Sinister Statistics: Do Left Handed People Die Young?

In 1991 Halpern and Coren published a famous study in the New England Journal of Medicine which appears to show that left handed people die at much younger ages than right-handed people. Halpern and Coren had obtained records on 987 deaths in Southern California–we can stipulate that this was a random sample of deaths in that time period–and had then asked family members whether the deceased was right or left-handed. What they found was stunning, left handers in their sample had died at an average age of 66 compared to 75 for right handers. If true, left handedness would be on the same order of deadliness as a lifetime of smoking. Halpern and Coren argued that this was due mostly to unnatural deaths such as industrial and driving accidents caused by left-handers living in a right-handed world. The study was widely reported at the time and continues to be regularly cited in popular accounts of left handedness (e.g. Buzzfeed, Cracked).

What is less well known is that the conclusions of the Halpern-Coren study are almost certainly wrong, left-handedness is not a major cause of death. Rather than dramatically lower life expectancy, a more plausible explanation of the HC findings is a subtle and interesting statistical artifact. The problem was pointed out as early as the letters to the editor in the next issue of the NEJM (see Strang letter) and was also recently pointed out in an article by Hannah Barnes in the BBC News (kudos to the BBC!) but is much less well known.

Percentage of left-handed peopleThe statistical issue is that at a given moment in time a random sample of deaths is not necessarily a random sample of people. I will explain.

Over the 20th century, left handers have increased as a fraction of the population. Left handedness may be relatively fixed as a genetic matter but in the earlier decades of the 20th century children were strongly discouraged from exhibiting left-handedness. As a result, many “natural” lefties learned right-handed behavior and identified as right-handed adults. Over time, however, the cultural suppression of left-handedness declined and the proportion of adults exhibiting left-handedness increased, as the figure, at left, illustrates (fyi, I believe British data).

Now suppose you take a random sample of people who died in 1990. In this sample, some people will have died old and some young. Among those those who died old, however, fewer people will be identified as left-handed because the old grew up in a time when left-handedness was suppressed. As a result, the old deaths in your sample will tend to be have more right-handed people and the young deaths will tend to have more left-handed people causing you to incorrectly conclude that left-handed people die younger. Studies show that this statistical artifact can easily explain a 9 year difference in apparent mortality rates.

To make this crystal clear consider the following thought experiment (offered by Chris McManus). Imagine you take a sample of people who died recently and asked their surviving family members, Did the deceased ever read the Harry Potter novels? One would clearly find in such a sample that those who died tragically young (at age 12 let’s say) would have been much more likely to have read Harry Potter than those who died in their 90s. Despite what some might argue, however, we should not conclude that Harry Potter kills.

Hat tip: Tim Harford.


Far too wonkish for the "women earn 78 cents on the dollar as men for doing the EXACT SAME WORK" crowd.

Comments for this post are closed

Well, an interesting point - unless you feel that essentially everyone left handed is very well aware of being left handed, and that their families, at least, would also knew, especially as the social strictures concerning left handedness faded away.

The point about left handers being forced to use their right hand in a typical fashion for eating and reading two generations ago is completely true - my mother was raised that way. The only way to know she was left handed in such a context was to ask her - in which case, she would tell you.

Or to have been a member of the family, as my mother never made any secret of being left handed, regardless of how she ate or wrote.

This seems like a cultural artifact searching for a statistical grounding.

The Harry Potter quote is good as an illustration - though far too restricted. Ask someone if they have read a book about magic, and the results might still skew towards the younger set due to the ubiquity of Harry Potter. Thankfully, you didn't pick alcohol as an example to make a valid point - that would have been far too easy to shoot full of holes.

So, in the name of anti-MR contrarianism, you believe that the life expectancy gap is real. At least you're consistent.


Comments for this post are closed

Nope - I am disputing the idea that family members of people who are left handed would be competely unaware of that fact, to the extent that it would statistically skew a study in more than a minor, though understandable, fashion.

I did not bother to cite the additional fact that everyone in my family was equally aware that my youngest brother was also left handed, though never forced to do anything with his right hand, in large part because of my mother's own experience, and not merely due to a changing zeitgeist.

The reality that people were forced to conform to the expectaction of using their right hand in social contexts two generations ago does not necessarily mean that left handers actually hid from other family members the reality of that fact.

After all, both my grandmother and grandfather, the ones who forced my mother how to learn to eat and write with her left hand, were also fully cognizant about my mother's left handed nature (she was an only child), and it was not a secret from the grandchildren on their part.

I neither agree nor disagree with more than 2 decades old research (it seemed a bit dumb when it first came out, to be honest, and not really worth investigating when reported inthe media) - I am merely disputing, based on personal experience, that family members would be completely unaware when someone else in the immediate family was left handed.

It is an unsupported leap to base on an argument on the assumption that family members would be so ignorant of each other, based on my own experience. How this becomes 'anti-MR contrarianism' as compared to the actual reality of living in a family with multiple left handed members escapes me.

Though it isn't exactly hard to use factual information to be an 'anti-MR contrarian,' after all.

Comments for this post are closed

Comments for this post are closed

Yes, some people still identified as being left-handed despite being made to switch or be closeted. Surely the graph and study cited provides evidence that there still were fewer people who identified as being left-handed, though.

The discussion from Prof. Tabarrok was that family members identifying those who had died as left handed were likely to be inaccurate due to previous social training of left handed people to use their right hands in certain social situations.

Based on personal experience, it isn't as if family members woudn't notice that apart from several rigorously trained actions such as eating and writing, that another family member was right or left handed.

And at the time of the survey, the stigma of being 'sinister' had faded over at least 3 decades - in other words, family members would be more likely to accurately report left handedness than previously.

Of course, anyone who has lived with a left handed parent and/or sibling is welcome to explain how that parent/sibling was able to successfully conceal such a basic aspect of how they did things in daily life.

However, I realize that this a site where most commenters are apparently unlikely to notice the absurdity of the 'benefit' of a cell phone app telling them when to kiss someone. Or to feel that having a vanguard of marketers leading them to a bold new future where they are promised the dream of feeling good about themselves is not exactly all that different from the last decades of American history.

And yet the actual data disagrees with you. More people identify as left-handed to themselves and their family than did previously, despite natural inclinations.

And for what it matters, I am left handed.

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

An excellent illustration of why not to read Harry Potter!

I love this kind of thing - we live in the age of big data, which means logical fallacies involving statistics are more common than ever.

Comments for this post are closed

In their reply to Strang, Halpern and Coren claim that there's been no historical trend toward more frequent left-handedness:

In 1932, Pyle and Drouin2 tested 13,435 elementary school students (who would have been about 70 years old at the time of our study) and found that 6.4 percent were left-handed. In 1953, Karpinos and Grossman3 tested 12,159 U.S. army registrants (who would be in their mid-50s now) and found that 8.6 percent were left-handed. In 1987, Chapman and Chapman4 tested 5825 college students (who would now be in their mid-20s) and found that 6.6 percent were left-handed. Thus, for more than 50 years of recent history, there has been a negligible change in the percentage of left-handers. This lack of a historical trend was confirmed by a statistical analysis of 34 studies of handedness conducted between 1913 and 1976, using only white subjects and preference measures for handedness.5 The results showed a slight trend toward increased left-handedness in the more recent studies, but it clearly failed to reach statistical significance. Furthermore, analysis showed that even had the trend been statistically significant, it would only have predicted a decline in the population of left-handers with age, which would account for only one quarter of the change actually observed in the life-span data.

Interesting claim, but what about this discussion of the Pyle and Drouin study:

"She supported this view with data from Pyle and Drouin (1932) showing that 7.5% of the children in the Detroit school system in 1932 were left-handed writers compared with 2.2% of the children in other school systems. The Detroit school system was notable at the time for the absence of pressure towards dextrality exhibited by teachers."

That to me completely destroys their claim that Pyle and Drouin (1932) support their conclusion. In fact, it supports the opposite, that social pressure greatly affects how many people who may tend towards left-handedness identified to themselves and family as sinister.

Thanks. Halpern and Coren's defense did sound implausible to me.

Comments for this post are closed

Comments for this post are closed

The 1932 Pyle and Drouin study apparently actually found that 7.5% of children in the Detroit school system were left-handed writers, but only 2.2% of children in other school systems, and noted that the Detroit school system was unusual at the time for not encouraging students to switch. Therefore, far from supporting Halpen and Coren, it undercuts their argument.

Comments for this post are closed

Please tell me they used an equal number of males and females in the army number

Comments for this post are closed

Comments for this post are closed

The graph also suggests that if you don't control for gender you will partially attribute the higher mortality rates of men to lefties. But that won't be enough for explaining a nine year gap, I believe.

Comments for this post are closed

Excellent. A similar example: how long does the average unemployment spell last? A bad way to estimate that number is to look at the currently unemployed people and see how long they've been unemployed, similar to how you don't want to estimate life expectancies by looking at the people who died in a given year.

The existing pool -- or any snapshot -- of unemployed people will be overweighted with the long-term unemployed; the short-term unemployed leave that status quickly, and will be perpetually under-sampled unless one takes care to look at a longer-term sample which follows unemployment spells longitudinally.

Larry Summers some decades ago wrote a paper which pointed out the different results that you get depending on how you estimate the length of unemployment spells. For some reason he had trouble explaining it to his macro students -- or maybe they were just being dense: "the length of the average unemployment spell among those currently unemployed" vs "the length of the average unemployment spell among those who ever became unemployed".

Comments for this post are closed

Classic truncation it!

Comments for this post are closed

The only thing deadlier than left handedness is Louis XIV: Everyone that ever met Louis XIV ended up dying.

Comments for this post are closed

Comments for this post are closed