By LIN - 12/25/2016
I am confused with how to calculate Cronbach's alpha of IAT ,
and I think that the calculation of Cronbach's alpha of IAT and ST-IAT will be the same ,right?
It will be kind of you if you can tell me how to resolve this problem.
Thank you very much!
By Dave - 12/26/2016
If you want to calculate Cronbach's alpha, you essentially
- compute difference scores for every single trial of the combined blocks,
- treat those as separate items, and
- calculate alpha over that.
I'll note that the more common (and easier) approach to assess IAT reliability is some variation of split-half reliability, e.g. calculation the correlation between the D-score for the 1st combined blocks (da) and the D-score for the 2nd combined blocks (db). https://faculty.washington.edu/agg/pdf/Schnabel%20&%20al.Hbk_2008.proofs.pdf gives a short overview of the various approaches.
By LIN - 12/26/2016
Thank you for your response,
it's really a great help to me!