Validation

I have been so busy with moving preparations that I haven't had much time for blogging. Between boss screaming for more data in different ways, packing, loading and unloading, I have been run ragged.

Moving details in it's ugly details later....

Last night, boss called me at home. "I'm not coming in tomorrow, so please send me the file you've been working on when you finish it."

"Yeah sure."

He shows up anyway. Which is good since webmail didn't send my attachment anyway.

That wasn't the only miscommunication. He wanted the data in a different way.

Why didn't I think to try it that way? Because you CAN'T. I'm showing untreated controls and how they grew respective of prior nicotine exposure. He wanted that information normalized.

Normalized to WHAT? There's nothing to normalize it to.

But he's still screaming for the data. So I compromise. I take averages of a set of ranges. I normalize the average to the non-smokers. Since the template pretty much was there, it was simply plugging in the new numbers, then getting rid of error bars and hitting print.

I knew it wasn't what he wanted. Sure enough, I show him the data, and he said "what happened to the error bars?"

I take in a deep breath and try and explain that there is no other internal control to normalize to. Since each smoker was done separately, they can't be normalized to non-smokers done at a different time.

He isn't getting it. I say "let the post-doc try and explain since I consulted him on this data." Really thinking "let post-doc repeat what I just told you."

Sure enough, post-doc backs me up and repeats what I just told him. He says "Oh, okay."

What a sexist jerk. I even joke with post-doc that it isn't because he has a PhD, but something underneath the shorts that makes him believe post-doc over me.

So glad to be leaving. Speaking of which, since I have a moment, I better put into writing that I'll be leaving.