What they really mean is "no scientific evidence. If Crazy Carl says that aliens abducted his prize turnip, there is still no scientific evidence of aliens.
It's not like there is a "scientist" card that some people carry and some don't... and it's not like none of the hypothetical card-carrying scientists would ever have crazy opinions.
A publications doesn’t give you a free pass from scrutiny.
Sorry to seem pedantic, but I think it matters here since we're talking about specific meanings of terms.
Imagine Carl ran an experiment where he put a sign in front of various groups of turnips asking for things like aliens, Bigfoot, or faeries to take the turnips.
After running this experiment, only the turnips with signs associated to aliens were taken.
In this scenario, Carl would have some scientific evidence suggesting aliens took his turnips. It'd be weak evidence that likely wouldn't be replicated, but still evidence.
An analysis of compounds in DNA finds that there are, within a small margin of error, identical quantities of adenine and thymine, and identical quantities of guanine and cytosine. You have not yet been able to experimentally verify your hypothesis of a base pair rule. It is nevertheless not true to say there is "no evidence" of a base pair rule.
Sufficient evidence to form a reasonable hypothesis, but insufficient evidence to elevate your hypothesis to a theory, is still some evidence, and it is a lie to say otherwise.
Really? The scientific method is flawed.
The OP had a funny alien example, so someone stretched and came up with a silly 'test' for a hypothesis of the alien ate the turnip.
That doesn't mean 'testing' is flawed way of determining 'evidence'.
Then of course there are gradations below that. Observational studies, are just observations after all.
But because I can say "there is 'no evidence' that there are no aliens" is then the lowest bar, it doesn't mean there 'are' aliens.
But the point here is that 'no evidence' is being used as a rhetorical technique to promote actually false findings.
Kind of along the lines of simply injecting doubt into a debate in order to detract from real evidence. Like what has been done for cigarettes, sugar, etc...
(They probably didn't have turnips though, I hope that wasn't the point.)
The problem here isn't whether this is scientific evidence, but that the evidence isn't sufficient for scientists to form a reasonable hypothesis about what gives rise to the phenomenon (which could be (a) aliens are real, (b) some form of mass hallucination, (c) some form of synchronized fraudulent claim, (d) etc. etc. ).
I speculate that if somebody actually does a "literature review" of abduction reports they could pick out some signals to form some hypothesis, but presumably scientists have better things to risk their careers over.
2. This gets back to the point Scott is making - now define "scientific evidence". For any reasonable definition of scientific evidence, there is a lot of evidence for almost everything. I can find scientific evidence for aliens if I want to - to a terribly low standard, true enough - although what Scott picked on was homeopathy which works a bit better.
People underestimate lies, damn lies and statistics. There are facts that back up any worldview, no matter how insane. Carl could find scientific evidence to back up his theory. There'll be anomalous weather station readings, weird sounds, academics papers published by crazy academics that kinda support him. We've all met monomaniacal people; they aren't shy of finding scientific evidence for their case. Still bunk though.
Slight tangent, but I've seen a similar thing when trying to decide which religion, if any, is true:
Every one I looked at had at least a few very intelligent, seemingly rational adherents. It seems like there's a subjective element to what evidence each person finds persuasive.