I'm not a fan of the "less scientific" phrasing, but the point of science is to be convincing to even a skeptic. That's why reproducibility (try it yourself in your own lab!), statistical analysis (there's basically no chance this is a coincidence!), control groups (it's not just an endemic property of the lab), random sampling (it's not selection bias) and other things are so important.

In some kinds of inquiry, it becomes difficult or impossible to apply certain standards. It's flat out unethical and immoral to not treat a man's syphilis just to have better quality evidence. So if you're studying syphilis, you need to find other ways to be convincing. It's impossible to have a statistically significant sample of Earths or a control group of Earths, so the bar for convincing is also higher in climate science.

"Less scientific" probably means "doesn't have access to many standard scientific techniques, so stronger evidence in other ways is essential to be convincing".