The IAAF's research study into performance and testosterone (as affects Caster Semenya) is affected by significant data errors and problems, and it has to be retracted. This is the result of a three month back-and-forth, where a group of us requested data from the IAAF, and found significant problems.

Obviously, suggesting a retraction is a big step, but when we compared the published data to what was eventually sent to us, we found between 17% and 33% of the data was flawed... or erroneous. This could have huge implications for the conclusions of the paper, though nobody can say whether they'll strengthen or weaken the findings. Regardless, that's enough to say that the paper must be retracted, and to call for more transparency in the future.