I'm also curious as to how testosterone level variance is only 56% decided by X or Y.
That claim in the article is intellectually dishonest and completely misrepresents the findings of the cited article. It didn't even twist the interpretation of the findings: it made up shit and attributed it to the paper. Any PhD student of mine did that would be in for a serious talk about critical thinking and research ethics.
Fwiw, the line in the blog was:
And while testosterone exhibits the largest difference between adult males and females, heritability studies have found that genetics (X vs. Y) only explains about 56 percent of an individual’s testosterone, suggesting many other influences on hormones.
The actual study cited for this "fact" was a twin study looking at how individual differences in testosterone were biologically heritable. In effect, whether being in the high versus low end of the normal male range of testosterone runs in the family. Twins are useful to study these things because they share a common environment, but monozygotic twins share 100% genetic profile, whereas dizygotic twins share ~50% genetic profile. The study found that both genetic and environmental influences affect testosterone levels, and 56% of the variance in testosterone can be attributed to genetics alone.
But it's still not about X and Y. The whole study used normal population male (XY) twins but didn't sequence their genomes at any point. The authors think that what really drives the heritability of hormone levels like testosterone are genes for receptors and feedback mechanisms in a complex hypothalamic–pituitary–testicular hormonal regulatory system... and genes for coding these things have already been identified all over the chromosome. Not just on X or Y.
Sheesh.