Protocols Once More
1986; National Council of Teachers of English; Volume: 48; Issue: 7 Linguagem: Inglês
10.2307/377377
ISSN2161-8178
Autores Tópico(s)Information Systems Theories and Implementation
ResumoEach side in the debate over protocols claims that the other doesn't fight fair. Erwin Steinberg complains that the critics don't agree on their criticisms. The critics complain that the descriptions of protocol analysis are too slippery to get hold of. Erwin Steinberg has helped, because he has made strong claims for protocol analysis. Still, I think, both sides have missed an essential fact. Any form of empirical investigation presupposes a model of the phenomena being investigated. The utility of the empirical method depends entirely on the cogency, accuracy, and validity of the model. When, for instance, people believed that a heavy object falls faster than a light object, they had a model of how objects behaved. With this model, experiments with feathers and stones, as Thomas Kuhn has said, made perfect sense. When another model came along, these experiments stopped making sense. If you don't realize this, then Steinberg's reply must seem vapid. He would be construed as saying merely that protocol analysis is useful. Who could deny that? Scardemalia and Bereiter's findings alone could help a teacher restructure assignments so that parroting becomes far more difficult. If you do realize this, however, you must see that Steinberg hasn't described his model. (Perhaps it would have helped to give us a transcript of what he was thinking while he was trying to express his thoughts.) In this article, therefore, I'm not going to reply directly to him. Instead, I'm going to describe this model in some detail and thereby clarify the case he ought (in my view) to be making. I will then criticize that case. If I do nothing more, I will at least describe the actual issues involved, and that might give the debate some clarity and decrease its rancor.
Referência(s)