PI Telling Me to Falsify Data

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
don't do it. personal morals notwithstanding, if you are caught, your academic career is completely over (especially if you fudge your data so much that you can submit to a top tier journal). You may not feel the ramifications until several years down the road, but you will regret this so much. Also, it is very possible that you will suffer immediate consequences when knowledgeable reviewers begin inquiring.
 
I wouldn't do it because it goes against the whole scientific method deal.

Now, if he/she really thinks that there is some trend that is not significant to the statistical test, then do a different test. If that doesn't work, then explain the problems you have with falsifying data and propose an alternative (like explaining the possible trend in your discussion).

Making up data would just put you back in the days where science was still subjective. Plus, it just looks bad.
 
You really shouldn't do it, the risk is too great. You'll never have scientific credibility if you somehow get caught. Don't be abrupt about it, either. You don't want to piss of the PI.

Is this a faculty member who is asking that of you?
 
explain! sounds interesting. if you are funded by NIH, then the data is of government property, so i would just be wary of that.

i have seen a lot of people under-develop westerns and not back up results with rtPCR b/c the blot doesn't look significant.

this has been known to happen though. if you falsify data now, you probably have to keep doing it to keep the research relevant
 
brooklynblunder said:
explain! sounds interesting. if you are funded by NIH, then the data is of government property, so i would just be wary of that.

i have seen a lot of people under-develop westerns and not back up results with rtPCR b/c the blot doesn't look significant.

this has been known to happen though. if you falsify data now, you probably have to keep doing it to keep the research relevant
Click to expand...
This type of stuff happens all the time in science, particularly in really high impact work, and it is too bad.. although in my lab RTPCR [steady state mRNA concentration] is no way to confirm results by western [steady state protein concentration].. there are a number of cases where a protein could be post-translationally destabilized and would look normal by RTPCR..
 
obviously not, but the theory is that if there's mRNA, there should be protein.

if you knock down a receptor, but the wt to siRNA western is unclear, back it up with the quantitative rt-PCR, at least

now if western is clear, its absolutely unnecessary

obviously, functional assays w/ controls are the preferred way to go (esp since you probly want to publish)
 
DendWrite said:
PI is telling me to fudge some numbers to make insignificant results significant. What to do?
Click to expand...

Welcome to the world of modern science. If you have any self respect, don't do it. Unfortunately, it seems like many scientists don't have a whole lot of self respect.
 
if the PI wanted numbers fudged why wouldn't s/he do it him/herself?
 
brooklynblunder said:
explain! sounds interesting. if you are funded by NIH, then the data is of government property, so i would just be wary of that.

i have seen a lot of people under-develop westerns and not back up results with rtPCR b/c the blot doesn't look significant.

this has been known to happen though. if you falsify data now, you probably have to keep doing it to keep the research relevant
Click to expand...

Hell that would suck! I mean, how do you know where to keep fudging the numbers to after a couple months roll by, then years, then decades..... Eventually, generations of that research will have to start creating numbers with some new equations just to keep the trend going. The experiment would create a whole new age of science where science in the mind becomes real versus imaginary.
 
DON'T DO IT.

Your career is on the line. Seemingly small incidents like this have ruined scientific careers in the past. Not just the PI, everyone in the lab gets screwed. Trust me, we had to take an entire course on Ethics in Research. This is serious stuff.

In fact, you have an ethical responsibility to report such misconduct. It's your call on whether or not you want to do that. But do not become a victim yourself.
 
redlight said:
if the PI wanted numbers fudged why wouldn't s/he do it him/herself?
Click to expand...

so if anyone finds out the PI can just blame it on the student and say it was done behind his back

i'm absolutely shocked by the responses to this thread. the only two appropriate responses which noone has even come close to suggesting are:

1. bring this up with the head of the department or alert the proper academic integrity personnel at your university. they'll thank you for it because the last thing they would want would be for this person to publish papers with false data that would have to be retracted.

2. get as far away from that lab as humanly possible. a lab is only as good as the PI running it...and if this guy is crooked then you don't want to be in his lab. also if he gets caught falsifying data you will be under suspicion as well and all the time you spent in lab will be at best worthless and at worse a detriment to your future
 
DendWrite said:
PI is telling me to fudge some numbers to make insignificant results significant. What to do?
Click to expand...

Pllleeeeaaaseee tell me this post was in jest. Otherwise the simple fact that you were sooo confused as to what to do in this situation that you had to pose it to an anonymous internet forum, well it doesn't speak well about ya.

This answer should be an obvious no. Not a "well maybe", or even "I don't know". Simply "No".
 
skinMD said:
so if anyone finds out the PI can just blame it on the student and say it was done behind his back

i'm absolutely shocked by the responses to this thread. the only two appropriate responses which noone has even come close to suggesting are:

1. bring this up with the head of the department or alert the proper academic integrity personnel at your university. they'll thank you for it because the last thing they would want would be for this person to publish papers with false data that would have to be retracted.

2. get as far away from that lab as humanly possible. a lab is only as good as the PI running it...and if this guy is crooked then you don't want to be in his lab. also if he gets caught falsifying data you will be under suspicion as well and all the time you spent in lab will be at best worthless and at worse a detriment to your future
Click to expand...

QFMFT. end.
 
skinMD said:
so if anyone finds out the PI can just blame it on the student and say it was done behind his back

i'm absolutely shocked by the responses to this thread. the only two appropriate responses which noone has even come close to suggesting are:

1. bring this up with the head of the department or alert the proper academic integrity personnel at your university. they'll thank you for it because the last thing they would want would be for this person to publish papers with false data that would have to be retracted.

2. get as far away from that lab as humanly possible. a lab is only as good as the PI running it...and if this guy is crooked then you don't want to be in his lab. also if he gets caught falsifying data you will be under suspicion as well and all the time you spent in lab will be at best worthless and at worse a detriment to your future
Click to expand...
ha, "it's my study but i did not know subordinates were fudging data in my favor."

that would definitely not fly. im sure the PI would take a lot of flak either way.
 
I hate when people ask me to lie to somebody for them. That is so rude. I would be insulted if I was asked to lie to the scientific community. That person is obviously not professional and doesn't care about science, but rather cares about being some sort of nerdy rockstar with "significant" results. Figuratively speaking, we need to hang people like that. We don't need people like that in the world. They will take us down the wrong path and hold us down, all for their own personal gain. Report that person. 👎
 
Absolutely not worth it. As has been said, you actuaally have an ethical duty to not only refuse but also to report ethical misconduct. It's a big deal and, if caught, WILL END your academic/research career as well as bring some very serious questions to the table concerning medical licensure (i.e., medical boards are going to wonder whether you can be trusted to practice medicine safely, with good judgment & with integrity).
 
If you are caught, it is over for your professional career.
 
Don't do it, and get out of that lab as soon as possible. Try to join another lab if possible. If you don't want to confront your PI, then make up an excuse. But definitely change labs.
 
The benchmark for 'significant' or 'highly significant' are, sadly, subjective viewpoints taken as scientific fact. If the p-value is 0.06 then argue that such a small variation in the p-value doesn't make too much of a difference. I've heard doctors argue that a 0.10 p-value for some life/death treatment is good enough because death is imminent enough that it's worth the risk. Your PI could certainly lie about things and get away with it--but the guys behind cold fusion lied and we saw where that went.

The saddest thing that could possibly happen is that your PI was onto something but his results were off and he lied. Then no one believed him and, by proxy, didn't believe the results.
 
What does P.I. stand for? Something equivalent to "research advisor"?
 
Wow your PI is so cool I wish I could have one like that...

seriously though, dont do it.if you are the one who make changes, it might come down to you who is to blame...and the PI just might get away with unethical thing like this...
 
If the P value is simply to high, run a few more subjects or w/e it takes w your specific experiment. Esp. If there's a strong trend already and assuming the results really are significant, you simply need to increase the power (sensitivity) of your tests. P is somewhat arbitrarily set but the idea is based on a 95% (or 90% or 99%) degree of confidence in the results. It has to be set somewhere or we'd get studies with varying degrees of significance and you can imagine the mess! As it is there's a definite bias toward positive results being published.
 
Thanks for all the advice. No, unfortunately it's not a joke. Obviously he wasn't like "hey, make up some numbers," but he did tell me how to quantify western blot bands, and to basically "ignore" visible bands on some samples and not others. I know his hypothesis, what we're looking for, and omitting these bands obviously strengthens the results (we're talking a p of 0.01 vs. 0.15 in this particular instance). In other cases he's done some of the analysis, but I ran the films, and after looking at the film and comparing it to the data, I have doubts that they match up. Would a reviewer catch this? Maybe, maybe not. I'd lean toward the maybe not side...there aren't many papers out about what he's researching, and the data hasn't been tampered with so much that it looks completely unbelievable.

It very well could be a case of not enough statistical power, as a lot of our data shows "trends" (which may or may not be statistically valid). This particular PI is leaving for another research institution in the next few months, so there is a lot of pressure to get the project finished.

As far as what to do...I feel like reporting him will get me into serious trouble. I don't condone what he's doing by any means, but on the other hand, I don't feel like it's my place as an undergrad to potentially tarnish this guy's reputation. Ideally I ensure that he doesn't give me any credit when he publishes and leave the lab.

Sorry for the post and if it reflects poorly on my own ethical standards. Just to be clear, I don't condone what he's doing at all, but I'm just really taken aback by all this. I knew that science wasn't this idealistic enterprise that I once thought, but I hoped that I'd never run across people who just make stuff up. I guess it's just a sad fact of life that not everybody's honest. I was hoping to see if any of you had this happen to you as well and how you dealt with the situation. Thanks for the input.
 
what should you do?

write about it in response to a 2ndary prompt about moral and ethical challenges
 
If its any consolation, at least you have a strong contender for the "moral or ethical dilemma" essay.
 
i see where you're coming from. its of course CLEAR what is right but its not that easy to report somebody just like that. i think its something to do with being brought up not to "tattle" because in childhood we all hated that kid that was self-righteous enough to do it. still, it is the right thing to do in such a situation. its just hard to think about killing someone's career...

i would say though, to quit. right now. maybe you won't report him, but you shouldnt be part of this and who cares if he has to get the project done? let him know he's doing wrong and its against your morals/principles and you think its time for you to leave. you don't need this right now.



but, i donno. personally, i 've never dealt with such a situation. i do have to deal with a graduate student that doesnt seem to know what he's doing, i dont know how he got in our lab, holding us back, and only shows up for like 3 hours out of the day and probably making up data as well cuz he can't work the damn technology. but *sigh* i just make up for his lack of work, and feel i can't say **** because i'm "just an undergrad". so the rest of us talk about him behind his back.. but our PI is too nice to admit he made a bad choice accepting this guy to our lab.
 
Where are you heading to musafirah? 😉

On topic, what a weird thread. I can't even imagine my PI saying such a thing. Maybe thats cuz he's a renowned, celebrity-status researcher around here.

Man, these ethical questions are tough. If it happened to me I'd let him know right away that I'm not cool with that. But is the best option to go a step further and report on him, and devastate his career?! I dunno if I could do that. Where's Compass at? 😀
 
I must point out that while I can analyze moral viewpoints of THEORETICAL standpoints very well, that taking all this advice in REAL LIFE is quite different. I'd like to point out that IF statements are not actually occurring. When it happens in real life, you will undoubtedly be whacked by emotions. Viewing a real-world scenario without emotions is dangerous. Unlike moral dilemmas where it's not actually happening, some people WILL find it difficult to actually turn in a cheater they see. Would you turn in a deranged psychopathic murder for cheating if you knew that 10 years down the road he'd kill you? Morally, yes, realistically, you'd be running like hell.

Moral ethical dilemmas are never black and white in real life.

The morally correct thing to do is to report, but you will be ruining your PI's life if you do, and/or putting yourself at jeopardy for what people might see as a "fake allegation." You need to be careful and cover your arse in real life as well. I wouldn't know how to approach this directly having never done so, but a lawyer and SEVERAL indepedent committees and review boards need to be consulted and/or informed if you don't want to have this bite you back.

This is a situation that sometimes appears in the news: whistleblowing
http://en.wikipedia.org/wiki/Whistleblower

Reprisal is QUITE dangerous for you in this case.
 
Compass said:
Viewing a real-world scenario without emotions is dangerous.
Click to expand...

wow compass, you really do give good advice! reading this sentence really resonates with me because I believe I usually do just that, and get confused when people act strongly with their emotions. you have a great moral "compass".
 
At a minimum, you need to leave the lab and hopefully find another one. Whether you should report him or not is more ambiguous.

schrizto said:
wow compass, you really do give good advice! reading this sentence really resonates with me because I believe I usually do just that, and get confused when people act strongly with their emotions. you have a great moral "compass".
Click to expand...

I think he was trying to say it's foolish to assume people will be rational. In this case, don't expect to ruin somebody's career without people hating you, even if it might be for the greater good.
 
As everyone else has said, don't do it and report him. However, I would add that you should go talk to a statistician to ensure that you are running a test that is truly testing what you are interested in. For example, you said that there was a trend--there are special tests for trends (as opposed to means). Running a proper test might make your data "significant."
 
Working out things in an ideal world and working things out in the real world tend are on completely different scales. The OP could very well accidentally ruin his future career and life, and I'd rather that not happen. There are a LOT more possibilities in the real world that would do more harm than good for the OP.

Bringing such allegations foolishly without preparation would be like saying that you have evidence the PI is an alien. Tread CAREFULLY. VERY CAREFULLY. Cannot stress this enough.

Calling this PI out could be the equivalent of "shouting wolf" 3x with no wolf. There's no proof, so you have nothing to back up your claim.
 
What's the point of these theoretical examples if they are on completely different scales from real life?! I suppose there's two right answers: an abstract, idealistic textbook (or interviewer) right answer.. and then there's the practical right answer. I'm interested in what you should actually do if x particular moral dilemma befall you.. not what they would to in planet perfect, or planet emotionless, etc.

What I think I would do (in this PI situation) is tell him I'm not cool with changing the numbers if those stats were part of an insignificant test/experiment/etc. If he was asking to change some pretty important numbers, then I'd refuse and tell him that this is completely unethical/unscientific/wrong and kinda judge from his response and expression and from my conscience where to go from there.
 
Compass said:
Bringing such allegations foolishly without preparation would be like saying that you have evidence the PI is an alien. Tread CAREFULLY. VERY CAREFULLY. Cannot stress this enough.

Calling this PI out could be the equivalent of "shouting wolf" 3x with no wolf. There's no proof, so you have nothing to back up your claim.
Click to expand...

given OP's updated post detailing the situation i agree with this. it is not as clear cut as OP made it out to be in the beginning so be very cautious because you are playing with fire and little proof.
 
YeOldeMan said:
You really shouldn't do it, the risk is too great. You'll never have scientific credibility if you somehow get caught. Don't be abrupt about it, either. You don't want to piss of the PI.

Is this a faculty member who is asking that of you?
Click to expand...

what if the risk were not as great? Would your recommendation change? :meanie: 😀
 
Honestly IMHO I think that if you have to ask, you *know* that it would be wrong to do so. Definitely don't do it. Suggest doing more tests or something, or leave if you have to. Unfortunately, I wouldn't suggest reporting the behavior unless you can do it anonymously.

I do think though that ilikekiwis is spot on about writing about it or talking about it when asked about moral dilemmas 🙂
 
DendWrite said:
Thanks for all the advice. No, unfortunately it's not a joke. Obviously he wasn't like "hey, make up some numbers," but he did tell me how to quantify western blot bands, and to basically "ignore" visible bands on some samples and not others. I know his hypothesis, what we're looking for, and omitting these bands obviously strengthens the results (we're talking a p of 0.01 vs. 0.15 in this particular instance). In other cases he's done some of the analysis, but I ran the films, and after looking at the film and comparing it to the data, I have doubts that they match up. Would a reviewer catch this? Maybe, maybe not. I'd lean toward the maybe not side...there aren't many papers out about what he's researching, and the data hasn't been tampered with so much that it looks completely unbelievable.

It very well could be a case of not enough statistical power, as a lot of our data shows "trends" (which may or may not be statistically valid). This particular PI is leaving for another research institution in the next few months, so there is a lot of pressure to get the project finished.

As far as what to do...I feel like reporting him will get me into serious trouble. I don't condone what he's doing by any means, but on the other hand, I don't feel like it's my place as an undergrad to potentially tarnish this guy's reputation. Ideally I ensure that he doesn't give me any credit when he publishes and leave the lab.

Sorry for the post and if it reflects poorly on my own ethical standards. Just to be clear, I don't condone what he's doing at all, but I'm just really taken aback by all this. I knew that science wasn't this idealistic enterprise that I once thought, but I hoped that I'd never run across people who just make stuff up. I guess it's just a sad fact of life that not everybody's honest. I was hoping to see if any of you had this happen to you as well and how you dealt with the situation. Thanks for the input.
Click to expand...

If the PI's asked you, it's also possible that he's asked other students. If that's the case, and one of those instances is discovered (and most scientific misconduct eventually is) then your association with this PI could be damaging to your career. I think that the best case is to leave the lab, or, as you've suggested, to not put your name on the paper if you don't agree with the results. He's probably asking you to do falsify the data so that he can cover his career -he'll take flak for "allowing" it to happen in his lab, but your career will be over, as he would be able to prove that you falsified the data. I agree with other posters that without proof, going forward with an allegation as serious as this will be more damaging than productive. A confrontation with the PI could also be very dangerous if not treated VERY carefully.
 
You must log in or register to reply here.

Similar threads

JSmith42
What data does AMCAS transmit to medical schools?
Replies
2
Views
601
Mr.Smile12
Mr.Smile12
S
Questions to ask PI (CRC Interview)
Replies
3
Views
2K
GoSpursGo
GoSpursGo
Mayajae
  • Locked
Why are schools not accepting urm's in the 2024 msar data?
Replies
24
Views
2K
wysdoc
wysdoc
S
MSAR Waitlist Data
Replies
6
Views
3K
gyngyn
gyngyn
Z
Psych me up to withdraw
Replies
8
Views
1K
Hollow Knight
Hollow Knight
Top