[sklogic]

Given a huge skill gap and inability to factor out methodology differences, I cannot see how such a study can be done at all.

Anecdotes are the best we have. Far better than a pseudoscience with agenda.

[lolc]

Science can be analyzed for how pseudo it is, anecdotes cannot.

[sklogic]

Anecdotes are case studies. In social sciences, for example, it's often the only thing you have. Do not dismiss this kind of evidence when you do not have any other options.

[lolc]

Not all anecdotes are case studies, but all case studies are anecdotal evidence.

The effect of type systems is probably not that big. Otherwise the proponents of type systems would have had an easy time proving it.