When validating assessment models built with data mining, generalization is typically tested at the student-level, where models are tested on new students. This approach, though, may fail to find cases where model performance suffers if other aspects of those cases relevant to prediction are not well represented. We explore this here by testing if…

The authors present work towards automatically assessing data collection behaviors as middle school students engage in inquiry within a physics microworld. In this study, the authors used machine learned models that can detect when students test their articulated hypotheses, design controlled experiments, and engage in planning behaviors using…

We present "Science Assistments," an interactive environment, which assesses students' inquiry skills as they engage in inquiry using science microworlds. We frame our variables, tasks, assessments, and methods of analyzing data in terms of "evidence-centered design." Specifically, we focus on the "student model," the…

Data-mined models often achieve good predictive power, but sometimes at the cost of interpretability. We investigate here if selecting features to increase a model's construct validity and interpretability also can improve the model's ability to predict the desired constructs. We do this by taking existing models and reducing the feature set to…

We present a method for assessing science inquiry performance, specifically for the inquiry skill of designing and conducting experiments, using educational data mining on students' log data from online microworlds in the Inq-ITS system (Inquiry Intelligent Tutoring System; www.inq-its.org). In our approach, we use a 2-step process: First we use…