Is there a way to tell SPSS to use summary data for a factorial ANOVA? Given the group Ns, means,and std devs only? Have seen for one-way (using MATRIX DATA command syntax), but not yet for factorial. Thanks.
Answering my own question... yes! The trick is to reverse engineer the raw data. The Box-Muller method can do this - it will generate pseudo-random, normally distributed data (a parametric assumption) by drawing random scores from a normal curve described by your mean, standard deviation, and N. You can use Excel's INVNORM function to do this, but I wrote a C++ program (based on existing code) to automate this for factorial ANOVA. Please let me know if interested in the program -- Open Science and all that! Does this method work? Yes. It's also helpful in analyzing guesses not in your summary stats...
https://en.wikipedia.org/wiki/Box%E2%80%93Muller_transform
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA