Overestimate of ESDs from Pilatus detector

[Andrew]

I have synchrotron SAXS that was collected off a Pilatus detector. The data are very nice, but the Chi values for a scattering curves regularised using Gnom is typically ~0.5, suggesting that the ESD are on average overestimated by a factor of ~2. Similar data collected from a Bruker NanoStar using their VANTEC detector yields values ~0.9.

Is it something inherent in the detector (Maybe the Gain of the detector needs correcting for), or is it likely that the data reduction is doing something a little strange?

It isn't a huge issue, but a reviewer is questioning the quality of our data based on the error bars (which are overestimated). I don't like arbitrarily scaling ESD's as I don't know whether the effect is independent of q or intensity. I would like an explanation that I give in the responses.

[Andrew]

It turns out that the data reduction program was outputting 2 standard errors, and I was assuming it was 1 standard error.