It is a natural human instinct to look for patterns. While entering the birth and death dates for all the people involved in Scott’s and Shackleton’s expeditions into a database, I couldn’t help noticing familiar dates. I spotted two people with the same birthday as me (Robert Selbie Clark and Henry McNish), as well as men born on my wedding anniversary (Huberht Taylor Hudson), my mother-in-law’s birthday (James Murray) … and even one prescient soul (Thomas Taylor) who was born on Antarctica Day some 130 years before the day was even inaugurated!
However, I also noticed what seemed to be a disproportionate number of birthdays on Christmas day. Was this a real trend? How could I tell if it was? Are Polar explorers really more likely to be born on Christmas day? Or rather, are you more likely to become a Polar explorer if you have a Christmas birthday?
Of the 230 different people in our database, we have the full date of birth for 105 of them (46%). Four were born on Christmas day: Arthur Samuel Bailey (born in 1878, took part in the Terra Nova expedition), Walter Ernest How (1885, Endurance) and William Lashly and William Lofthouse Heald (1867 and 1875, both Discovery and Terra Nova). That’s not even counting the two near-misses who were born on Boxing day: Frank Debenham (1883, Terra Nova) and Leslie Thompson (1886, Aurora). In fact, 16 people were born in December – a lot more than in any other month. The graph below – which uses polar co-ordinates, appropriately! – shows this strikingly:
Bailey, How, Lashly and Heald make up 3.8% of the 105 known birth dates – about 14 times bigger than the 0.27% probability that somebody will be born on any given day (1/365.25 = 0.0027). So you might think this is proof that polar explorers are more likely to be born on Christmas day – and even that there is some link between these two circumstances. Surely this is more than coincidence?
It is not actually that remarkable to find two people who were born on the same day, even among quite a small group. A famous maths problem asks how many people you would need at a party for there to be a 50% chance that at least two of them share a birthday. The answer, somewhat counter-intuitively, is 23 – that is, if you have a group of 23 people, there is an even chance that at least two of them will be born on the same day. By the time you have 50 people at your party, there is a 97% chance that there will be at least one shared birthday. (To put it another way, there is only a 3% chance that all 50 people will have a unique birthday.) Given our group of 105 people, there is a probability of 99.9999% that some of them will have the same birthday – it would be really remarkable if none of them did.
This is not quite the same as our problem, however. The birthday problem assumes that it doesn’t matter which day is shared, which hugely increases the chance that you’ll find two people with a birthday in common. To find out how likely it is that four Polar explorers will have birthdays on Christmas day, you need to use the binomial distribution. This tells us that the probability of at least 4 people out of 105 being born on Christmas day is 0.02154% – very unlikely indeed.
Before I get too excited about this, a word of warning: this is a relatively small sample, and the date of birth is only known for 46% of our 230 Polar explorers. If none of the others were born on Christmas day (quite possible given that you’re more likely to record your birthday if it is on an “important” date), then the probability of at least 4 Christmas-born explorers goes up to 0.39% – nearly 20 times greater (but still quite unlikely). I have also assumed in my calculations that there is an equal chance of being born on any given day. In fact, birth frequencies fluctuate slightly throughout the year, with the most common birth month in Europe being July. We can’t assume that this was also true in nineteenth-century Britain – or elsewhere, given that some of the men on Scott’s and Shackleton’s expeditions were born in other countries. Nevertheless, it looks as if the high number of Christmas birthdays is more than just coincidence.
Why is this? It is already well known that your month of birth can affect everything from academic and sporting achievement to health or choice of career. Maybe having a significant birthday (like Christmas day) makes you more likely to stand out among your peers, and to follow an adventurous career? Or perhaps babies born in the depths of winter have an affinity for bleak, icy places?! Whatever the reason, it seems that Christmas babies really are more likely to become Polar explorers.