Let’s say you see a chart with two lines. An orange line starts well above a blue line — but as you move right on the X-axis, the orange line dips well below the blue line. The blue line, meanwhile, has slight peaks and valleys, but more or less remains unmoved.
The two lines are meant to represent the outcomes of a certain policy — the rolling 7-day average of coronavirus cases in counties enforcing a mask-wearing COVID-19 ordinance vs. those that aren’t enforcing it. And let’s say the chart reflects the number of cases per 100,000 population. (Totally random hypothetical, I assure you.)
The orange line represents the counties where the mandate is being enforced. The blue line represents the counties where it isn’t. You’re supposed to draw an obvious inference from this: That enforcing mask mandates leads to fewer cases of the coronavirus per 100,000 population than not enforcing mask mandates. You’re still following and this seems reasonable enough, right?
However, this conclusion rests on a couple of bedrock assumptions. First among them, I would hope, is the assumption that the Y-axes values for both lines are the same. In other words, that both lines chart the exact same numerical value. That’s the only way the orange line and blue line could be considered comparable.
But that’s not how Kansas Department of Health & Environment Secretary Lee Norman rolls.
Norman was out in front of the media for a COVID-19 update last week when he presented what might be the single most deliberately misleading pieces of coronavirus information put forth by a major elected or appointed official in the during the annus horribilis of 2020 — and it has competition.
First, though, a short bit of background.
Kansas is a reliably red state in presidential elections, but the state Democratic Party has been reasonably successful in winning statewide office, particularly for governor, going back well over a half-century. However, most of these Democrats are what you usually get when you look at Democrats who win office in the Great Plains: More eggs-and-bacon and less latte.
Gov. Laura Kelly, who managed to get elected after the state GOP imploded in the run-up to the 2018 midterms, is definitely more in the latte mold, having ushered through Medicaid expansion and now a universal mask order in a large state like Kansas, where infection rates and ability to distance can and do differ wildly.
However, according to The Week, the legislature limited her ability to enforce it so it’s up to each individual county.
Some counties have enforced a mask order, others have declined. Kansas Health Secretary Norman, in his Wednesday news conference, went out of his way to show the graph with the orange line and the blue line, each representing what I said they hypothetically might.
According to The Sentinel, a non-profit news service that’s a subsidiary of the conservative Kansas Policy Institute, it was presented as proof to Norman that the orange-line counties abiding by the Gov. Kelly’s mask protocol were “winning the battle” while the blue-line counties weren’t.
The Sentinel saw something different.
“But he deceptively doctored the chart to justify mask mandates,” The Sentinel reported.
Here’s video of the news conference. At the 15-minute mark, a reporter asks, “If the no-mask counties would start masking, would it [blue line] drop, and would it dip down below the mask counties?”
“I think it would,” Norman said.
Here’s the problem, however: The chart was wildly misleading.
The problem is that the orange and blue lines on the X-axis each had its own Y-axis values. The orange line — the cases per 100,000 in counties which were enforcing the mask mandate — had a range between 15 and 25 cases. The blue line — the cases in the counties not enforcing mask protocol — was between four and 14 cases per 100,000.
That all sounds complicated, but the Twitter post below shows the graph with the Y-axis values the same when you click on the picture on the left and how it was presented when you look at the picture at the right.
On the left, both lines are essentially similar — except the counties where the mask order is being enforced has a much higher rate than those where it’s not being enforced.
Data plotted on the same axis vs KDHE Secretary Lee Norman’s manipulated chart for comparison.
— Nathan Hill (@nj_hill) August 8, 2020
On the right, it’s clear that the graph was meant to mislead — to make people believe the mask order had gotten COVID-19 cases under control, provided it was followed.
One official of the Kansas Policy Institute, a free-market think tank based Wichita, told WIBW-TV that Norman’s chart was deceptive.
“It has nothing to do about whether masks are effective or not. What it has to do is making sure Kansans can make sound conclusions from accurate information,” Michael Austin, director of the Sandlian Center for Entrepreneurial Government at the Kansas Policy Institute, told WIBW-TV.
“And unfortunately, the chart that was shown prior in the week strongly suggested that counties that had followed Dr. Norman’s mask order outperformed counties that did not and that was most certainly not true.”
According to The Sentinel, Kansas House Majority Leader Dan Hawkins, a conservative Republican, blasted Normal and Kelly for the misleading data.
“Governor Kelly and her administration have failed Kansans time and again, but manipulating data to intentionally deceive the entire state is a new low. Tens-of-thousands of Kansans have lost their jobs and businesses as a direct result of Governor Kelly’s politics-first response to the COVID pandemic, and these individuals struggling to make ends meet deserve to know the truth,” Hawkins said in a statement.
“It is reprehensible for a public servant like Dr. Norman that we trusted to protect our health and safety in a nonpartisan way to intentionally spread misinformation. The Kelly administration has lost all credibility.”
On Saturday, Norman responded to criticism from the Kansas Policy Institute on Twitter saying, according to KSNT-TV, that “the data is clear. He said it does show that cases dropped in mask-mandated counties and remained level in non-mask-mandated counties.”
That’s not really what it shows at all — and it still demonstrates counties that don’t follow the mask mandate do better than those that do. There are alternative explanations for that. There shouldn’t be alternative Y-axis values to make those explanations.
Those are what I believe the mainstream media used to like to refer to as “alternative facts.”
This article appeared originally on The Western Journal.