So we were allowed to leave early(ish it was only a quarter til four) yesterday because the air conditioning has once again stopped working and it was anywhere between 85 and 90 degrees on the floor. The part they need to fix it just rode up in the elevator with me at 10:30, so hopefully it should be fixed soon. Currently it's 88 degrees and it'll take a while to cool down, so Dr. Marchetti told me to work until I get to a stopping point and then go home.
The NMR is like the MRI that you get in to take cat scans and such, which basically makes it a giant magnet. To make the magnet work better, it's cooled, first with liquid helium and then with liquid nitrogen which keeps the helium cool. Thus, when it's really hot, like say when the a/c has gone out, it's a bad idea to use the NMR. So, I can't even set up any samples to run this weekend. The backlog, for anyone keeping track, is now up to 10 and rising.
So I spent all of this morning getting advise from
pretty much everybody about what type of statistical analysis I should do on
mydata to tell if the changes I want yo
make to my experiment changes the results.If I can make these changes to my experiment, it will shorten each run
to just a quarter of the time (half an hour instead of 2 hours) which will let
me catch up on my backlog.If it changes
my experiment, then all my data us no good and I'll be even farther
behind.So it's really important that I
do the right stat test to tell me if I can make the changes.
The only problem is, no one can agree on what test I
should use.Everybody can tell me which
tests are (next to) useless, but not which test I need.The two people who could tell me immediately
exactly what to do are not here for the next two weeks.
I know my second major is math, but statistics is a whole