I would like to use the allocation routine to assign new subjects to group using the Kullback-Leibler divergence (KLD) index. My first cohort of 10 participants were assigned to the experimental and control groups at random. I would like to assign each subsequent participant using the XL toolbox allocation routine. However, it assigns all of the participants (i.e., the 10 who were randomly assigned and have already completed the protocol and the new participants). How can I modify the procedure so that it maintains the group assignments given to prior participants and only assigns the new participants using the KLD?
it is possible right now, although with some limitations. You need to disable the ‘presort’ as well as the ‘optimize’ and ‘use chance’ options. With these three disabled, you will get the same allocations for your existing subjects, unless you change the order in which they appear on the sheet (which, btw, is what the ‘presort’ option does internally).
Thank you. I tried to follow your suggestions, but perhaps I did it incorrectly. I have uploaded pictures of the the spreadsheet before and after the allocation and with a picture of the Allocation options I selected. I would greatly appreciate it if you could take a look at the pictures and let me know where I went wrong.
hm, it works quite well over here. I performed four runs; each time, I gave a different range for “Allocation criteria” that was one row larger than in the previous run. Here is the output: