I love it when my hometown rag, The New York Times, attempts to publish something provocative about the state of mathematics and teaching. Yes, the Times has not aged well, and I often wonder if the editors of the various lifestyle sections, including “Style,” “Home,” “Food & Dining,” and “Travel” section live in the same economic climate as the bottom 99% of us, but I do admire the investigative reporting, especially when they decide to do something like pantsing that mendacious asshole, Andrew Cuomo.
So it was with curious interest that I had a look at Elizabeth Green’s article in the July 23rd edition of the New York Times Magazine, “Why Americans Stink At Math.” Me, I love a good screed disparaging Americans as much as anyone else, although I would like to point out to Ms. Green that “America” is composed of two continents, North and South, and that even if she were referring to North America, we do share this continent with Mexico and Canada. Oh, and then there’s that issue about those people who were here before there was an “America,” but let’s not get technical, okay?
This article is clearly well-researched, and I hope Ms. Green’s book sells well and that “Chalkbeat,” the website where she is the chief executive, gets a bazillion hits. However, it appears that Green herself is pretty poor at math herself, and the Times let her get away with it. Of course, perhaps the Columbia School of Journalism doesn’t require that their students take a basic course in statistics.
Where should we begin? Perhaps the title is a good place: if you’re going to write about how bad “Americans” are at mathematics, it’s probably a good idea to get some reliable and relevant data to back it up. Unfortunately, the “data” that is thrown down in this article is, to put it politely, piss poor.
The first place where Green goes wrong is when she cites “national test results” about mathematics achievement in the U.S.. First, I wonder which “test results” Green is referencing here (you have to be suspicious when, in the days of the omnipresent interweb, a link is not included to the data supporting this point.) It may be significant that 2/3 of all 4th and 8th graders are not “proficient” in math, but again, this is a national standard, not an international standard, so this only points to the fact that U.S. children are not achieving according to some standard that was created where, in some dark cave where Dick Cheney and his family reside?
Green goes on to state that half the 4th and 8th graders taking the National Assessment of Educational Progress could not read a thermometer, or that 3/4 of the test takers could not translate a simple word problem into an algebraic expression. Note that this is the National Assessment of Educational Progress – it doesn’t say anything about whether U.S. children are better or worse than anybody else around the globe; for all we know, 7/8 of the children in Helsinki and 11/13 of the children in Ibaraki couldn’t successfully answer these questions either. Look, I’m not the sharpest pencil in the box, but even I know these numbers are insignificant without a context.
The one mention of an international comparison is that students in Massachusetts, which many of us know is a “state,” lag two years behind their counterparts in Shanghai, China, which if I understand it correctly, is a “city.” Yes, and would you be surprised to learn that the students at the Bronx High School of Science outperformed those in the village of Zhuangjiashan, China? Cherry picking data is never a good idea, except if you can use it to back up a sensational headline (or you are a best selling author.) Oh, and if Ms. Green bothered to do her research, she would have found out that the scores in Shanghai are cherry picked themselves: according to an investigative report by The Guardian, many children in Shanghai are barred from taking these kinds of tests.
Green goes on to cite how the picture does not get better into adulthood (I wonder if she is talking about her own understanding of statistics?) Green uses a 2012 study (which remains unspecified) about how U.S. adults ranked in the bottom 5 of 20 countries in numeracy. Which countries these are, we do not know. What the sample size is, we do not know. How the samples were chosen, we do not know. The only study I know of that was done comparing math proficiency in adults internationally was by the “Program for International Assessment of Adult Competencies” that compared 23 countries and put the U.S. third last. However, that result had so many methodological flaws that I had to tear it down in a separate blog post that I titled “Is this the most ignorant article about the Common Core ever written?” If Green has some better data on this, I would love to see it.
Green follows this up with some scary data about miscalculations in the medical profession, including doctors and pharmacists, just to help show how very afraid we should all be. However, this “evidence” is junky as well: does the U.S. rank higher in medical computational errors than other countries or not? Isn’t this an article that is putting forth the thesis that “Americans” are worse at math than everyone else? I wonder if other professions have higher rate of alleged innumeracy, including writers reporting on mathematics education? Ms. Green, would you mind calculating the integral of X squared for me?
I won’t even go on about the A&W burger story and fractions, besides the fact that I was unable to find a reference that it was actually true… the only mention I could find was a case study described in a blog post in 2013, which also, curiously enough, was unreferenced…. It sounds like a good story, but like many other urban legends, this might be one that stands alongside the worms in McDonald’s hamburgers?
Besides the dubious statistics Green uses to promote her cause, I would also like to take her to task by referencing the Louis CK debacle on “Common Core Mathematics.” I followed this story very carefully, and can tell you that the issue being described had nothing to do with the Common Core or mathematics education. What it really was about, in my estimation, is that parents should not really be helping their kids with math homework: it is a subject that is bound up in all sorts of emotions concerning intelligence and anxiety, and since the teaching of it changes every few years, parents inadvertently end up transferring their anxieties to their children. As someone who has worked in the field of teaching mathematics for the past 3 decades, I can tell you right now that the first piece of advice I give parents about promoting their children’s achievement in mathematics is butt out of helping with their homework assignments.
The rest of the article, when it avoided any mention of data, was a good read, and I always enjoy hearing about the work of the wonderful Magdelene Lampert, who is a legend in the world of education. Ms. Green may not know much about statistics, but you can’t beat her narrative ability; my only hope is that she’ll hire a statistician next time to help support her claims.