My PI wont follow the evidence - (Jul/29/2009 )

We have interesting strong results but my PI won’t let me follow up on them because they do not fit his theory. Does this happen often?

I guess im just here to rant a bit i wonder if others have the same problems.

I obviously have to leave as soon as i can but this just sucks so bad.

Or perhaps it's because he doesn't feel the evidence is as strong as you do. Not knowing the guy, it's difficult to comment on this...

However, one ignores strong data contradictory to a current theory at one's own peril -- if the contradictory evidence is correct, and you pursue the other theory, you at best waste time and money, and at worst get shown to be in error in the literature when someone else publishes experiments that establish the new theory.

true, not knowing the situation you should side with the PI who should be more knowledgeable and have more experience.

I would like to know if others have the experience of their PI either

A.) flat out misrepresenting data

B.) push too hard to fit the data to their way of thinking

C.) not follow the evidence

But, this situation is not a danger to science. It will go nowhere, nothing will be published. (He will retire soon having published more opinion pieces than papers). To my mind this actually shows science does work, but it’s hard to be tied to something like this. Also i see there is something here that could be followed up. A strong result that i could try to elucidate why it happens, unfortunately i won’t be able to order the necessary probes... Of course it would only confirm the consensus view, still it’s something.

And yes i know i look bad even writing this, criticizing a superior and all, but it is what it is. It’s the situation as i see it and very stressful. I’m a lab technician, i do a job, i get paid so i guess i should be happy but I’m not. I can’t associate with this sort of work.

ejim on Jul 30 2009, 06:32 AM said:

I have once been warned off a PI by a technician who claimed that the PI had published results after swapping the data around and justified this by claiming that the technician must have miss-labeled the samples.

As for B & C, I suggest that it would be hard to find someone in science that is not guilty of one or the other from time to time. Just try not to get too stressed when someone interprets the data differently than you. Although, I can see how it would be frustrating not being given the resources to test your ideas; are you sure you could not sneak in a few additional experiments on your own time ?

I think your PI is one-track mind. By doing this the PI could get himself into trouble, and also could miss some important discoveries. Many important discoveries were made by mistake or by accident.

Not knowing the PI is hard to give an opinion, but I can tell you what happened to me.
During my PhD I started a mini side-project of my own because some of my data suggested it was a good idea to go that way. When I showed my PI the initial data he thought it was interesting, but didn’t want me to follow it because it wasn’t in "the original project proposal". Even after having presented the data at an international conference with a really good response to it.
To this, I nodded and kept going with the "proposed project", while doing plenty other experiments in my own time for the side project. Now, of course this meant long extra hours in the lab, and resources that luckily I had at hand.
By the end of all this extra hard work, when I showed my PI the final story, he loved it and the examiners said it was the best chapter in my Thesis; also my favourite. I haven’t managed to publish the stuff yet, but it is the piece of work I’m most proud of as it was all mine

In your situation, I don't know how much "freedom" to do extra experiments you have, but if you can sneak in a few to prove your point I'll go for that. If it doesn’t work, or doesn’t show what you want, just don’t tell the PI and there’s no harm made. If it works, and shows a nice story, he/she won’t really care that much in the end as long as is good publishable data.

I don't agree that you should do the experiments on the sly- it is the PIs lab, his funding and his perogative to use it as he wishes.

The situation is different if you are a PhD student, and obviously worked out for almost-a-doctor, but as a technician it is SO not your job to do experiments that you have specifically been told not to persue.
If I were a PI, I would be PISSED if a staff member went did something I asked them not too, using up valuable resources behind my back!!

Without knowing the details it is hard to know if what he is doing is wrong- or if you just have differing opinions on the same data. I don't think there is anything wrong with a PI not wanting to follow up on some results, quite often interesting things come up in your work but unfortunately funding and time are limited- maybe he just doesn't think it is interesting enough to follow?? Surely that is his decision, after all it is his lab so he can decide what direction he wants the research to head.

If I were you, I would try not to take it personally and just get on with your work the best you can. Of course if you really can't handle working like that then I think your only option is to leave and find a PI who does things in a way that is compatible with you.

Good luck figuring it all out.

Or alternatively, if you think that you've got the aptitude, willingness and drive (of course, time and money too) you might want to consider getting into grad school so you can have the opportunity to start your own lab eventually.

I have experienced

ejim on Jul 29 2009, 12:32 PM said:

by my supervisor/PI. While more experienced and knowledgeable, PIs/supervisors are still human. They can make mistakes and sometimes get blind sided by their own preconceptions of how the model/system/protein should work/react.

And yes, the scientific method does triumph in the end. But between now and then you waste a lot of valuable time, effort and depending on experiment design a chunk of money. It is little consolation to being able to say "I told you so!"

In this situation I believe one should do three things,

1 - at the very least, conduct experiments that would invalidate the false idea.
2 - run more/different experiments to validate your "interesting" observations. (They could be an artifact of the original experiment design.)
3 - try to run mini experiments that would validate/invalidate your own theory. Keep in mind, that your own ideas could be wrong.)

Sadly, running off on your own is easier said than done (almost a doctor, you lucky lucky thing)

I was unable to run my own mini-experiment. I had to run experiment after experiment to demonstrate conclusively that my PI ideas were not only wrong, but there was no quick fix/amendment that could make things right again. That cost me 5/6 of my PhD. The remaining 1/6 of my remaining time was used to restart the project from scratch. My own idea worked but the system I was working on wasn't completed due to lack of time.

So, I would say, if you are willing to stand by your data, you must fight. If you need to purchase new material, try to come to an understanding with your PI. If you can run your mini-experiments with material in hand, you should do them, even if you have to keep things quiet.

perneseblue on Jul 31 2009, 03:17 AM said:

Hi perneseblue,

Hope you're already out of that lab bringing with you that tiniest consolation of "I told you so". . Ejim's case is I think a bit different, he's a technician and altho he's a valuable member of the lab, how strong do you think his position is that he can fight for his data and results and that the PI would even listen to him? Probably with the technical and over-all lab management aspects, yes...but to take a different direction from his original plans, re-evaluate his initial hypotheses etc? There would be a lot of ego-related issues (Mt. Olympus) that will be stirred in this and that's definitely a part of being human too.

And logistically, how much free time does ejim have to be able to pursue these experiments on the side and what about the approval for purchasing what he needs when the PI had already decided not to continue with it? If he feels so strongly about his data, he can always make one last try of convincing his PI but if the PI absolutely refused to even consider them then he (ejim) doesn't have a lot of options open for him...he could stay and endure, or he could leave and go somewhere else or as I suggested before- go into grad school...

I guess you're in a better lab this time, pernese, with a cake-eating PI .......

casandra