I can't reproduce the results because they are fake

[trisha90]

Long story short:

Someone was working next to me and everything "worked" for them, after they left, I just could not reproduce that. The protocol is wrong, other things probably as well. But I have no proof. 

 

I am done with trying, my contract has ended and I did not get a new one (I guess that it might be related to the fact that I am "not able" to use the method). Now I really don't know what I should do, just leave and let it be and look like I am lacking the skills, or pushing the but-it-has-never-worked-before issue (though I should have noticed, the person was working next to me in the same room, but honestly, the work I was doing was so extremely time consuming that I was unable to keep an eye on her, plus she could not speak proper English and would just say "yes, yes, of course" instead of "I don't understand") when I don't have any hard proof? 

 

I hate it, I was working so hard and it was good for nothing (at this point I am actually not sure if I can publish a paper and thus graduate with what I have). I cannot even search for a job properly, because I don't know if I am going to get a Ph.D. or not. 

[Humulus_lupulus]

I don't know exactly what kind of experiment you are doing, but in my lab, it is not uncommon that people who have collected massive amounts of data have done it wrong (scary!).

The best advice I have is to think through the method and see if there is anything at all that doesn't add up in the materials you are using or the protocol that was given. Can you compare to published literature to see which set of results seems more accurate? Or see if a colleague can run the experiment? If you think it is wrong, there must be a way to justify that.

[spectastic]

maybe they're just infertile

[trisha90]

On 6/20/2016 at 9:42 PM, Humulus_lupulus said:

The best advice I have is to think through the method and see if there is anything at all that doesn't add up in the materials you are using or the protocol that was given. Can you compare to published literature to see which set of results seems more accurate? Or see if a colleague can run the experiment? If you think it is wrong, there must be a way to justify that.

 

I actually found quite some mistakes in methodology, but was repeatedly told to reproduce it with the flawed methodology first, and then we can move on. 

I don't know if I got all the mistakes, because seriously, some of them were really WTF? and one would not expect them from a senior high school student. 

 

It is not my problem anymore (except for the fact that I look like I cannot reproduce it, while the stuff is just plainly wrong), someone else is already working on it (already got a bit more freedom with changing the protocol in direction "correct"), but it is still totally annoying.