How unisex is your baby’s name?
One of my children has a unisex name. As we wander around local playgrounds and kids’ clubs we come across a fine mixture of girls and boys who share her name. A friend told me that she has three other mates who have given their kids this name — two girls and two boys in total.
Recently, I started to wonder: had we in fact given our child the UK’s most unisex baby name? I decided to have a look at the government data to find out.
Baby name data is published separately for England & Wales, Scotland and Northern Ireland. So the first job was to laboriously merge the data sets. A couple of notes: names are recorded by exact spelling, so for example Isabelle and Isobel are recorded separately. Also for privacy reasons, only names given to at least three individuals were used. I used the latest data which is fully available, for 2017.
I compiled a UK dataset for girls, and one for boys, and searched for matches. Bingo! In 2017, UK babies were given 555 distinct unisex names — those given to at least three girls and at least three boys.
OK, so… that’s a lot of names. But which is the MOST unisex? My first thought was just to calculate which of the 555 names was given to most babies overall. And on this measure, the UK’s most unisex baby name is… Noah.
Hmmm. I’m not sure Noah really qualifies as a unisex name. Although six parents did call their baby girls Noah, this pales into insignificance when compared to 4,780 boys.
So what about looking at the names which were used in most similar numbers for boys and girls? It transpires that 26 names were used for exactly the same number of girls and boys. Here are the most popular:
Aaaaand again, I’m not happy! How can the most unisex baby name be one that was given to only 22 children across the nation? You could easily go through life and never meet a Darby or a Lamar.
At this point it became clear to me that only a combination of frequency and ratio could determine which name was the UK’s most unisex baby name. We want to maximise the number of individuals given a name while preferring those where the ratio is closest to 1. I created an index by dividing the combined count of name frequency by the ratio of boy:girl or girl:boy, whichever was greater than 1. I then ranked names by greatest index.
WE HAVE A WINNER! The UK’s most unisex baby name in 2017 was Frankie, with an index of over 600, followed by Quinn and Harley at more than 400 apiece. Here’s the top 10:
This list is much more satisfactory. The names on it are reasonably popular and reasonably equally distributed. From this I conclude that this index is a good measure of a name’s unisexity. (Yes, I made that word up.)
But there is one final problem. The index as thus constructed is not comparable across populations. This is because it uses the absolute number of instances of a name, meaning that more populous countries will inevitably see higher index numbers. So the final step was to scale the index by the total number of babies born in the year in question. This resulted in very small numbers, so I then multiplied the total by one million to make the results more manageable.
Modestly, I have named this the Canovan Index of Unisexity™. Just to recap, it is defined thus:
We can now give a table of the UK’s most unisex names which can be compared with the results of other datasets. As we have just divided all the indices by the same amount, the names are in the same order:
In summary, this is a sensible definition of how unisex a name is. It takes into account how common it is amongst its own population and the spread between girls and boys. It is comparable across datasets from different countries, and also across time from the same country. Watch this space for more analysis!
And my own child’s unisex name? Well, it’s not Frankie. In fact, it’s not even in the top 10. But I’ve just given birth to a lovely little index, so I’m happy all the same.
