Discovering Prerequisite Relationships Among Knowledge Components.
Richard Scheines, Elizabeth Silver, Ilya M. Goldin · 2014
Knowing the prerequisite structure among the knowledge components in a domain is crucial for instruction and assessment. Treating Knowledge Components as latent variables, we investigate how data on the items that test these KCs can be used to discover the prerequisite structure among the KCs. By modeling the pre-requisite relations as a causal graph, we can then search for the causal structure among the latents via an extension of an algorithm introduced by Spirtes, Glymour, and Scheines in 2000. We validate the algorithm using simulated data.