Millisecond Forums

Hello,

We have begun running our experiment, but D scores are not calculating as we'd expect. We hand-calculated some and the results are not coming out the same.
I'd appreciate it if you could let me know if this is a problem in the programming, or a problem in my manual calculations? 

These are the steps I'm going through manually (after removing the necessary trials):
1. Compute the standard deviation for each participant’s “practice” trials
2. Compute the standard deviation for each participant’s “real” trials
3. Compute the mean difference between practice blocks and divide for each participant and by each individual’s pooled SD
Practice Difference Score = (Mean (Practice incompatible) – Mean (Practice compatible)) / Mean SD for Practice Trials (Step 1)
4. Compute the mean difference between the real IAT-blocks and divide by pooled SD
Real Difference Score = (Mean (Real incompatible) – Mean (Real compatible)) / Mean SD for Real Trials (Step 2)
5. Create D score by averaging the practice difference score (step 3) and the real difference score (step 4)

I've attached my program here so you can take a look. All the calculations look the same to me and all my lab members. I also checked that the right values/expressions were in each block and it all looks good to me, so we're really stumped as to why we're getting such different results. Any help is greatly appreciated.

Thank you so much!

Cassandra