I'm writing my thesis for an Engineering PhD. Our department has a tool for measuring trace metals (an ICP-OES) that I was supposed to use to collect my data.

Unfortunately this equipment proved to be unreliable, and at the time I did not know whether it was my experiment/hypothesis or the equipment that was at fault. I raised concerns about the equipment for years, but everyone brushed me off. It turns out that the equipment was defective after all.

I was awarded extra funding so that I could get samples analysed externally. These results were exactly what I had expected all along. My concern is that my thesis now seems pretty insignificant. If the equipment had worked from the beginning, I would have pursued several additional lines of inquiry.

At this point, there is no time/money to do more work; my advisor wants me to finish up. So: how should I address the rather modest scope of my research when writing and defending my thesis?

  • 2
    Hi Scaaahu. It was used by many people, and I was the only one to report a problem. For this reason I was dismissed for years until the technical team finally investigated and saw that all data collected over the time period was affected; I was the only one to notice that it was unreliable. So when I repeated my analysis and sent it externally, my results came back as expected. – user18483 Aug 4 at 9:15
  • 6
    Your advisor and/or committee are the best people to ask. – Nate Eldredge Aug 5 at 0:37
  • 2
    Maybe add a section on the consequences for research of trusting measuring equipment without regular checks? – Patricia Shanahan Aug 5 at 2:39
  • 2
    @Patricia Shanahan thank you, yes I'm compiling a short chapter on method development, it will contain some of the data that determined that the equipment wasn't working correctly, along with preliminary results and detailing the checks required to ensure that the equipment is analysing samples consistently. – user18483 Aug 5 at 2:52
  • 9
    Keep in mind, your advisor is the single most important person for getting a PhD. If they say you should get the degree awarded, you will generally get it. Talking with them about this should be your number one priority right now. – Roland Aug 5 at 6:20

I'm afraid there's no magic bullet here; all you can do is proceed as transparently and reasonably as possible.

First, present clearly the proof that the machine was broken. After years of debate about whether the machine was broken, your committee has probably stopped following the interminable discussion. So, you should explain very clearly and unambiguously that the debate is over and the machine really was broken.

After establishing that, I think the main question you should be prepared to address is: why did it take you years to notice that the equipment was broken? Naively, it seems like you should have done calibration runs in the first few days and identified/fixed the problem right at the beginning. Clearly that didn't happen here, so you should proactively and clearly explain why not. Hopefully there is a good technical reason (e.g., there is no way to calibrate this equipment) rather than just a mistake on your part.

Third, consider whether there is scientific value in the procedure you eventually followed to diagnose the faulty equipment. If identifying the problem was not trivial, it follows logically that your successful diagnosis is a significant step. You want to find the right balance: don't inflate your achievement or gloss over any mistakes, but at the same time, don't sell yourself short.

Finally, you should definitely discuss this with your committee in advance. If your professor wants you to graduate, you will probably pass your defense. Still, your defense is not good to surprise anyone. Since you have firm time/budget constraints, any potential issues or objections should be identified as early as possible.

| improve this answer | |
  • 6
    I am upvoting and would upvote another dozen times, if it was possible, because of your second paragraph! As an analytical chemist, I am almost speechless at how the instrument did not get calibrated for years on end. Apparently, not even spot checked with a secondary standard of some sort. This is like the horror of the old days, when an instrument was installed by a field service engineer, a calibration was run, plotted, taped to the wall over the instrument and then used year after year! – Ed V Aug 5 at 19:35
  • 1
    Great answer! thank you for all of your help. My supervisory team and whole research group know about the issues, and how for years I was trying to demonstrate to the powers at be that the equipment was not working as intended. That'll be my answer when asked what I'd do differently if I had my time again; run calibrations and known samples at the very start to ensure I definitively know that my analytical equipment is working, and to ensure that I knew how to operate it correctly. That's my big regret looking back. – user18483 Aug 6 at 3:32
  • 1
    @user18483 Best of success going forward! Also, as a three time grad program director in my department, my prognosis is that everything will work out fine: the most important thing, by far, is having your advisor actively on your side. – Ed V Aug 6 at 17:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.