COS 8-2 - Assessing scientific reasoning in a liberal learning curriculum

Monday, August 8, 2011: 1:50 PM
9AB, Austin Convention Center
Sarah E. Jardeleza1, Diane Ebert-May1 and Megan Donahue2, (1)Plant Biology, Michigan State University, East Lansing, MI, (2)Physics-Astronomy, Michigan State University, East Lansing, MI
Background/Question/Methods

Michigan State University (MSU) embarked on a process designed to articulate a set of five liberal learning outcomes (LLO) and associated competencies for all graduates to achieve; one of which was scientific reasoning (SR). A first step in the process was to create specific objectives about SR for all students regardless of disciplinary major. Faculty input from seven colleges resulted in consensus about the major SR objectives and outcomes. Assuming that students’ mastery of scientific concepts and competencies is assessed differently in various science courses, we focused our efforts on the “reasoning” dimension of scientific literacy; that is, students’ understanding of the scientific process, independent of domain-specific scientific contents. MSU faculty designed assessment items that aligned with the objectives and learning outcomes for SR. We asked, how can SR data enable faculty to translate and align broad institutional learning goals into their own course objectives, instructional practices, and assessments? We began by analyzing incoming students’ performance on the SR instrument taken during freshman orientation.

Results/Conclusions

On average, incoming students answered 65% of the SR items correctly. The data show differences in performance on the SR instrument related to gender, ethnicity, and major. Findings include: 1) male students outperform female students, 2) students of Caucasian and Asian ethnicities outperform Hispanic and Black students, and 3) students in science, technology, engineering, and mathematics (STEM) majors outperform students in non-STEM majors. We also found that students perform at a high level (84% correct) on items that assess formulating hypotheses and designing experiments. However, students did not perform as well on items related to the other SR learning objectives (percent correct): describing methods of science inquiry (53%), making predictions and drawing conclusions based on data (69%), interpreting data (57%), solving real world quantitative problems (46%), evaluating the credibility, use and misuse of scientific information (57%), and illustrating the interdependence of science, social, and ethical issues (61%). As part of this longitudinal study, we will administer the instrument to students after completion of their general liberal education courses. These data are driving the professional development activities of faculty involved in non-majors science courses and guide curricular changes in liberal education at the institutional level.

Copyright © . All rights reserved.
Banner photo by Flickr user greg westfall.