Tag: statistics

Damned statistics

At a conference talk on the “reproducibility crisis” in psychology, the speaker quoted a relative issuing the commonplace anti-statistics apothegm “You can prove anything with statistics”. It’s a funny sort of claim, because it is self-undermining. Outside of a seminar on Popperian scientific philosophy no one would say “you can prove anything with numerology” or “you can prove anything with astrology”. Those who are not in thrall to these methods of divination find them either entertaining or ridiculous, but 

Is it because statistics is too abstruse for ordinary people to criticise? No one says, you can prove anything with quantum mechanics. Or, for that matter, mathematics.

Statistics is sufficiently precise and rigid and generally reliable to be authoritative, but leaves enough flexibility for experts to disagree and for manipulative misapplications to still hew close to standard procedure, and sufficiently abstruse that most people can’t figure out whether they’re being manipulated.

An interesting parallel is the Shakespearean dictum “The devil can cite scripture for his purpose.”

Don Quixote on sampling bias

Continuing my series on modern themes that were already thoroughly treated in Don Quixote, here is the passage where Don Quixote and Sancho Panza discuss whether it is better to be a knight errant or a monk:

“Señor, it is better to be an humble little friar of no matter what order, than a valiant knight-errant; with God a couple of dozen of penance lashings are of more avail than two thousand lance-thrusts, be they given to giants, or monsters, or dragons.”

“All that is true,” returned Don Quixote, “but we cannot all be friars, and many are the ways by which God takes his own to heaven; chivalry is a religion, there are sainted knights in glory.”

“Yes,” said Sancho, “but I have heard say that there are more friars in heaven than knights-errant.”

“That,” said Don Quixote, “is because those in religious orders are more numerous than knights.”

Rapid growth

A lot of EU citizens who live in Britain are worried that they will be forced out if the UK voters decide next month to withdraw from the EU. The Leave campaign dismisses this, and all concerns that anyone might have about this radical step, as “Project Fear”:

Clearly any EU citizen that is legally here if we come out of the EU would absolutely have the right to remain here. Any other suggestion is just absurd.

Given that the main point of Brexit is to reduce immigration from the Continent, and given that tempers are likely to flare when the fate of said migrants (on both sides) are negotiated, and given that current UK law clearly would not give most of the EU citizens who are here the right to permanent residency, it’s clearly not absurd to worry. To adapt an old saw, even those whom political campaigners are trying to make paranoid, have real reasons to worry.

Well, from the NY Times, here’s some non-evidence:

Rose Carey, the head of immigration at Charles Russell Speechlys, a global law firm based in London, said she had seen an “unprecedented amount” of applications for British citizenship in the last few months.

“Historically, E.U. nationals didn’t really bother applying for a British passport,” she said. “It used to be a couple hundred a year to now five queries a week.”

From a couple of hundred a year to five a week — that’s pretty rapid growth!

Don’t do the maths!

Journalist Simon Jenkins has launched a broadside against the teaching of maths in school, or at least against taking it seriously. He goes further than Andrew Hacker, who argues prominently for a focus on more concrete mathematical skills.

No one would argue that pupils should not be able to add, subtract and multiply. But I studied higher maths, from calculus to number theory, and have forgotten the lot. All the maths I have needed comes from John Allen Paulos’s timeless manual, Innumeracy. It is mostly how to understand proportion and risk, and tell when a statistician is trying to con you.

Presumably, once you know how to count to 1000 you’ve learned enough. (I’m wondering about this claim about his having “studied higher maths”. At least according to Wikipedia his university subjects were philosophy, politics, and economics. Now, I have no doubt that some people can learn very advanced mathematics in their spare time and understand it wonderfully. I wouldn’t even object to them saying they had “studied” the subject. But if your private study of mathematics left you with no memory of what you thought you had learned, that suggests that perhaps the fault was in your mode of study, and not in the subject. It’s rather like someone who says, “There’s no point learning to swim. I spent years on it, and I still can’t cross a pool without drowning.”

And why is it that statisticians are always accused of trying to “con” people? Is it that statisticians are particularly dishonest? Or is it that statisticians make things sufficiently clear that you can see where you might disagree with them. What subject would you study to understand when a journalist is trying to con you? There isn’t one, because the journalist’s con is ambiguous, and for the most part his claims are clouded in rhetorical smog.

Then there’s this:

I agree with the great mathematician GH Hardy, who accepted that higher maths was without practical application. It was rather a matter of intellectual stimulus and beauty.

Now, GH Hardy was indeed a great mathematician. He probably knew more about higher maths, from calculus to number theory, than even Simon Jenkins in his prime (before he forgot everything). But I think we can also agree that the man who wrote in 1940

No one has yet discovered any warlike purpose to be served by the theory of numbers or relativity, and it seems unlikely that anyone will do so for many years

did not have the most acute vision of the scope of mathematical application. In any case, Hardy’s goal was not to argue for or against the potential utility of mathematics, but rather to defend mathematics against the charge of uselessness — basically, to defend it against people like Jenkins.

Any league table that has China at the top, Britain at 26th and America at 36th tells me something more important than merely who is good at maths. If the US and Britain – among the most vigorous economies and most successful at science – are so bad at maths, it suggests their young people are applying themselves to something more useful. Chinese students are rushing to British and US universities to join them….

Maths is merely an easy subject to measure, nationally and internationally.

 

I am reminded of a bumper sticker I saw in Florida, responding to the popular boastful messages that parents would paste on, saying “My kid is an honor roll student at Dingdong Middle School”; the response said “My kid can beat up your honor roll student.” This is that bully-boy bumper sticker expanded to a national scale. Let the inferior races waste their time on mathematics. Our kids will learn how to be “vigorous” and kick their asses.

Jerome Karabel’s wonderful book The Chosen describes how elite universities in the US in the first half of the 20th century, dismayed at how the meritocratic elements of their admissions process were being abused by Jews, who were simply outperforming their gentile compatriots on admissions tests, leading to the freshman class at Harvard in 1922 being more than 20% Jewish. The response, driven by fear that Jews would “drive away the Gentiles” (in the words of Harvard president A. Lawrence Lowell) was to de-emphasise quantitative measures and tests, in favour of the all-important “character” of applicants, a quality husbanded mainly by WASP families in exclusive boarding schools.

There’s kind of a Nietzschean flavor here: Mathematics has replaced Christianity as the intellectual tool used by the weak (nerds) to dominate their natural superiors (men of action and vigor like Jenkins). The soul-breaking catechism has been replaced by the binomial theorem. The priests are statisticians and bureaucrats, obsessed with counting and what can be measured. I am reminded of a remark by CS Lewis (I can’t find the exact quote now), that soft virtues like love and mercy had come to be more discussed than rigid virtues like chastity and courage, because it is easier to persuade yourself that you have been loving than that you have been chaste or courageous.

Politicians debate statisticians and philosophers

I should have known the writing was on the wall for my career in Canada when, at the first federal election debate in 2006, the Bloc Québécois leader Gilles Duceppe said

We don’t need inspectors. We don’t need statisticians. We need doctors and nurses.

The rest of academia kept their heads down, hoping the storm would blow over. But now, not even a decade later, just south of the border, presidential candidates have another academic discipline in their sights. In yesterday’s Republican presidential debate Marco Rubio said

Welders make more money than philosophers. We need more welders and less philosophers.

As is pointed out here, the first statement isn’t actually true. Whether it should be true is another question. We might say, a philosophical question; although, in a serious dispute over the issue between a philosopher and a welder, I would not be surprised if the latter came out the better for it.

First they came for the statisticians…

Fat and spinach

There’s an old joke — I’ve seen it attributed to Clarence Darrow, but I have no confidence in this attribution — that goes

I don’t like spinach, and I’m glad I don’t, because if I liked it I’d eat it.

I thought of this in reflecting on the lessons of Nina Teicholz’s book Big Fat Surprise, about the sorry history of public health recommendations about dietary fat, mainly in the US. This will surely go down as one of the most embarrassing disasters in public health history, so Teicholz’s efforts to uncover how a supposedly self-correcting process was able to go so badly wrong holds important lessons for all of us who care about either science or public policy. (It’s sort of The Innocence Project, with observational studies in place of eyewitness misidentification.) Continue reading “Fat and spinach”

Annie Get Your Prior

I was reading Sharon McGrayne’s wonderful popular (no, really!) book on the history of Bayesian statistics. At one point it is mentioned that George Box wrote a song for a departmental Christmas party

There’s no theorem like Bayes’ Theorem
Like no theorem I know…

A bit later we read of Howard Raiffa and Robert Schlaifer singing

Anything frequentists can do, Bayesians do better

(More or less… the exact text is not reproduced.) So it seems the underappreciated role of Irving Berlin in the development of Bayesian thought has yet to be adumbrated. Perhaps researchers will some day uncover such hits manqués as “How High is the Bayes Factor?”, “I’m Dreaming of a Conjugate Prior”, or even “Bayes Bless America”.

Qualified majorities

The new Conservative government has announced plans to make strikes by public-sector unions more difficult, including

A strike affecting essential public services will need the backing of 40% of eligible union members under government plans.

Currently, a strike is valid if backed by a majority of those balloted.

There will also need to be a minimum 50% turnout in strike ballots.

The new Business Secretary Sajid Javid said “What people are fed up of is strike action that hasn’t been properly supported by the members of the relevant union.” It is notable that this complaint comes from a government that received less than 37% of the votes in the last election, accounting for less than 25% of eligible voters.

Every three minutes

From a Guardian article on a new theory about the aetiology of Alzheimer disease:

It is thought that this year one person every three minutes will develop dementia.

It’s hard to explain statistics in a way that they feel real to people, but is “one person every three minutes” really a useful way to think about a disease that develops gradually over many years, as opposed to, say, muggings? Perhaps they meant to say “one person every three minutes will be diagnosed with dementia”.

Prenatal sex ratio

A paper that I’ve been involved with for a dozen years already has finally been published. We bring together multiple data sets to show that the primary sex ratio — the ratio of boys to girls conceived — is 1, or very close to 1. Consequently, the fact that more boys than girls are born — the ratio is about 1.06 pretty universally, except where selective abortion is involved — implies that there must be a period in the first trimester when female embryos are more likely to miscarry than male.

This is one of those things that is unsurprising if you’re not an expert. The experts had developed something close to a consensus, based on very little evidence, that the sex ratio at conception was much higher, some saying it’s has high as 2 (so that 2/3 of the conceptuses would be male), with excess female mortality throughout gestation. (We know that male mortality is higher in the second half of pregnancy, and after that… forever.)

The paper has its problems, but I think it’s a useful contribution. It’s also the first time I’ve been involved in research that is of any interest to the general public. Several publications have expressed interest, and an article has already appeared in two German magazines online, including the general news magazine Der Spiegel.

Update: Guardian too. This makes it interesting, in retrospect, that we had such a hard time getting a journal even to be willing to review it. One said it was too specialised.