Oh no...
I found out yesterday that I am becoming someone I never wanted to be. Specifically, I wrote a letter to the editor.I believe that I wrote about this article before, but there is an infuriating article by Duncan Greenberg on billionaires.
The letter I wrote is a little, well, snarky. I hate to be the person who hides behind the anonymity of email. On the other hand, I decided to be snarky because I think that Mr. Greenberg is very slightly dangerous. I feel justified.
Here it is:
Dear Editors,
I just read Duncan Greenberg's article "A Recipe of Riches" online at http://finance.yahoo.com/career-work/article/107927/a-recipe-for-riches.html?mod=career-leadership. I have two comments:
First, it seems like Mr. Greenberg should provide a citation when he states that mathematical prowess is hereditary. If it is true, then I would like to read the study that concludes this. If false, Mr. Greenberg is perpetuating a stereotype that some people "just aren't good at mathematics," and is helping young people to give up in mathematics classes by giving them genes to scapegoat.
Second, even if mathematical prowess is hereditary, it appears that Mr. Greenberg did not inherit this gene. "Want to become a tech titan or hedge fund tycoon? Up your chances by dropping out of college..." Mr. Greenberg goes on to state that 14% of billionaires dropped out of college. This means that 86% of billionaires--over six times as many that dropped out--finished college. Really? You can up your chances of being a billionaire by doing something that almost 9 in 10 billionaires specifically chose NOT to do?
I sincerely hope that no children read your magazine. Duncan Greenberg just gave them false excuses to give up in math class and--worse yet--drop out of school.
Sincerely,
Dirk Awesome
posted by Dirk Awesome @ 3:34 PM
5 Comments:
I think you did the mathematical community a service by challenging Duncan.
However, both of you made a Bayes-theorem error. You discussed the probability of being a college dropout given that you're a billionaire, but the phrase "up your chances by dropping out of college" should be based on the probability of being a billionaire given that you drop out of college (which is probably much lower).
Thanks, Aaron. While I grant that Mr. Greenberg made an error, I cannot find any sort of assertion that I made (other than Mr. Greenberg not being good at math). I was actually trying to avoid making any sort of statement about how to determine the probability of this---only to mock Mr. Greenberg for his method.
I agree, though, that it would make the most sense to compare "the probability of becoming a billionaire given that you finish college" to "the probability of becoming a billionaire given that you do not finish college." If this is the case, the former is certainly higher, since you would divide the number of billionaires with college degrees (86% of all billionaires) by the number of college graduates (which is only roughly 30% of Americans), which is bound to be larger than the latter since 86% > 14% and 30% < 70%.
How did I do at weaseling out of owning up to making a mistake? I honestly don't think that I made any assertion about a proper way to compute the probabilities (although I did suggest that it was sketchy by using not rigorous reasoning).
You challenged Greenberg's quantitative literacy (comparing 14% to 86%) without challenging the more fundamental logical mistake that makes *any* such comparison not support his claim. But I agree that you didn't actually make the mistake yourself.
I can agree with that.
Actually, you were more right than I initially gave you credit for. When I first wrote the email, I thought for a split second about writing, based on the 86% and 14% numbers, that "you are actually six times more likely to become a billionaire if you DO finish college." I quickly realized that this reasoning was flawed, and changed what I was going to write. I think that I was a little defensive because I wanted credit for this realization. I had also very superficially thought a bit about how the number of college graduates and non-college graduates plays a role, but I decided that relative populations worked in my favor (I was right, but I didn't think through it as much as I should).
However, you are correct that I never really examined what evidence would be needed to support. I should have done this, and specifically refuted Greenberg's base premise. That would have done the most good (assuming any good at all comes from this email).
Thanks, Aaron.
Post a Comment
<< Home