I am co-authoring a paper and I am the 2nd author on this paper. We submitted the paper and it got generally positive reviews. The reviewer comments were not difficult to address. I have put a lot of time into this paper, and I am in an early and important stage in my career, so publishing is important to me.
The issue is that I don't believe the results of our study. My coauthor has sent me the data to address some of revisions, and some of the effect sizes are uncomfortably large. I worry that there has been a severe methodological issue that is now impossible to diagnose. I have spent a long time trying to figure out what could be wrong, with no success. My concerns are limited to one study only; the other studies replicate this effect but do not have impossible large effect sizes. I have asked my colleague about this and am awaiting a reply. Revisions are due in a few days.
Presuming they want to proceed with the submission, my options (as I see them) are to:
Do nothing. The paper will likely get published and I can ignore it. The paper is in my field, but I don't plan to follow up on this particular effect.
Retract my authorship and let my coauthor go it alone. Have nothing to show for my effort. Risk making a rash decision based on something potentially minor-- but without discovering any new information, I may never know.
What should I do?