Statistically informed ideas on how to make the world work better.

A proposal to substantially reduce school shootings

The school shooting problem in the US is a hard nut to crack. It could be largely solved with draconian restrictions on guns, but these are politically impossible to pass given the political divisions and the need to control the presidency, the House of Representatives, and 60% of the Senate in order to achieve anything. This won’t happen soon for Democrats.

Republicans counter that the right solution is to fortify schools. I don’t think they’re right, but here’s how Democrats can win: give the Republicans a chance to be right. Offer to enact all the policies that Republicans want, on the condition that, if they don’t work, then there is an automatic trigger for draconian gun control legislation. It’s win-win: if the Republican policies work, so much the better. And if they don’t, then we achieve far more gun control than we ever could otherwise. And politically, Republicans might be forced to accept: It’s pretty hard to counter the Democratic offer of, “If you think your policies will work, fine, we’ll give them a try. And if you’re not willing to try our solution if yours don’t work, that means you don’t care about our children’s lives.” The down side, of course, is that many children will be killed before gun control takes effect, but they are lives that will be lost regardless, due to the political gridlock.

Read the rest of this entry »

How to reduce police violence against minorities

Protests over George Floyd's death by a Minneapolis police officer ...

With the ongoing protests in the aftermath of the police killing of George Floyd, it’s worth thinking about how we can reduce police violence against minorties. Many people say that rioting doesn’t solve anything. It’s not yet clear if that’s true, but the rioting is at the very least a legitimate expression of anger about a situation that has not gotten markedly better, despite a lot of attention over the last 5 years. Perhaps the rioting will call sufficient attention to the problem that something might change. But the protesters agree: this is not one bad apple, this is systemic. So how should the system change? Without an answer to that question, the rioting will not solve the problem. With an answer, the current crisis might provide a push to major policy changes that could actually help.

Here is my answer: Punish the police collectively for the failures of some members of the force.

Read the rest of this entry »

The best reasons to favor diversity in hiring: A response to Tristan Kromer’s “Why Meritocracy Doesn’t Work”

Tristan Kromer just published a blog post titled “Why Meritocracy Doesn’t Work” in response to the recent anti-diversity Google manifesto. He makes some good points, notably around how small amounts of bias can amplify and threaten diversity without increasing the merit of employees. However, to my mind, he misses the most important reasons to have active diversity policies at the expense of a pure meritocracy. Here are the key points, developed in a bit more detail below:

  1. Diversity increases new ideas and decreases groupthink;
  2. There is no obvious reason to believe that hiring the best individuals necessarily produces the best workforce – there can be emergent properties based on how people interact, etc.;
  3. There is no fully objective metric of merit, and forcing people outside their comfort zones in defining merit may open up new ways of seeing things;
  4. A cultural change to ensure and value diversity could create new ideas of merit that are less individually based.
  5. There are advantages to both company performance and diversity. It seems unreasonable to expect one or the other to completely dominate the other, particularly as there are likely to be diminishing marginal returns at the extremes of which we value.

Read the rest of this entry »

An Open Letter to Canada’s Health Minister

July 8, 2016


Hon. Jane Philpott, 

Minister of Health

House of Commons

Ottawa, Ontario, Canada K1A 0A6

Dear Dr. Philpott,

I was very pleased to see that concerns of the medical research community are being expressed, and that you are convening a panel to explore next steps. There are many serious issues to be addressed. However, while I thank the writers of the original open letter for getting this process going, and while agreeing with the general sentiment of frustration expressed in that letter, I (and I suspect many others of members of the medical research community) do not agree with the specifics, and in particular with the suggestion to return to a system of face-to-face peer review. While I cannot claim to speak for others, I suspect that many signatories of the letter agreed with the frustration expressed more than the solution proposed.

Broadly, I would raise four points in response to the open letter. Read the rest of this entry »

Might societal acceptance of homosexuality drive it extinct?


The precise causes of homosexuality are not completely clear, but most scientists agree that there is a genetic component, perhaps 35% for gay men and about half that for lesbians. There is unlikely to be a single gene that makes people gay, but there may be some genetic variation that increases the probability of becoming gay, or is a necessary precondition but also requires a certain environment (e.g. intra-uterine). (I have written previously on why natural selection could produce homosexuality, counter-intuitive as that sounds.)

In any case, the simple fact of genetic variation associated with homosexuality means that natural selection can act on it, and this has a major and paradoxical implication. The more biological children of gay people, the more transmission of these genes, and the more gay people there are. The less biological children of gay people, the fewer gay people there will be in subsequent generations. But in what societies do gay people have the most biological children? Almost certainly those societies that forbid homosexuality, forcing gays to marry and have children. Modern societies with liberal social policies will likely see many gay couples adopting, but fewer having biological children. Over time, this means that societal acceptance of homosexuality might actually drive it extinct, or nearly so. Unfortunately, natural selection in this case will function such that every society will achieve, long-term, the opposite of what it wants. Read the rest of this entry »

The probability that God exists: A Bayesian perspective

Many people are very certain either that God exists or that (s)he does not. This post is based on the premise that, from an empirical perspective, such certainty either way is almost certainly unwarranted, but that a combination of statistical theory and the current state of scientific knowledge can shed substantial light on the probability. After all, there is a rather big difference between an agnostic who thinks there is a 1% chance God exists and one who thinks there is a 99% chance. Obviously, there is no way to calculate a precise probability, but there certainly are ways to establish a likely range. In a nutshell, I will show that, while all the scientific advances in the world cannot disprove God’s existence, the accumulating capacity to explain the world through science substantially diminishes the probability that God exists. Read the rest of this entry »

In support of women who speak out about sexual assault: a man’s perspective, and a call to arms


Image credits: Brendan Smialowski/Getty Images for Meet The Press and Darren Calabrese/Canadian Press

Like so many people, I have been upset by the stream of recent news stories on sexual assault: Bill Cosby, Jian Ghomeshi (a popular and seemingly pro-feminist radio host here in Canada), the University of Virginia story from Rolling Stone, several stories relating to violence against aboriginal women in Canada, and stories about the US military, to name just a few. Throughout this string of stories, I have felt admiration and support for the women who have come forward, but I have also wondered why not everyone comes forward, and why so many women waited so long to do so.

In the ensuing debate, many people (mostly women) have decried a culture that permits sexual assault, and a few people (almost exclusively misogynistic men) have defended the status quo and denigrated women, often through trolling and other unsavoury means. But there is a perspective that has been largely missing: that of the average man. Most men are not misogynists, and I would venture to say that most are actively against misogyny and do generally respect women. But the misogynists are loud, obnoxious, and everywhere on the internet. Most men shut up when the subject shifts to sexual assault, partly because it’s seen as a women’s issue, and partly because they are scared of saying the wrong thing and coming off as misogynists themselves, even when they are not.

But sexual assault is not a women’s issue, it’s a human issue. White people can see the injustice in how blacks are treated by the police, and adults can be against pedophilia. When something unjust happens to one person in our society, we should all be concerned. It is time for the large majority of men to speak out, and to support women. My hope in writing this piece is that it will be read by men, forwarded and posted by men, and discussed by men.

Read the rest of this entry »

Idea: How to stop sleazy media companies from screwing us on our bills

One of my co-workers has been complaining a lot recently that Bell, the internet/TV provider, has been intentionally messing up her bill for 3 months in a row. Everyone I know has at least one story like this, of being really screwed by some phone or TV company. It’s not that she’s being targeted, it’s that companies like this know they can wear consumers down if they make enough “errors.” Eventually, we don’t have time to fight, and if we don’t pay up, our credit rating suffers. So we give in. The company doesn’t even have to try to screw us individually – they can just build an imperfect system that tends to make errors in their favor.

Well, I have a permanent solution to this problem. The only leverage these companies have over us is our credit rating. But many people who have these problems have perfectly fine credit – like me, or my co-worker. If we chose not to pay our bills it wouldn’t reflect a risk of defaulting on a mortgage, our car payments, or a credit card. It would just reflect the fact that the phone company is awful and gave usa faulty bill.

Credit scores are generated by companies that have developed algorithms to predict the chances that someone will default on a payment. There is competition among consumer credit rating agencies to have the best prediction. Banks and used car salesmen need to know this risk so they can offer appropriate credit. So anyone who comes up with a better prediction algorithm could corner the market for credit scores.

Enter my idea: Non-payment of a cell phone bill of $500 – even though your normal bill is $45 and you regularly pay every other bill – is NOT a sign of any risk that you will default on anything else. It is a sign that the phone company screwed you. So, a more accurate credit score would not penalize people for not paying such bills. A credit score that took this into account could outperform other credit scores and become the industry standard. And at that point, the evil media companies would have no leverage to force us to pay, except to go to court. Going to court is expensive and difficult, and they would usually be at risk of losing. So they would be forced to stop screwing us if they want to be paid!

Note that consumers who refused to pay all their bills would still have worse credit scores. It’s just specific non-payment of one bill that would contribute nothing (or next-to-nothing) to the score. Obviously, a good algorithm would have a sliding scale (a continuous measure) for credit score decrement based on how aberrant the bill was, how often you didn’t pay (relative to other people), and so forth. So the system couldn’t really be abused by consumers in the other direction either. The effect would just be to shift the balance of power in a dispute to midway between the consumer and the company. The company can refuse service, the consumer can refuse payment, but there’s no strong external force limiting the consumer.

Unfortunately, this idea takes an entrepreneur to implement, someone to develop the new credit score and start up a company to sell it (or integrate it into existing companies’ algorithms). While I’m confident I could improve on the existing algorithms, I have other statistical questions that occupy my time, and I’m no businessman. So someone else will have to do this…

A strong, immediate impact of the demographic transition on human evolution?

This blog post is different than most others. I had a Eureka moment in the shower, and I just need to write the idea down so I don’t forget it, and to share it with potential collaborators so we can maybe develop it into a paper if it hasn’t been done already by someone else (a quick search in Google Scholar suggests no).

Over the last 150 years, many human societies have undergone what is known as the demographic transition, first dramatically reducing mortality rates and subsequently reducing fertility rates. In simple terms, your great-great-grandma had 15 kids and 3 lived. You will have one or two, and they will live. What is the impact of this change on human evolution? I think it could be profound, and here is why:

Read the rest of this entry »

XKCD on small risks

As always, xkcd comes through at the right moment. In support of my last post, here is why we shouldn’t worry too much about taking a baby out of a car seat for a few seconds while driving, or other miniscule risks:


Increased Risk