Mathematics is about abstraction and proof, not data.

That’s a weird definition of a common English word. “Arithemtic isn’t mathematics.” Right. Wait. What?

It’s just silly to say that mathematicians don’t work with data, even more so to say that when they do, they aren’t doing mathematics.

The point is that Koray is correct and what you refer to in the article as “mathematics” would be better considered “statistics and arithmetic”. Mathematics is about abstraction and proof, not data.

I find your comment unintelligible, Nepenthe. Sorry. No idea what your point is.

That’s an apt point in many cases (certainly in Condell’s, which was Gabriel’s point), but not every negative reaction to such a statistic is innumerate. 1% of six hundred thousand people (the number of Muslims in the city of London) is 6,000 people. That’s a very large number of would-be murderers to have inhabiting a city. The error is not being alarmed that so many people in your city want to kill you, but in thinking that because 6,000 Muslims want to kill you, that therefore all 600,000 Muslims want to kill you, or that we should respond to the 6,000 by punishing all 600,000 as if they agreed with the 6,000 (the “kill ’em all and let God sort ’em out” strategy). It’s also erroneous to infer that because 6,000 people agree gays should be killed, that therefore those 6,000 people are actively killing gay people.

Fixed. Thanks!

I sincerely hope that you never make the mistake of asking the mathematician at your dinner table to calculate the tip. Then you will find out how relevant numbers are to mathematicians. As a rough guide, if you–a non-mathematically trained person–understand anything that’s written, see any numbers greater than 5, decimals or (*shudder*) percentages, or don’t see a tombstone or QED mark at the end of each segment, it’s not mathematics.

Alex Gabriel wrote recently ‘Almost no British Muslims – one or two percent – support execution for homosexuality.’

Some ‘sceptics’ must not understand how small a percentage one or two percent is.

Somehow this tiny percentage gets inflated by some ‘sceptics’ into a claim that Muslims are hostile to gays.

Reality check! Only 1 in 100 British Muslims want homosexuals to be killed, according to Alex’s figures, which he carefully compiled.

But people don’t understand risk when it is expressed as a numerical term.

the fact that the larger group will have more total poor members is irrelevant

Unless the number of them is relevant. Which is precisely Ross’s point. And mine. A point you seem to have missed.

FOX News could pull the same trick with obesity: if twice as many blacks as whites were obese, convincing us (wrongly) that this meant most obese people were black, therefore obesity is a black problem, therefore white people can ignore it.

That’s not irrelevant. It’s how people in the real world are being manipulated. Which is as relevant as relevance gets.

…it makes sense to regard black poverty and white poverty as two different conditions, very likely calling for two different remedies.

That is a non sequitur. It makes sense to ask and find out if there are different causes. It does not make sense to assume there are other causes.

And IMO, I suspect you would find they don’t have different causes, other than that racism suppresses (and has suppressed) black advancement significantly more than white advancement. In other words, racism is the problem, which is precisely Ross’s point. How racism has that effect is multivariate, but it all comes back down to racism somewhere in the system, either deliberate, unconscious, or institutional. That would certainly be the first hypothesis I would test…it having so much evidential support already (e.g. studies showing significant bias in hiring and promotion based on black vs. white names on resumes), compared to any alternative I know. But as a good empiricist, I’m open to evidence to the contrary.

In what way is statistics and statistical reasoning not a “real world application of math”?

But never mind that. It’s a distraction. Bizarre. But a distraction.

You skipped my point that I’m not talking about “statistics and statistical reasoning.” I’m talking about doing counts and ratios. (Look at all my examples…nary a mention of p-values or confidence intervals or anything to do with statistics as a specific branch of mathematics…although I agree that’s something we should all have some familiarity with, and it’s a part of numeracy, and I agreed with you that some mathematicians aren’t fully numerate that way).

Surely if you got a compsci degree you learned how to count things and develop and question and test ratios, and about the folly of building routines on false dichotomies or counting the wrong things, and so on. You know, the sorts of things I actually talked about.

At least, good gods, I hope they taught you that stuff!

Mathematicians cannot suck at math.

Now you are contradicting yourself. You opened by saying many mathematicians don’t know the particulars of statistical reasoning and thus suck at it (by which you mean the advanced tools used by statisticians, since that’s a particular branch of mathematics that not all mathematicians study). Now you are saying that’s impossible.

Huh?