First of all, do you have any formal training in research? I do.
Ditto. Graduate-level, as if it matters. Knowledge is knowledge; you're displaying a lack of understanding of basic methodology, so it wouldn't matter if all I had was a high school diploma. But your appeal to authority/snark is cute.
You're mixing and confusing a lot of terms in this paragraph. Why can't an anecdote have rigor? Do you think anthropologists realy on rigorless anecdotes? What about field tests of pharmaceuticals?
Again, it's data collection.
I'll not speak to anthropology, but I know there are several branches with varying degrees of scientific rigor. Regardless, I doubt most physical anthropologists, archaeologists, social anthropologists, and the like would prefer to describe their work as "a collection of anecdotes."
Don't conflate "collecting anecdotes" with "surveys." They're very different.
Field tests of pharmaceuticals aren't even remotely anecdotal. They often involve surveys, but surveys != anecdotes.
The problem in this situation would be failing to identify anecdotes in which those cures were not successful.
Again, it's not whether or not the cures were s successful that's important. It's the control and methodology used in the collection of data. This really isn't hard.
Performing a controlled experiment in most of the examples you give is completely impossible. For instance, it's not feasible to randomly select 1000 people between two groups, give half of them cancer, and then treat the cancer and the non-cancer groups with a randomly selected treatment, either placebo or the experimental treatment.
In the real world, you have to deal with people who already have cancer, and who agree to be part of an experiment. You compare the experimental treatment to some sort of control, usually the "industry-standard" or "gold standard" alternative, maybe both. You can't give cancer treatments to people who are well, so your only comparison is whether the new treatment is significantly better than what you were doing before. Which will not tell you directly, whether the new treatment works at all, or even if what you were doing before was better than no treatment at all.
I am starting to doubt any research credentials you claim.
Let's say we're doing a cancer study for a treatment. We'll take 200, 1000, whatever people; the methodology is the same regardless.
A simple double-blind would be, "Half get the real treatment, half get a sham treatment." Neither the experimenters nor the patients know which one they're on. Collect data at the end - surveys, results of medical examinations, whatever - and compare your treatment to both the placebo effect of the sham treatment and known industry standards.
None of the above is "collecting anecdotes" in any way, shape, or form. Information and knowledge are controlled from start to finish.
Similarly, we can't run two controlled experiments, one in which WotC opens up the OGL to 4e material, and one in which they don't.
That doesn't mean there is no data.
Of course you can't. And of course there's data. But neither "people on forums said this" or "my game shop owner said that" is meaningful data. You could approach such a thing with a degree of rigor with, for example, surveys with randomized or bias-controlled participants, but what you'd find is data on what people in forums and game shops are saying. If you think I'm saying the only way to find meaningful data would be to force WotC to make changes, then you're displaying a lack of imagination here.
A quick overview of qualitative research:
Qualitative research - Wikipedia, the free encyclopedia
You're welcome.
Sigh.
-O