In attachment a script for the data in the "fried.csv" file + a data matrix below in the script. It is based on your first https. Hope it might help.

(Make sure to place the csv file in the write directory; you might want to rename it.)

The aligned ranks version of the Friedman test (friedmanAlignedRanksTest) has the most power.

Catharine,

There's a fairly clear example of Friedman test and post-hoc tests at the following link. There are a couple of options for post-hoc tests: pairwise sign test, and Conover test. Each is easy to conduct.

http://rcompanion.org/handbook/F_10.html

You might also look into Quade test, which is used in similar situations, but is sometimes preferable:

http://rcompanion.org/handbook/F_11.html

Do you all know how I need to change the coding if I have more than 12 or 14 treatment factors across 373 blocks?

Just adapt your input file (e.g. csv file); for example:

c1,c2,c3,c4,c5,c6,c7,c8,c9,c10

500,400,98,200,250,406,407,408,409,410

etc.

or the matrix:

y

I think 14 treatments with 373 blocks should work without trouble.