3

I'm almost at the end of my research dissertation for my MSc program and my results could not prove any improvement in the techniques already well established. I relied too much on a paper talking about some interesting application and results but when I went deeper to apply the technique in a more careful and thorough way, it appeared to be flawed. I still think I did a great deal of work to reach these results and I learnt a lot during my path. Should I try to fix it or just show what I got?

Buffy
  • 363,966
  • 84
  • 956
  • 1,406
Mr Frog
  • 657
  • 8
  • 12

2 Answers2

5

While your advisor can probably help, don't misunderstand the purpose and process of research. It isn't making some unsupported claim at the beginning and then "proving" (dammit) that your claim was accurate. Research is a reach into the unknown. The answers you get depend on what is, not on what you want them to be.

You are actually seeking knowledge (truth), not verification. If the results of your research disprove your original hypothesis that is still a valid result.

If your "new" technique isn't actually better than the old, then it is useful (maybe vital) to know that.

In an ideal world you should be able to write that up. We hypothesized X. We used process Y to explore it. We found no evidence that X is true. That is still knowledge.

Hopefully, however, your advisor has such a view. And hopefully your process Y was sufficiently sophisticated that you haven't fallen in to easy errors.

If every hypothesis came up with positive (supporting) results, we wouldn't be working very hard and only making trivial initial hypotheses.

But, start out with "what is true here" rather than "this is true and I'll beat it until it yields". That leads to propaganda and isn't research. It is often the basis of misconduct. Tobacco and pesticide "research" too often fall into such errors.

So, it isn't "bad" at all if you reach truth.

Buffy
  • 363,966
  • 84
  • 956
  • 1,406
  • Related: https://academia.stackexchange.com/questions/30995/what-to-do-when-you-spend-several-months-working-on-an-idea-that-fails-in-a-mast/31082#31082 – Ethan Bolker Jun 07 '20 at 21:58
  • 1
    I would also expand on this for OP by suggesting that, at least if the MSc is your first dive into research, half of the battle is learning what research really is and how you navigate yourself conducting research. Now, when you approach your PhD, you'll be significantly more equipped with better habits, efficiencies, and skills that you can now use. I have found (me being no exception) that the grand focus on results is something hammered into us as undergrads and there is much more to an MSc thesis than getting '"good" results. Do not lose sight of how your struggles have made you better. – GrayLiterature Jun 07 '20 at 22:17
1

It's not bad at all

That sounds like a nice piece of research to me. You tried a new method to do something, and found out it doesn't work as well as the existing methods. Useful stuff. Write up your results and be honest about the outcome; see if you can explain/hypothesise why your new method didn't work as well as you were hoping. That research will be really helpful for anyone else who comes along and wants to develop this new method along similar or different lines to you.

Ben
  • 68,453
  • 9
  • 142
  • 263
  • Perhaps a useful answer for future readers but a year and a half late for the OP. – Ethan Bolker Jan 14 '22 at 00:44
  • Late answers are totally OK -- as you say, they may be useful for future readers (and in theory, we only take questions that may be of interest to future readers). For the curious: our late answer queue exists because a large fraction of late answers are also bad answers, not because late answers are intrinsically bad. – cag51 Jan 14 '22 at 03:14