Data Agrees with 1 Author/Paper not the Other Author/Paper

Go to the profile of ops959
May 18, 2017
0
4

Please...I understand that I can argue or publish my hypothesis if the evidence and statement/claim are able to be understood as logical, valid, and not able to be currently disagreed with...meaning, that there is nothing that presents reason for doubt (yet).

What do I do if I have 1 author's model or hypothesis published that fits my data and one that doesn't. Do I still publish my paper or report or do I withdraw the validity? Or, do I state that it 'could' be valid and attempt to explain why one author's hypothesis and model are correct and the other's not? Do I argue in favor of my evidence without mentioning this (unethical)?

THANKS for any help/inputs!!!!!!!

4 Comments

Go to the profile of relaxin
relaxin 8 months ago

This is what science is about. The results should be reproducible by other investigators. In your case, you can discuss your results in comparison with the two groups of investigators, and try to explain why the results are similar or different, such as difference in methods, reagents, or experimental design.

Go to the profile of ops959
ops959 8 months ago

Thanks!

It would seem that each product comes with the instructions and recommended procedures...I can't honestly say that there is even one that I wouldn't in some way have had to modify or cleverly devise a way to complete the work due to problems or lack of results and what I felt were invalid results... e.g. using a lighter to burn off bubbles in a assay plate before reading it with the laser plate reader...

Go to the profile of relaxin
relaxin 8 months ago

The results should be obtained from triplicated experiments. Bubbles on ELISA plates would produce large standard deviations on the data, and thus will cast doubt on the validity of the results. This can be discussed in the "Discussion" of your paper

By the way, I use a hot inoculating needle to puncture the bubbles on the ELISA plate before reading. This will be less likely to add any artefact on the results.

Go to the profile of ops959
ops959 8 months ago

Yes definitely. I'll need to run the curves for each sample so as to indicate any error (also control).

Thanks for the advice...I'll pass it on to friends' to use as a proper technique!