Most educational innovations work. At least they do when they are first introduced. This generalisation may be because we never hear about the ones which are total disasters, but the positive aspect is likely to be because they are promoted by their advocates, and advocates are enthusiasts.
We have known since Rosenthal and Jacobson's Pygmalion in the Classroom (1968), whatever its faults, that education is a world of self-fulfilling prophecies. The initial proponents, for example, of the Initial Teaching Alphabet in the 1960s were sure that it would revolutionise the learning and teaching of reading. It was good enough to be widely adopted: but that was its downfall. When it was required of teachers who were not its advocates, the enthusiasm was lost, and it has been abandoned. An ERIC search yielded 286 sources in total, but only 10 of them since 1990, of which seven were by the same person (milking his PhD thesis, it appears!)
Note that of course teaching children to read is still a contentious area, as the current debate about the claimed results for "synthetic phonics" in the UK demonstrates: it's a bit off topic, but go here for the story, and here for the research report of the National Reading Panel in the US.
I have yet to read an action research study which concludes that the innovation was a total waste of time and effort. The same caveat applies as to the general innovations pointóbut there is also an alternative explanation:
The most significant (and unaccounted) variable in educational innovation is enthusiasm. Given enough of it, most things work, however wrong-headed. When teachers haven't got it, they might as well pack up and go home, regardless of the sophistication of their strategies, tactics and underlying theories. Period.