It’s not an understatement to say that the corporate news media failed the American public in almost every way possible during the Covid-19 pandemic.
From actively suppressing information about the origins of the virus … (while the Wall Street Journal is busy pretending it “broke” the Wuhan lab leak story this week, alternative outlets like this one released the very same information over a year ago) … to politicizing, suppressing and ridiculing studies showing both Hydroxychloroquine and Ivermectin as promising early treatment methods (see my post on that here; also watch this Tucker Carlson interview about it with one of the top five most published medical researchers in the nation) … U.S. “journalists” turned, seemingly overnight, from friend to foe of the American people when Covid hit.
But there’s one other way the media failed the American public during the pandemic: by not properly processing, managing or publishing the data we all needed to truly understand the disease and its actual dangers. By failing to do even the simplest of jobs correctly – share real, straightforward and meaningful information on a serious disease – the U.S. news media has managed for more than year now to stoke fear, create mass confusion and, in the process, pit Americans against one another.
I found a solid example of this while searching online recently for a simple summary of current national Covid data – cases, death rates, etc. It came in the form of a CNN Health webpage, titled: Tracking Covid-19 cases in the US.
The page itself is a product of something people in the news industry like to call “Data Journalism;” this basically refers to journalists who research, read and interpret data and then report on it, sometimes in a regular article, but oftentimes in the form of charts and graphs, like the ones we’re looking at on the CNN page. But the big question I often find myself asking after reading a piece containing data journalism is this: Is the data the media’s sharing in this piece useful? Or is it overwhelming, confusing and / or misleading? For my money, this CNN webpage is an excellent example of Covid-19 data that’s all three - “overwhelming, confusing and misleading.” Let’s take a look.
The first thing you see when you get on to this CNN Health webpage is a very colorful map showing every county in the nation and the number of Covid-19 cases and deaths in each one:
Take a minute to click through it – it’s a pretty impressive compilation of data on its face: both cumulative and current (7-day average) numbers of Covid-19 cases and deaths in every single county in the nation.
But when you start clicking on each of the counties, you start seeing some interesting things. First off, you see that the total population of each county is never included anywhere. Why? Wouldn’t that be a pertinent number to know when trying to provide context for all these other numbers?
I’ll give you an example. Let’s look at Lorain County, Ohio:
When you divide All-time Covid deaths (489) by All-time Covid cases in Lorain County (25,406), the percentage of people dying from Covid there is 1.9%. But when you divide All-time Covid deaths (489) by the total number of people in Lorain County overall (309,833, per this official US Census estimate in 2019), the percentage of people dying from Covid in Lorain County is quite different. It’s 0.157% of the population - way less than one percent. Neither number is wrong, by the way. It’s just interesting that CNN only chose to focus on one and not both. Why?
The second thing I noticed when looking at this CNN page was that, instead of just showing us the number of cases and deaths in every county, it also shows us the number of cases “per 100k” people and the number of deaths “per 100k” people. I wasn’t sure what those “per 100K” numbers meant or why they were important, so I did a search on: “Cases per 100k.” I found two articles that (sort of) explained it.
The first, from this U.S. News & World Report article, explained why they think we need to know about “cases per 100K:”
“The COVID-19 case rate, or the number of cases of the disease caused by the novel coronavirus per 100,000 people, can be helpful in assessing to what extent the virus has taken hold in a community. It also is useful when comparing the coronavirus’ spread in different communities of different sizes, as an area with a high number of cases but a large population might have a similar rate to that of an area with a low number of cases and a much smaller population.”
The second, from this local NBC affiliate website in Columbus, Ohio, explained how they calculate “cases per 100K:”
“State health officials calculate cases per 100,000 people by adding up the onset cases of the previous 14 days, dividing it by Ohio’s 2019 population (11,689,100) and then multiplying that result by 100,000.”
So… call me crazy, but this all seems really, overly complicated. Wouldn’t you achieve the same thing if you just showed people the total population of each county? That way, they could divide the number of cases by the total population to see what percentage of that county’s population was catching Covid-19. Why all the complex rigamarole about “cases per 100K” instead?
Maybe journalists are trying to put every county on equal standing. Maybe statisticians or virus specialists want to know the case rate per 100,000 people because that’s a significant number in their worlds. But for the average person, like you and me, all of this extraneous data produces one thing: confusion.
Let’s set the numbers aside for a minute and just look at the colors on the map instead. Maybe those will provide more clarity. Did you notice all of those dark-colored squares? Based on the key at the top of the chart, my assumption would be those dark squares would show us places with high percentages of death rates.
Want to know something really bizarre, though? When you start doing the math, you start realizing that sometimes the death percentages in darker counties are higher than in lighter counties, but IN MANY CASES, the death percentages in the darker counties are the same as, lower than or not much different at all from the death percentages in neighboring light-colored counties. So, while I appreciate all of the work that went into putting the data here, what, exactly, are the colors on this map actually showing us, other than a vibrant mosaic picture of the U.S.?
Here’s another interesting thing to do: Put this map in “death rate mode” instead of “case rate mode.” The first thing to note here is the change in the key at the top. Even though the colorful bars are exactly the same size in both modes, the case rate categories are measured in thethousands (“less than 6,083,” “6,083 – 9,045,” etc.) and the death rate categories are measured in the hundreds (“less than 128.9,” “128.9 – 215.5,” etc.). I’m no expert on chart-making, but it seems to me that people who are skimming this chart likely will NOT notice this apples-to-oranges discrepancy, and instead will likely come away thinking case rates and death rates are happening in equal numbers. At the very least, this is confusing. But it’s also starting to feel a bit misleading, too.
*Also, while in death-rate mode, I noticed something else fascinating – I noticed that about 30 counties across the country had zero deaths (FYI I did not count the state of Utah, which seems to not be reporting information at all). I know most (if not all) of these “zero death rate” counties are likely in rural areas, but still - that’s fascinating! You know what would make a great story for a newspaper or magazine? These counties! Why have they suffered no deaths? What is their secret? What have they done differently from the rural counties that suffered way more deaths? But – strangely – no one in national media seems to have written anything about any of them. Why?
There are two more parts of this webpage that I find fascinating. The first is the section labeled: DAILY NEW CASES OVER THE PAST 14 DAYS. Let’s start with the statement at the top:
“The charts below show the number of new reported cases for each state, the District of Columbia and Puerto Rico over the past 14 days.
Gray bars
represent the number of daily new cases. The
dashed red lines
shows the seven-day moving average.”
So... right off the bat I have a question. I get everything they’re saying in that description except for one thing. What on earth is a “seven day moving average?” It sounds very fancy, doesn’t it? But nowhere on this page do we learn what it actually means. I know what a regular seven-day average might be; what’s the difference between that and a “moving” average?
The other thing I find interesting about this section is this: When you quickly glance at those little charts of each state, they appear to be comparing apples to apples, don’t they? But look closer, at the numbers on the left side of each chart. Some of them measure case numbers in the hundreds, and some of them measure case numbers in the thousands.
Why is this a problem?
Let’s take Louisiana and Maine as an example. Glance at them side-by-side and it looks from afar like Maine had MANY more cases than Louisiana. But if you look at the numbers more closely, you’ll realize that the opposite is true - Louisiana actually had way more cases than Maine, since Louisiana had cases measuring in the thousands while Maine’s only measured in the hundreds.
This is no longer just confusing – it’s flat out misleading. And again, none of this is relative to total population in both states, so we have no idea of the bigger picture here. Are there just more people in Louisiana than in Maine? Or did more people get sick – a larger percentage of the population - in Louisiana than in Maine? Without context (again – the overall population in each state), we have no idea which is true.
If you weren’t confused enough by the graphs and charts on this webpage, don’t worry - the icing on the cake of this whole webpage is that there is also a bizarre, eight paragraph narrative running through the whole thing as well, which includes totally out-of-date statements in it, like:
“Despite hopes of mass vaccination in 2021, the pandemic has continued to worsen.”
and
“With many hospitals filled to capacity, health experts are begging US residents to wear masks and stay within their social distancing bubbles until most people are vaccinated.”
Hello!
Contrary to what this crazy thing says, most hospitals today – May 28, 2021 - have few or no Covid-19 patients, mask mandates nationwide are in the process of ending and – as we all likely read in almost every news outlet in the nation this week - 50% of the adult population has been fully vaccinated. Oh – and Reuters reported Monday that the “U.S. Reports Lowest Number of New Covid-19 Cases in Nearly a Year.” Why on earth would CNN keep information from months ago on a webpage that is otherwise updated regularly? Are they lazy? Incompetent? Or are they making a last-ditch effort to keep you terrified of Covid-19? Whichever answer you choose, CNN isn’t looking too good right about now …
Maybe the most interesting thing about this webpage, though, is the information quietly sitting in the middle of it, almost like a buried lede. It has no fancy graphics or pretty, colorful mosaic charts. It’s simply a list, titled “Reported Cases and Deaths,” which is organized by state. It tells a pretty fascinating story to anyone who can do simple division problems. It tells the story of how every single state did during the pandemic overall. So, why on earth would CNN not make fancy graphs and charts or maps for this information, too?
I’m not sure what someone less cynical than me would say, but – speaking as someone who’s been watching up close as media people contort facts for nearly three decades now – I’d like to make a suggestion. I’d like to suggest that this data, for anyone who can do simple division, shows something that CNN, for some reason, doesn’t want you to see too clearly. It shows that states with severe lockdowns and mask mandates did not fare better overall than states without severe lockdowns and mandates. Want an example?
Remember all the fuss in the national media about how horrible Florida was for being relatively open all this past year compared to New York? Divide the number of overall deaths in Florida by the number of cases and you get a 1.58% death rate. But if you do the same with New York, something interesting happens. You get a 2.53% death rate. Do it with California and you get a 1.66% death rate. Do it with Michigan and you get a 2.04% death rate…
Are you starting to see a pattern here? Are you starting to understand that presenting the information in this section as an eye-catching chart or graph would appear to show that lockdowns and strict Covid rules either didn’t affect the Covid death rate much at all, or actually may have made it worse? I’m not saying that these things are definitively correlated; what I am saying is that this data, when presented in a very simple and straightforward way - with none of the colorful or complicated distractions of charts or maps - certainly makes a strong suggestion in that direction.
I could go on but I think by now I’ve made my point: Where Covid-19 data is concerned, the corporate news media is not your friend. When they use complex charts, graphs and data “journalism” like this, they are almost always hiding something or trying to distract or confuse their consumers.
The good news is - just knowing this is half the battle. My hope is that, moving forward, we can all help others see through the charade.