source of their claim. That’s when we learned they were using an entirely different database – the unemployment data – and extrapolating employment levels by subtracting unemployment from the total labor force.
We noted that without passing judgement because, by that time, deadlines were looming and we didn’t have time to get quotes from experts explaining the difference between the surveys and which is more reliable for which purpose. That may be fodder for a follow-up story.
– Peter Hancock, The Lawrence Journal-World
Have a tip you’d like to share? E-mail us.
READ MORE FROM:Fact-checking and accountability journalism project, How I got that fact
MORE ARTICLES ABOUT:Data reporting and visualization, Fact checking and accountability journalism, Political and government reporting