No, I don't think I'm arguing that. It's not a question of any one methodology always being better than others, or correlation studies always being wrong. The q

No, I don't think I'm arguing that. It's not a question of any one methodology always being better than others, or correlation studies always being wrong. The question is what we should reasonably believe in light of several analyses of various strengths.