Okay... I agree with this: while it's important that we do look at even what's working to see how we can make it better, it certainly needs to be purposeful, and with a clear cut idea of why we think it could actually affect change (whether initially based on research or not). When I changed approach to independent reading, it perhaps might've been a "new method" (though not really), but there was significant purpose behind reaching for the change: I had significant uncomfortabilities (pedagogy-wise, mindset-wise, and just in general) with how it was being addressed before that. I had heard that some of the conclusions this came to were not valid due to errors? Or something like that? Regardless though, I think one can definitely take the sum of various research - including that even if slightly flawed, combined with one's own experience and the experience of others (when they actually give the whole picture and not just color it with perfect rosy colors), to determine whether there might be an opportunity for improvement (even if success already exists).