I think if it were to make sense, it would have to be similar standards to perjury: wilfully making a materially false statement about your data or process in a published paper. I.e. only targeting outright fraud in statements of fact about what you did and observed. The conclusions from that data would not make sense to include in that umbrella: firstly because as you mention they are difficult to judge and are in fact for pretty much any paper up for debate, and secondly because that's the part that is expected to be assesed by the process of peer review and pulishing anyway.
I'd say that at the moment there's a bit of an issue with the way the community handles this kind of thing, in a way which is akin in structure (I'm not comparing severity/morality) to sexual assault in many communities (science also among them): it's sadly common that someone is widely known or suspected within their field to engage in scientific fraud, but it's only known within that because that person has enough power to make it dangerous to overtly make an accusation, as well as a general fear that it will discredit the field in general. And someone with a bad reputation there still often gets to engage with the community. It seems that only in the really high-profile cases are there actual consequences, and even then they often only come out long after the offender has retired.
(I'm not entirely convinced criminalising it will actually reduce the problem, though. The idea that harsher punishments = less misbehaviour is a bit of a fallacy in part because people who do this don't expect to be caught)
I'd say that at the moment there's a bit of an issue with the way the community handles this kind of thing, in a way which is akin in structure (I'm not comparing severity/morality) to sexual assault in many communities (science also among them): it's sadly common that someone is widely known or suspected within their field to engage in scientific fraud, but it's only known within that because that person has enough power to make it dangerous to overtly make an accusation, as well as a general fear that it will discredit the field in general. And someone with a bad reputation there still often gets to engage with the community. It seems that only in the really high-profile cases are there actual consequences, and even then they often only come out long after the offender has retired.
(I'm not entirely convinced criminalising it will actually reduce the problem, though. The idea that harsher punishments = less misbehaviour is a bit of a fallacy in part because people who do this don't expect to be caught)