No it hasnt - except by cable proponents who seek to undermine a rigourous scientific method that fails to corroborate their beliefs.

Please produce references for published work in the scientific community that shows DBT is flawed...

If you go for a table ranking kind of study say of 12 cables then you can get significant results even with a dozen listeners. Its all in how you build up the tests.

In edition students t can be used for n < 50 where the normal distribution approximation doesnt hold.