Bayesian network models can not only provide belief estimates for proficiencies and claims about the learner, but they can also be used to explain the basis of those estimates. The quantity defined as the weight of evidence associated with a task is useful for explanation, as well as debugging models (explaining unexpected results). Expected weight of evidence is useful for assembling assessments either adaptively or in fixed forms.
CITATION STYLE
Almond, R. G., Mislevy, R. J., Steinberg, L. S., Yan, D., & Williamson, D. M. (2015). Explanation and Test Construction (pp. 197–237). https://doi.org/10.1007/978-1-4939-2125-6_7
Mendeley helps you to discover research relevant for your work.