About the theory, I should say that at E=20eV, there are nowadays three calculations performed HRM-SOW, TDCC, and ECS. At the (e,2e) symposium F. Martin showed that for equal energy sharing, they all give the same results within a few % deviation (or less) to each other. Furthemore, the
Brauning et al data fits nicely to the calculations. So I would say at this stage, so far so good for the theory.... I forgot to mention that I would not suspect that there is a dramatic change at E=25 eV....
Today, I would rather think that we (at least I) do not control/understand the calibration procedure so well. But I am sure that we will fix that rather soon, definitely before any publication !! In a first step, I would normalize the data to the theory (Patricia's one for instance) to get the normalization factor at each theta1 (as I did in the file *_rel.opj), and then apply it to Hydrogen taking into account the ratio between hydrogen and helium cross section and the number of events.
I hope that you will read this message.