I'm not sure of the wisdom of this, but this article describes "how we can link educational datasets to theoretical constructs represented as causal models so formulating empirical tests of the educational theories that they represent." After all, "we might ask if the data being collected are fit for purpose." I get the idea - if you want to test your theory, you need a way to show whether and how data would confirm or disconfirm that theory. But I'm less sanguine about the utility of the theories themselves; why now just (as they say) follow the data? That said (and with more subtlety) "Arguments about the relative value of statistical-versus-theoretical modelling have arisen in many fields, under a wide range of guises, but fundamental to this debate is a difference in understanding about the purpose of our models. Are they for predicting system behaviour or for providing a causal understanding of the system being studied?" This paper is good grist for the mill, rich with examples, and deserving of a careful read and reply.
Today: 0 Total: 1433 [Direct link] [Share]
