Making classroom assessment more accountable to scientific reasoning: A case for attending to mechanistic thinking


  • This work was completed at the University of Maryland, and the ideas expressed in this paper were refined through lively, often heated and contested, discussions among the authors and members of the University of Maryland Physics and Science Education Research Groups.

    Any opinions, findings, conclusions, or recommendations in this article are those of the authors.


When teachers or students assess the quality of ideas in science classes, they do so mostly based on textbook correctness; ideas are good to the extent they align with or lead to the content as presented in the textbook or curriculum. Such appeals to authority are at odds with the values and practices within the disciplines of science. There has been significant amount of attention to this mismatch in the science education research literature, primarily with respect to experimentation and argumentation as core disciplinary means of assessing ideas. In this article, we call attention to another aspect of scientific reasoning: a focus on causal mechanisms in explaining natural phenomena. We highlight examples and research from the history and philosophy of science to clarify what scientists mean by “mechanism” and to make the case for its centrality. We then present an excerpt from a second-grade class in which a student provides an incorrect mechanistic explanation, and the teacher gives priority to textbook correctness. As the conversation proceeds, the student shifts from mechanistic sensemaking to quoting terminology she does not understand. We argue that attention to mechanism in the classroom would better support student reasoning and better reflect disciplinary epistemology. © 2008 Wiley Periodicals, Inc. Sci Ed93:875–891, 2009