Media and Data
Most of you know I have a significant contempt for television, and not a little for print and internet media as well. My complaint is not so much what is labelled 'bias', because any point of view can be called bias. rather, it is the trumped claims, sensationalist language, innuendos, epithets and exaggeration that bother me.
I just read a CNN story about H1N1 deaths. The headline is 'H1N1 deaths as high as 17,000'. That sounds awful, doesn't it? You have to read the article and have some statistical knowledge to put the number in perspective. The headline insinuates that H1N1 is dangerous. Most people (me included) have never seen 17,000 dead people at one time, so it sounds terrible.
In reading the article, the typical death toll from seasonal flu is 36,000! Put another way, the headline would have been just as truthful, but with totally different implications, if it had said 'H1N1 death toll half of normal flu season'. If you read a bit, you find the CDC is estimating the death toll, since flu cases are notoriously under-reported. The range calculated by the CDC is 8,300 to 17,000. It is just as statistically likely the death toll was 8,300 as it is 17,000, but which number is in the headlines?
When you read a little more, you find out that the number of REPORTED deaths is only 2498! The actual data we have is 2498 deaths! The rest is statistical projection, with a very broad range. When a range is that wide, it means the result has a large degree of uncertainty.
So CNN made a headline splash with the highest end of a wide range, saying 'up to 17,000 deaths'. The reality is 2498 confirmed H1N1 deaths, and a range estimated at 8,300 to 17,000. Most people don't understand statistics well enough to interpret the data range, and lots of people glance at a headline, don't read the article, and think there have been 17,000 deaths from H1N1, which sounds terrible.
Of course, politicians spin data even worse than the media. This is why I make a point of reading different points of view, and trying to get raw data where I can. I think I'm smart enough to look at the data, and make my own conclusions, then compare mine with the political headlines, and I think most citizens who read are smart enough also.
A favorite story that reflects the issues of spin and label comes from the original Star Trek series, an episode called 'Journey to Babel'. Spock, Kirk, McCoy and Spock's parents are at a dinner reception, and McCoy asks Spock's mother if he ever had a favorite toy as a child. She says he had a celot he was very fond of. When asked what a celot is, she replies it is a sort of Teddy Bear. She leaves, and McCoy looks at Spock with a grin (perhaps more of a leer) and says 'A Teddy Bear?' Spock, utterly flat faced, says 'Not quite doctor. On Vulcan, the Teddy Bears are alive, and they have 6 inch fangs'.
I always think of that reply when reading overblown, distorted headlines, and obvious political spin, where some self-aggrandizing 'public servant' makes a mountain of cotton candy out of 2 tablespoons of sugar. Not quite, doctor.
In closing, let us consider the data on our stimulus package of last year. I think everyone knows it was a total of 787 billion dollars. When it was enacted, the unemployment rate was 7%,and it was claimed it would prevent unemployment from reaching 9%. Really? That claim was missed. Some would suggest the stimulus was a failure on that basis, and we have certainly seen articles to that effect. It's impossible to know what would have happened without the stimulus, although a good economic argument can be made that running enormous deficits to treat recession just reduces economic recovery (not here, not now- we'll talk later). The administration has taken to saying the stimulus created or saved 2 million jobs.
Where does that piece of spin come from? From a theoretical calculation of employment based on economic activity. In making such a calculation, what is the assumed impact of the stimulus, knowing that governmental spending has a much lower impact than private spending? bets it was treated like private money in the economy, and ignores the fact that, without 787 billion dollars of debt, the economy might have trundled forth activity of its own.
So, I always look up data when I can, and read different points of view about the same event or claim. Eventually, reality does emerge, although finding it is like doing archeology on the present.
Ted
Most of you know I have a significant contempt for television, and not a little for print and internet media as well. My complaint is not so much what is labelled 'bias', because any point of view can be called bias. rather, it is the trumped claims, sensationalist language, innuendos, epithets and exaggeration that bother me.
I just read a CNN story about H1N1 deaths. The headline is 'H1N1 deaths as high as 17,000'. That sounds awful, doesn't it? You have to read the article and have some statistical knowledge to put the number in perspective. The headline insinuates that H1N1 is dangerous. Most people (me included) have never seen 17,000 dead people at one time, so it sounds terrible.
In reading the article, the typical death toll from seasonal flu is 36,000! Put another way, the headline would have been just as truthful, but with totally different implications, if it had said 'H1N1 death toll half of normal flu season'. If you read a bit, you find the CDC is estimating the death toll, since flu cases are notoriously under-reported. The range calculated by the CDC is 8,300 to 17,000. It is just as statistically likely the death toll was 8,300 as it is 17,000, but which number is in the headlines?
When you read a little more, you find out that the number of REPORTED deaths is only 2498! The actual data we have is 2498 deaths! The rest is statistical projection, with a very broad range. When a range is that wide, it means the result has a large degree of uncertainty.
So CNN made a headline splash with the highest end of a wide range, saying 'up to 17,000 deaths'. The reality is 2498 confirmed H1N1 deaths, and a range estimated at 8,300 to 17,000. Most people don't understand statistics well enough to interpret the data range, and lots of people glance at a headline, don't read the article, and think there have been 17,000 deaths from H1N1, which sounds terrible.
Of course, politicians spin data even worse than the media. This is why I make a point of reading different points of view, and trying to get raw data where I can. I think I'm smart enough to look at the data, and make my own conclusions, then compare mine with the political headlines, and I think most citizens who read are smart enough also.
A favorite story that reflects the issues of spin and label comes from the original Star Trek series, an episode called 'Journey to Babel'. Spock, Kirk, McCoy and Spock's parents are at a dinner reception, and McCoy asks Spock's mother if he ever had a favorite toy as a child. She says he had a celot he was very fond of. When asked what a celot is, she replies it is a sort of Teddy Bear. She leaves, and McCoy looks at Spock with a grin (perhaps more of a leer) and says 'A Teddy Bear?' Spock, utterly flat faced, says 'Not quite doctor. On Vulcan, the Teddy Bears are alive, and they have 6 inch fangs'.
I always think of that reply when reading overblown, distorted headlines, and obvious political spin, where some self-aggrandizing 'public servant' makes a mountain of cotton candy out of 2 tablespoons of sugar. Not quite, doctor.
In closing, let us consider the data on our stimulus package of last year. I think everyone knows it was a total of 787 billion dollars. When it was enacted, the unemployment rate was 7%,and it was claimed it would prevent unemployment from reaching 9%. Really? That claim was missed. Some would suggest the stimulus was a failure on that basis, and we have certainly seen articles to that effect. It's impossible to know what would have happened without the stimulus, although a good economic argument can be made that running enormous deficits to treat recession just reduces economic recovery (not here, not now- we'll talk later). The administration has taken to saying the stimulus created or saved 2 million jobs.
Where does that piece of spin come from? From a theoretical calculation of employment based on economic activity. In making such a calculation, what is the assumed impact of the stimulus, knowing that governmental spending has a much lower impact than private spending? bets it was treated like private money in the economy, and ignores the fact that, without 787 billion dollars of debt, the economy might have trundled forth activity of its own.
So, I always look up data when I can, and read different points of view about the same event or claim. Eventually, reality does emerge, although finding it is like doing archeology on the present.
Ted
As a long-ago graduate of the UT School of Journalism, I agree with you that sensationalism and distortion of facts seems to be the norm now. My favorite professor, now deceased, is probably spinning in her grave. The others are still living, and I'm sure they hate it like I do. And don't get me started on the piss-poor job of reporting done locally. When we had two newspapers, the competition resulted in great reporting, writing and editing. Now, it is truly pathetic and embarrassing.
ReplyDeleteAs one of my college history professors once said, "Statistics don't lie, but liars use statistics." That's a good aphorism to remember, even when reading medical literature.
ReplyDeleteDavid Graham