Agree 100%. People who engage in outright fraud won't be deterred, as they will still cheat. They can pre-register, make up data, then post the made-up data. Questionable research practices occur because of the incentives and norms. I elaborated here: https://paulspector.com/pre-plan-even-if-you-dont-pre-register/

Thanks for your reply, Paul, and for the link!

Any and all types of Bad Behaviour can only be reduced if the incentives to behave badly are offset by a high probability of being identified as a miscreant and suffering serious consequences. Benefits with low consequences mean that people of low moral character rise through the academic ranks to where they encourage others to behave badly and that feeds into a culture of bad academic behaviour.

Chris Wright

Open science may help to decrease scientific misconduct, but it cannot be the only solution. Some kind of misconduct is also possible with open science, and for various reasons not all science can be open (e.g., to secure patents).

Dear Dr Thomas E. Becker . It could be reduced the misconduct. But It has other advantages: it increases productivity in an era of tight budgets. Open science results in great innovation potential and increased consumer choice from public research. Open science promotes citizens' trust in science. Greater citizen engagement leads to active participation in scientific experiments and data collection. See the link: https://www.oecd.org/sti/inno/open-science.htm

Surely.

In your question also the answer: as long as "money at all costs" is the main objective of so-called "scientists", research will always be marred by bad practices (with everything that comes with it).

Totally agree with the main idea: to consider integrity and productivity at the same time.

Thanks to all for your comments! It sounds as though there is some consensus among us that open science is fine but, by itself, cannot be expected to have much affect on scientific misconduct. Instead, the incentives for misconduct need to be addressed. (If you disagree, by all means say so - I don't mean to put words in your mouth.) Now, let me give you an EXTREME position to react to, with no implication that I necessarily agree with the position:

"The only way to greatly reduce misconduct by assistant professors is to eliminate the tenure system. Tenure is the largest incentive for scientific misconduct because assistant professors get lifetime job security if they are 'successful' and lose their jobs if they are not. Although other incentives are important (e.g., promotion, grants, reputation), nothing much will change until tenure is eliminated or, at least, drastically modified."

Please note that the quotes represent the position of a hypothetical critic of tenure. They are not from a published source. Looking forward to your thoughts!

Tenure provides vital protection for academics who research unpopular topics, or take a teaching/research stand that threatens the wealthy/powerful, or aids those being preyed upon by the privileged. Societies that wish to progress, need to give their researchers fair and reasonable access to tenure.

In my experience, the greatest threat to academia comes from Universities that over rely on student evaluations--good lecturers should challenge their students and sometimes learning can cause difficult or even painful cognitive dissonance. Over reliance on student evaluations tends to lead to mediocre graduates who were over catered to during what should have been a challenging education.

Students want to be taught facts, which tend to have a half-life of 3-5 years, when they need to be challenged with learning the craft of identifying relevant questions finding currently useful answers, and combining the questions and answers into succinct and lucid reports (of no more than 1 to 1.5 pages long).

Thanks for your thoughts, Chris.

It depends on what you mean by scientific misconduct.

If you mean data dredging and p-hacking, it will definitively help.

If you mean harking, it can help in some cases, but with the possibility to add "explorative analyses" you can add non-planned analyses and still get your research published in prestigious journals. Maybe it is not exactly harking, but I am not sure it is so different, in the end. I think that additional help might come from multiverse analysis, but you need to add a lot of extra work, and often the times and necessities of modern research do not reconcile with this.

If you mean the production and use of fake data, there is no defence on this (in particular if the production of fake data resembles realistic data... anyone good in data simulations is able to produce realistic data samples with the desired effects).

In the end, it always depends on your personal ethics.

Probably, a different conception of the merit in research, not only based on productivity might have a way better effect in reducing scientific misconduct.

Gioacchino de Candia, I agree with you, there is no doing right in a (more or less and lesser subtle) capitalist order. It just ruins everything, also sciences. Moral became a subordinate. In therapy settings, I hear about the misery of the (so many) people bearing the weight of remorse about what they decided and did, when they felt "pressured by the context" to act against their intuition and personal beliefs, and they will never get rid of those feelings because many things can not be made right in retrospect. Inhumanity as the general system is at the end of the day simply inhuman, it's that simple. I am not sure whether harder punishment would help, because the subjects know and suffer already. With more or less slight modification within a system, probably little will change, as the principles remain the same.

FYI, as an update on this conversation: I recently gave a TEDx talk on the topic of scientific misconduct. If you wish, you can view it at the below link (or just search for it on Youtube). I also welcome any comments, pro or con, you may have on the talk. (Apologies if you have already seen it.) Tom

https://www.youtube.com/watch?v=PLgy-i-RwhI&t=481s