Peera, In general, the fewer transformations you do, the less additional noise you add, and the better you are able to detect differences between groups. If you do a subtraction between conditions, the variance of the difference is the sum of the two variances (or sd =sqrt(sd1^2 + sd2^2)). This isn't too bad, if you consider it is identical to pythogoras theorem--the standard deviation (and standard error) will be equivalent to the hypotenuse of a right triangle with legs equal to the individual standard deviations. Most of the transformations I have seen work that way, and I have typically used raw difference scores for comparison.
If you also scale using division and products of the components, this can have larger impacts on the variance and make the resulting scores less reliable. However, they may make them more interpretable, because 1.0 now means something now. In your formula below, do W and C below refer to error rate, accuracy rate, or response time? I think they probably The value of this depends on the scale of the two values and so I wouldn't use this unless you know what the inputs are supposed to be scaled with. It might only work if these values are prescaled to be between 0 and 1 or something. I'd suggest just subtracting mean congruent from incongruent RTs for each condition to get an interference score in MS for each person. Shane > > > Firstly, I would like to say thank you for your excellent PEBL tasks. I am > currently using Stroop task in my project and it works very well. > > However, I am wondering that could you please suggest on how to calculate > Stroop interference score from the given output? I am trying to convert to > fit this formula, but it seems the outputs do not allow me to do so. > > > "(W*C)/(W+C) = CW'; CW - CW' = Stroop interference score. W = raw > word score; C = raw color score; CW = raw color-word score. > > Your suggestions are very much appreciated. > > Best regards, > Peera > > > > ------------------------------------------------------------------------------------------------------------- > > Mr. Peera Wongupparaj, BA, MA > > PhD (Candidate) student > > Institute of Psychiatry > > Department of Psychology, > King's College London > ********************************************* > > Box PO78, ASB building (Fourth Floor), > > De Crespigny Park, Denmark Hill, > > London SE5 8AF > > Tel. 07564718478 > > Email: [email protected]<mailto:[email protected]>, > [email protected]<mailto:[email protected]>, or > [email protected]<mailto:[email protected]> > ------------------------------------------------------------------------------ > _______________________________________________ > Pebl-list mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/pebl-list > ------------------------------------------------------------------------------ _______________________________________________ Pebl-list mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/pebl-list
