I’ve written this post, and the adapted research I link to below, so that I can refer people to it when discussing some of the issues I have with SER, as well as this, this and this.
Briefly, a great deal of SER uses data from school systems to try to identify ‘effective teachers’, and starts with questionable assumptions about the complicated relationship between teachers and teaching, on one hand, and pupils and learning on the other. Have read a great deal of this kind of research, I’ve come to the conclusion that what SER advocates call ‘teacher effectiveness’ can, using exactly the same data but fundamentally different assumptions, be called ‘class effects’.
I’m not in a position to do my own research in this area, but I’m happy to prove my point by adapting existing research. The first of which is available here, and is ‘Adapted Research’ (the original of which is here) which shows that ‘class effects’ exist, and that some classes have better outcomes than others. As soon as you realise that, with slightly different assumptions, nearly all studies of ‘teacher effectiveness’ can be thought of as studies of ‘class effects’, they become pretty banal and obvious. Some groups of children achieve more - or less - than others, for vastly complex reasons; and teachers are simply one (awesome) part of a hugely noisy process of child development.