So I’m reading this article the other day.
“A comparison between teaching words through mere comprehensible input and through comprehensible input plus pushed output and corrective feedback was made.”
Naturally I’m intrigued. The authors were quite certain about the results: that input plus feedback and output made for better results than input alone. And this was published in a good journal.
And yet…when you got way, way down into the text, it turned out that the “input” he was talking about was…well, not what we consider to be “input”. It was language going in, I guess. But not connected language. (I guess that’s why it was “mere” comprehensible input?) Nope, here’s what this study considered to be comprehensible input: “The teacher would read each word aloud several times and provide students with appropriate definitions and examples. Then the participants were required to carry out some recognition fill-in-the blank exercises.”
And…the total sample size was 30. So, 15 for each group. Fewer than a typical American high school class. Anybody have two sections of the same course that behave in radically different ways?
And…the test of superiority was “passive acquisition of vocabulary”. Not connected language. Not proficiency.
And…well, that’s enough. I’m not ready to jump on the pushed output bandwagon based on that one, folks. If the authors have to say “because the time interval between the tests was rather long…” it ain’t acquisition, is it?