How Not To Be A Dumbass: Media Innumeracy Edition
I’ve not had a good “How Not To Be A Dumbass” post in a while.
I was going to put this in the next Taxing Tuesday, but this is so stupid, I can’t let it wait.
ORDER OF MAGNITUDE IS IMPORTANT
For those who don’t want to listen to the video (and it’s short), I’ll do a transcription below.
This is the stupidest thing that has ever aired on television. Congratulations to all involved. pic.twitter.com/vARi9yQ0Bv
— Timothy Burke (@bubbaprog) March 6, 2020
Mara Gay [NYT Editorial Board]: Somebody tweeted recently that actually, with the money he [Bloomberg] spent, he could have given every American $1 million.
Brian Williams [MSNBC host]: You got it, let’s put it up on the screen
[the following is put up on the screen:]
Brian Williams: When I saw this, tonight on social media, it all became clear.
[he then reads out the tweet]
“Bloomberg spent $500 million on ads. The U.S. population is 327 million.”
Brian Williams: Don’t tell us if you’re ahead of us on the math
“He could have given each American $1 million and have lunch money [sic] left over.”
Brian Williams: It’s an incredible way of putting it
Mara Gay: It’s an incredible way of putting it, it’s true. It’s disturbing …it does suggest, what we’re talking about here, just that there’s too much money in politics. And it makes it difficult because what we want in politics, the point is to have competition. And the point is to have the best candidates, and have people from all backgrounds to be able to run.
Yes, yes, thanks, Dr. Evil.
BREAKING IN: THE LESSON IS TOO MUCH MONEY IN POLITICS?
I want to break in for a second and point out that these people are on MSNBC, and maybe some cable subscribers are helping pay for the program, but I happen to believe that ADVERTISING is a huge portion of MSNBC’s revenue stream.
Am I correct?
Election Boom: Here’s How Much Ad Money CNN, Fox News And MSNBC Are Expected to Make in 2020
Kagan’s forecast, which was first reported by Variety, predicts that Fox News Channel will make $1.32 billion in advertising this year, while CNN will make $773.1 million and MSNBC $723.1 million. Each amount represents an increase from the previous year: up 7.9 percent for Fox, 11.7 percent for CNN and o.8 percent for MSNBC.
Maybe MSNBC is just sour that it falls behind CNN, though it gets better ratings.
[FTR, in looking at that article, I had to turn off the ad-blocker, because Mediaite, like many similar websites, makes money via ads]
[I don’t … but you can buy me a cup of tea….]
If they want to talk about too much money in politics, then perhaps they can talk about how their own channel feeds into that, hmmm?
But really, it’s that they’re throwing the numbers around as if the numbers have no meaning to them. They really have no clue.
COMMENTARY ON INNUMERATE TALKING HEADS
The ‘Million Dollars Per Person’ Affair Is Telling
Obviously, the math here is spectacularly off. If Michael Bloomberg had divided the money he spent on his presidential run evenly among Americans, we would each have got $1.53, not $1 million. For Bloomberg to give $1 million to each American, he would have to be worth $327 trillion (in cash), which, for context, is around 17 times American GDP and about five-and-a-half thousand times what he’s actually worth. The scale of the error here is galactic.
It’s also extremely telling. This, right here, is why so many left-leaning Americans think that “the billionaires” can pay for everything. It’s why Elizabeth Warren was enthusiastically boosted by the media despite her ridiculous pretense that she could pay for a series of gargantuan initiatives without raising taxes on anyone but the extremely rich. It’s why Democrat after Democrat promises not to raise “middle class taxes” while promising programs that require the raising of middle class taxes. How did this bad tweet make it onto TV to be endorsed? Why did Mara Gay agree with it? Why didn’t Brian Williams notice? Because the people involved in this clip thought it was true. This is how they see the world.
There are amusing comments on this Althouse thread:
Meade said…
Phase 1: Collect all of Mike Bloomberg’s $10/pair underpants
Phase 2: Redistribute
Phase 3: Every American gets 17.9 underpants
3/6/20, 8:28 AM
…..
iowan2 said…
To understand numbers you have to play with them.
Very few people “play” with numbers.
Yes, I know, but you don’t have to “play” with them. You do have to think about them, though, and many people don’t want to.
Especially if the numbers tell you unhappy stuff.
Small world, as they say…
— Spencer McCormick (@spmmccormick) March 6, 2020
The inability to grasp large numbers is also, as
charlescwcooke</a> pointed out, how so many Sanders/Warren supporters believed that we could fund a welfare state that consumes 60-70% of GDP almost exclusively by taxing the top 1%, who collect <20% of national income.</p>— Megan McArdle (asymmetricinfo) March 6, 2020
I have tried to explain “we ain’t got the money of” multiple times on multiple topics.
We definitely don’t have enough in rich folks to be able to tax them, whether by income or wealth.
But first, you need to show you actually know some numbers if you’re going to talk about them.
NUMBERS ARE NOT MAGIC, SO LEARN SOMETHING BEFORE DISCUSSING THEM
I am a numbers person, and I live in a world of numbers.
I have seen bullshit like this when it’s “Oh, the CEO should get their pay cut to under $1 million, and then the rest goes to the workers!”. When the amount being redistributed is something like $30 million for a year, and the company in question has 300,000 workers… they’re talking $100. For a whole year. That doesn’t really move the needle on wages.
People don’t understand why corporations may not pay taxes in a given year, when they have so much revenue! People don’t understand that corporations pay a hell of a lot of money in payroll taxes directly to the government, and then their employees pay a lot in payroll and income taxes. For most companies in a service business, their biggest expense is employee pay and benefits.
I have had to deal with non-numbers people who pretend numbers are magic. They never developed a sense of how much anything actually is.
No, that’s a lie. They know if they’ve been cheated on Amazon, most likely. Because that involves numbers they understand – they can tell if they’re being charged $50 for something really worth $30.
I don’t blame people for not understanding abstract concepts such as reserves or Deferred Acquisition Costs (which are amortized). That gets complicated.
I don’t even mind that most people do not understand the number one billion, or, even worse, one trillion. Which is where we’re going when we discuss federal finances.
The most I’ve really gotten in concrete terms, myself, is the number one million, because when I was a kid, my dad let me use his dot matrix printer to print out one million Xs (I wrote a BASIC program!) and then take the printout to school. Because it was that continuous feed of paper, we stretched it outside in the field. And we all got to see how much one million was.
If I remember correctly, I printed 5000 Xs per sheet… so I needed 200 sheets. 200 sheets of paper, at 11 inches per, is 183 feet or so. We had to use the whole class to stretch it out.
I had a very indulgent dad. Oh, and I was a weird kid.
But come on, if you had the chance to print out one million Xs (and somebody else was paying for it), wouldn’t you go do it?
Okay, that may explain a lot of things about me [and I have a story about the electronic voting program I wrote when I was 10], but back to our story.
KNOW YOUR LIMITATIONS
I have written about this before, with respect to just saying “I don’t know” when you don’t know, but the perennial problem people are the ones who think they know something about a topic when what they “know” is inane.
I’m here to help, by the way.
From my own experience of teaching math at the college level, I’m sorry to say that, no, your high school math scores are not a good indication of how well you know math or can reason quantitatively.
I do know many people forget the math they knew (no big deal), but more than that, the way math has been taught, a lot of people get very good grades while “learning” recipe lists. I remember the look on one prof’s face, teaching freshman calculus at NYU [I was taking notes for the text he wanted to write], and having a student ask why L=1 couldn’t be a solution to L = 2L.
I think he hadn’t taught non-majors in a long time.
I have dealt with lots of people who had good math grades in high school, and had to be re-taught basic geometry and algebra. I’ve also known people with crappy high school math grades, but managed to pull it together in college. The former were far more frequent than the latter.
WHAT YOU NEED TO KNOW ABOUT MATH
I will stick to just understanding quantitative reasoning. Usually, people call this “math” as an umbrella term, but it’s a hyuuuuge umbrella. It’s okay not to know every little nook & cranny of math.
But you should know some basics.
From the accompanying post, I mention the following:
What math do the journalists really need to understand?
They need to understand:
- how basic statistics work: mean, median, mode, standard deviation, and sample sizes
- the difference between percentage increases and percentage point increases
- The distinction between rates and absolute amounts
- how to read a simple income statement and balance sheet (and what the differences are)
- how units work in general (like miles per hour, etc.)
There is more, but I’m talking about some very basic ideas.
I didn’t think that I needed to talk about the following:
You need to understand:
- orders of magnitude (the distinction between million, billion, and trillion, for example)
- HOW TO DIVIDE
- How to “strike out” orders of magnitude, as long as they match up.
- Oh, and when to divide, because somebody seems to have figured the original tweeter just subtracted the two numbers, instead of divided them.
- Knowing the meaning of addition, subtraction, multiplication, and division in general is good. It’s fine to punch the numbers in a calculator. Okay, maybe not. But I will return to that another time.
Oh, as an aside, Ace of Spades relays the thought that some producer passed the tweet along, thinking a “HAHA WHAT A DUMBASS” conversation would ensue. I don’t think an experienced TV producer would think any such thing would ensue.
That they went in the direction of “Hey, ad buyers, don’t buy ads on our network! Clearly we think it’s wasted money!” is doubling down on a stupid I wasn’t expecting.
Yes, Mara Gay was talking about having people from all backgrounds being able to run, but the point of the original tweet was that $500 million couldn’t bag Bloomberg more than a few delegates. And that the money would have been better used raining down upon the American people in general, as opposed to specific people like Brian Williams and Mara Gay, whose salaries are paid by ad revenue.
It wasn’t just the math they weren’t thinking about.
PICTURING ORDERS OF MAGNITUDE DIFFERENCE
Here’s an example of comparing orders of magnitude — $250 million versus $5 billion.
Now, with “million” and “billion”, note that I changed the $250 million to $0.25 billion. If I want to know how many times $250 million goes into $5 billion, I can either write it out like this:
or
5 billion divided by $250 million = 5,000 million divided by 250 million = 5000/250 = 500/25 = 100/5 = 20
Basically, I pair up “factors” from the dividend (5 billion) and the divisor (250 million) at each step, and just cancel them out. That’s how that works.
That they took $500 million divided by 330 million and got an answer that was in the millions… THEY SHOULD HAVE KNOWN SOMETHING WAS WRONG.
[Again, maybe none of the people involved thought about dividing. Just subtraction.]
DEVELOPING NUMBER SENSE
What I’m calling for is to develop number sense, along with all sorts of other basic skills (like communicating in written form, which is not natural), as a functioning adult, and especially as anyone wanting to make any sort of public comment regarding quantities.
Mainly, you need to be able to explain numbers you see around you, and be able to determine whether or not those numbers make sense. Do you understand what they mean?
What I used to encourage my own students to do, for my non-actuarial science students, would be to look at news stories that had numbers in them — once you start deliberately looking for it, you’ll start noticing it. It doesn’t even have to have a lot of numbers in there.
Example, here is an article I just read:
India coronavirus cases spike, Switzerland has first death
Twenty-nine people have contracted COVID-19 in India, including three who have recovered, CNN reported. Many of the cases are linked to a group of travelers from Italy — the hardest-hit European country, according to Health Minister Harsh Vardhan.
Thoughts:
- Is the “spike” just going from 0 to 29 confirmed cases?
- How many of those 29 were linked to that group of travelers from Italy (and how many people were in the group of travelers?)
So, I start out with questions, but these aren’t answered in this article. [nor in the linked CNN article]
Meanwhile, Switzerland has reported its first coronavirus-related death, authorities said Thursday.
She was a 74-year-old woman — a “high-risk patient suffering from chronic disease” — who had been hospitalized since Tuesday, authorities said.
A total of 58 infections have been confirmed in Switzerland.
Okay, out of 58 [confirmed] infections, one has died, giving us a raw case fatality rate of 1/58 = 0.017 = 1.7%.
I would think about reasonableness of that 1.7% — that’s about 2%, which is the same as 1/50, so that calculation “smells” right.
Then there’s the “compare against reality” aspect – how much does this compare to the death rates we’ve been hearing thus far? It’s in the range.
The country has close relations with its neighbor Italy — where about 3,100 people have been infected and 107 have died — as well as France and Germany, where coronavirus cases have also been reported.
Okay, let’s calculate 107/3100 = 0.034 = 3.4%… that’s twice 1.7%, but when it’s only one person, and you don’t know if you’re missing cases, well, it’s the right order of magnitude.
Also, I know it’s in the right range because 107/3100 looks like 100/3000 = 1/30 which is 3.33% so yay.
Anyway, you get the drift.
Yes, I’m generally not going into this level of detail, but I don’t really need to develop my number sense. I’m mainly subconsciously letting the numbers bubble behind my eyeballs. I don’t always catch minor errors, but order of magnitude errors, yes.
IT’S OKAY TO BE WRONG, IF NOBODY IS PAYING YOU HUGE AMOUNTS OF MONEY FOR YOUR WRONG THOUGHTS
The main thing to remember, though, is that it’s fine to be wrong. Keep working at it.
But you’re probably not getting paid huge bucks to blather about stuff you understand nothing about.
Brian Williams and Mara Gay are probably safe, in any case…. unless those sweet, sweet ad dollars really get put at risk. Then they’d better watch out.
PRIOR DUMBASSERY
Behold the stupidity:
January 2019: Math Stupidity: Comparing Pizzas and What’s Important
April 2018: Divestment Dumbassery: Know What You’re Protesting
Jan 2017: How Not to Be a Dumbass, Take N: It’s Okay To Not Know Something (or Somebody)
June 2015: Down with the Struggle: Survey of Current Dumbassery
March 2015: How Not to Look Like a Complete Ass When Doing Your Job… AS A REPORTER
January 2015: How Not to Look Like a Complete Ass: Saying “I Don’t Know” When You Don’t Know
October 2014: How Not To Look Like a Total Ass
October 2014: How Not To Look Like a Total, Desperate Ass, Part 2
Fun.