This work stems from the fact that objectively assessing student “mastery” of science concepts without truly understanding how they are making sense of these concepts, continues to be one of the most difficult tasks I face as an educator. A model eliciting activity (MEA) is an instructional tool that provides students and teachers with plenty of opportunities to express, test, and refine their thinking while simultaneously providing a document trail of thinking. Model eliciting activities allow teachers, students, and researchers to gain valuable information about how students construct, test, and revise models. Essentially, they are rich metacognitive tools that encourage students to express and refine their own thinking while simultaneously providing an opportunity for teachers and students themselves to gain insight on how their students are learning. However, two difficulties arise in the implementation of MEAs: (1) assessing the quality of the tasks involved in MEAs, and (2) assessing student knowledge demonstrated through MEAs (Wang et.al., 2009). This report reviews the literature on assessing MEAs and focuses on the development of a generalized assessment framework for model eliciting activities in a middle school science context. / text