Xjyuk

I am a final year PhD student in chemical engineering. Since the start of my PhD, I was clear on one personal philosophy. I will put my name only on those papers where I have made substantial contribution. And I will put my colleague's name on a paper only if he/she has contributed substantially. That's what collaboration means to me and anything else seems fraudulent.

However, my group members frequently put each other's name as co-authors in their papers even if they work on completely different topics and have no contribution in the paper. Now, this is unethical in my perspective, on the other hand, they have 10-15 papers by the time they graduate which increases their chance to secure a postdoc position or a tenure track position.

My advisor says, granting of co-authorship is entirely up to the first author and she doesn't interfere with the process.

I have experienced similar sharing of authorship during my masters degree as well. Past PhD members or postdocs were given co-authorship in spite of not contributing anything.

I will be getting 4 (first author) + 2 (co-author) papers from my PhD, which is far less than my colleagues' output.

My colleagues often say that I should have been more collaborative (i.e. share authorship without contribution) as that would have increased my publication count and helped everyone. I simply can't see myself doing that.

What's your opinion about this? Have I severely affected my chance of a future in academia by not taking part in the authorship sharing practice?

I can tell you from the perspective of a person who was on hiring committees that these kinds of ethical indiscretions are easy to spot and are not well received. To the point, we rejected several applicants because we suspected they weren’t sufficiently independent after graduation as they weren’t lead authors on enough publications.

First of all, word gets around. If a PI lets their group slap their names on random papers, hiring committees will take notice, and will not take these applicants seriously. Second of all, we ask questions! If you obviously know nothing about papers you coauthored (yes, it’s obvious), that reflects very badly on you. Finally, if I were you I’d emphasize that on the few publications you have you are the lead author.

That said, your colleagues can partly get away with this: publishing is a numbers game and I don’t know how hiring committees think in other places or fields.

Fraudulant "collaboration" is caused by perverse incentives

One of the main things that contributes to this problem, in my opinion, is having ridiculous anachronistic academic and citation metrics that do not adjust for authorship (e.g., the raw h-index). Ironically, there is a large and well-developed academic literature on metrics that adjust for authorship contribution, but this has not made its way into the practice of universities. Most universities still look at crude citation metrics that do not adjust for the number of authors on a paper, and this creates an incentive for the kind of fraudulent cross-authorship of papers.

Authorship adjustment is quite a complicated field, owing to the fact that the contributions of authors on a paper may be unequal, and the authorship order can give information on this (depending on the field). Nevertheless, the basic principle of every proper metric is that the "total value" of a set of papers to a set of authors should not be able to be increased (or decreased) by spreading the authorship in a different pattern. Even the most crude authorship adjustments remove the perverse incentive problem that leads researchers to game the system by cross-authoring papers where they haven't done the work. If you adjust authorship metrics for authorship then this problem disappears --- being the full author of one paper with 100 citations is as good as having fifteen papers each with 100 citations and fifteen co-authors.

I can understand why you are concerned about this problem. It punishes researchers that do papers as sole authors (or in small groups) compared to similarly situated researchers who do research in large groups. Over time, it would be my hope that university practice will catch up to the development of the literature on this problem, and authorship-adjusted citation metrics will start to be used in practical decision-making (e.g., hiring, promotions, etc.) instead of the anachronistic indexes. If that occurs then you will eventually see this practice dissipate, since there will no longer be a perverse incentive to do it.