Biased Psych Reseearch

I'm trying not to post, but this is bugging me.

Testing the ‘Extreme Female Brain’ Theory of Psychosis in Adults with Autism Spectrum Disorder with or without Co-Morbid Psychosis http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4466563/

They found lower empathizing bias (EB) scores for those with higher FSIQ (a measure of IQ), but moreso for the ASD-NP group than the ASD-P group (Figure 2). However their downloadable spreadsheet has five ASD-P entries with FSIQs of exactly 500 or 600. It also shows empathizers (Type E) scattered about the spreadsheet, whereas Table 2 reports that none of them were Type E.

Perhaps the modulation by IQ wasn't quite significant, so they fibbed it. I suppose it wouldn't have had a huge effect since IQ was split into discrete categories anyway, but these fakes comprised 5 of the 13 in the ASD-P high IQ group. The fakes were mostly empathizers labeled "Balanced". Correctingly, there would have been 18 in total if they had been included. Alas their exclusion from the analysis could have been the intent.

I'm just confused because their spreadsheet doesn't seem consistent with their findings at all.

Edited December 2, 201510 yr by MonDie