A link between the size of a father’s testicles and how active he is in bringing up his children has been suggested by scientists.
Researchers at Emory University, US, said those with smaller testicles were more likely to be involved with nappy changing, feeding and bath time.
They also found differences in brain scans of fathers looking at images of their child, linked to testicle size.
This is all over the news but I am always leery until I examine the paper. So very often what the press release says is not supported by the paper. There may all sorts of other confoudning things.
That appears to be the case here.
The mantra – correlation is not causation. Even if these (testosterone and caregiving) are correlated, something else, like sleep deprivation, might be the ’cause.’
For example, lack of sleep reduces testosterone levels. Tremendously. And men getting less sleep have smaller testicles on average. Might the correlation be between sleep and caregiving? Men who are doing more caregiving are getting less sleep. That explains both the lower testosterone, testis size and several other things.
But saying tired fathers are better caregivers is not as titillating as comparing the size of their balls.
So I looked at the paper itself. Sleep is not mentioned at all. That is a worry as it would be an obvious thing to examine.
They looked at 70 men and measured testicular volume with MRI. First some buried numbers here.
The mean volume for all the men’s testicles was 38,064 mm2 ± 11,183. But 43 out of 70 has testicles smaller than average. This means the median (the point where half are above and half are below) is smaller than the average, indicating that their population of men was not ‘normal‘. In a normal distribution, the mean and the median should be the same. Here, it appears they had a small number of large testicles that pulled up the average.
This could just be a function of the small study they did. With a larger study, one might expect the mean and median to be closer. And this might affect the overall numbers.
Because the numbers themselves are not all that strong. For example, the effect of testes size on caregiving is just barely statistically significant (p<0.05). This means that there is a 5% chance the data just happen to be correlated when they are not.
But the correlation is not outstanding. Here the correlation is -0.29.
Sure there may be some correlation but not a lot. A correlation of -1 would absolutely predict that a man with large testicles would be a poorer caregiver than a man with smaller balls. A correlation of 0 means you could not predict based on testicle size.
But -0.29? This means that predicting whether a man will be a bad father purely by the size of his balls is not very likely. The data simply do not support the ability to tell whether a man is a good or poor father simply by looking at the size of his balls.
In fact, the ‘worst’ father in the whole study has the same size balls as the best. And the guy with the biggest balls was just as good a father as the one with the smallest. Individual men were all over the map.
This study is just not predictive at all. Even if there was a small correlation between testicle size and fathering, it is not strong enough to determine what individual men will do based on the size of their balls.