Actually, this advice is spot on. Something that I keep reflecting on myself about the whole situation. I guess my fault in this case is that I became critical & belligerent about the issues that I perceived and I became the naysayers and then got excluded from the team. Although my view was validated by the reviewers, I feel that if I just got along, then I would be a co-first author by now. What should I do actually? Should I just get along even when I know that the ideas don't make sense?

Also, to be fair, "letting people try something and fail" may not work in this case because the PhD kept saying what he did was working until the paper got rejected. Then, I need to give a hard number basically proving that the PhD is wrong/misleading and it alienates him. Also, on another hand, in my first year, I kept saying the model was saturated, and my PI didn't believe me and he added the PhD student instead. After two years, we revert back to the original model. So, letting people fail can take 2 years of work..

Also, before the PI submitted the paper, I kept telling him that this paper going to be rejected because the reviewers going to ask a lot of questions about the PhD's ideas. He submitted it anyway and the reviewers did reject the paper because of it. Afterward, I didn't really want to help the group anymore because I felt that I was never being heard and I didn't want to help people who didn't want to be helped. Is this the correct way to react? Or I am not reacting professionally? Should I just help them after they fail?

Also, in this case, it did harm someone because the paper got rejected.. when it doesn't need to be rejected because we can just take out all the PhD's additions without any decrease in performance. But, the PhD was stubborn enough so he didn't want to do it and kept pushing it for submission.

By harm I mean physical harm or significant harm of any other kind. Not getting a paper published wouldn’t check that box IMO. Someone losing their job might (I say might b/c sometimes you can’t stop someone from walking off a career cliff). A whole team losing their jobs would be a significant harm and worth significant naysaying, though even there there’s a time say “I’m not the boss, they are” and accept that the responsibility for success or failure lies with them and you can’t change their mind if they’re unwilling to let it be changed.

If your only option is between naysaying so often you get sidelined, and watching your career get sidelined, then my point number four may come into play. That may be the case here. Your university may differ but from my memory there’s very little incentive for a professor to ever prioritize an undergraduate researcher’s advancement over or even on par with a PhD’s advancement. So you’re automatically going to be quite low in the pecking order no matter how amazing a job you do. It may simply be time to go to grad school. As others said you don’t usually need publications to do so.

Also sometimes you can help the truth of your predictions become evident faster by helping speed the work along, though this doesn’t always work (sometimes you’re strong armed away from the tasks that would let you do this) and it can be risky b/c you could set yourself up to be a convenient scape goat if you turn out to be right. Knowing how to navigate this type of situation is a skill that comes with practice. And like I said, sometimes you can help fix it, sometimes you can’t and you either keep your head down or move on.

One final thing I may add to my answer: take a look at how you’re communicating and maybe read up on learning how to speak persuasively, especially to people who have authority over you. It’s a difficult skill to learn but very important and it may also be in play here. You may be communicating the information accurately but not persuasively. Note that some people will not be persuaded no matter what and I’m not saying that’s your fault. But there is a skill that can be learned to at least improve your odds of success when communicating.

@justhalf "You can still be critical but not simply following along" Can you please tell me how concretely what I should do in this case, especially with someone with little power? If they want to finalize the model that doesn't make sense, then I need to use the model, right? I needed to help write the manuscript, create figures, benchmark for something pointless that I knew going to be rejected and it did get rejected and they rewrote all the stuff after discarding the PhD's additions. So, it's wasting my time as well.

@justhalf I can be diplomatically critical, but in the end, I need to follow along.. Also, besides me, two postdocs, and a master's student also questioned the PhD ideas, and the PI diplomatically asked the PhD to confirm whether it worked. The PhD said, "Yes, it worked". Then, I tried the idea myself on my code and it didn't work. So, a diplomatic way (even not from me) may not work with the PhD.

@bob so basically you are saying that I should get along even when I know they are objectively wrong and wait until the paper gets rejected by the reviewer and then propose a better idea, is it? Also, what I did was not naysaying before the ideas are tried, but I tried the ideas myself and it didn't work on my code and the PhD also could not reproduce his own result which was known before the submission. And the PhD is still pushing for submission. I guess I care too much about this project, if they want to submit it that way, I should just get along and go that way until it gets rejected.

@bob that's true as well, I could help speed the work until it gets rejected, which would take another year, but I might get my cofirst author. I am just the kind of person who cannot be a "loyal soldier" who can do something that I know is objectively not correct, especially since we are doing science. But, it is an opportunity cost to work on a project that I know is going to fail when I can work on another project that might be successful.

@bob Yeah I agree that I don't communicate correctly and diplomatically as well so I alienate people. I guess I became too much of a contrarian then when the PhD lied saying that things worked when they didn't, my PI believed him instead because I was too critical generally. So, that was my fault as well.

@bob I feel that no publications work when you apply right after your undergraduate, if I do research for several years and have no publications, it becomes a red flag instead...

It’s hard to say exactly what you should do not having all context (not being there with you) of your situation, but it sounds like there may not have been a path to success in your situation.

So either you needed to / need to find a different advisor or do something else, or you needed to find a way to work with them and try to make the best of a bad situation. One thing I’ve learned from many years in the workplace is that there are some situations where you just cannot win, but you can avoid making them worse and you can almost always leave if you decide it’s best to do so.

And I feel you on the passion for doing a good job. Just make sure not to let that accidentally drive you into unproductive working relationships with your coworkers. Sometimes (usually) the relationship is more…

…important than the specific project. Again a general suggestion. Not really in a position to help you figure out exactly what to do in your specific situation.

Yes, one and only path of success in hindsight was just to get along until the paper got rejected, then probably they would listen to me after the rejection and I would keep my authorship. Is that what you had in mind when you said I needed to find a way to work with them? Honestly, I was hoping that after I proved that the dataset was different and the result was not reproducible, the PI would be on my side.

but it turns out that the PhD (shamelessly) still keeps pushing to submit the papers even when he could not reproduce his own result... I was like, seriously? I feel that it's hard to work in research with someone who doesn't care about evidence and objective data-driven facts.

Yes, you are right, one learning point here is getting along sometimes is better than being right.. If I was just getting along, when the reviewers reject the paper, I could propose a better idea. But, since I don't get along, even when I was right, I could not propose a better idea since I got excluded.

To be honest, I also feel that they prefer the paper getting rejected than getting my help sometimes. So, I feel that they step on me sometimes (or maybe I am just too emotional). Also, the PhD clearly wanted to "win" by running away even though he knew the ideas that he proposed didn't make sense and we would be better off taking them off.

I also guess that it became my focus to prove my point because the PhD denied what I said, so the PI didn't believe me instead and it impacted my credibility as the PI kept cutting me off or mad when I said something didn't work or that the dataset is not consistent. So, it forces me to prove the PhD wrong to gain back my credibility which in turn alienates the PhD as I prove him wrong. Not sure how it can go differently though