Researchers in the Harvard Smithsonian Center for Astrophysics are developing and validating an assessment instrument that addresses the life sciences for students and teachers in grades 9 through 12 based on the Misconception Oriented Standards-based Assessment Resource for Teachers (MOSART). The project is developing 400 new test items that are based on core content domains for life science and are aligning these items with the previous National Science Education Standards to provide a connection to the earlier MOSART assessments. The project is also developing and validating two test instruments that address the cross cutting concepts of energy and matter for grades K-12, with a specific focus on flows, cycles and conservation. The new assessments will be made available to other researchers and practitioner through the project website and their on-line assessment system.
The assessment development is based on the process used in prior work that has produced the other MOSART instruments, including design efforts of assessment specialists, teachers, and learning scientists. Pilot items are tested through crowd-sourcing with online adult test takers. Classic test theory techniques, item response theory and Bayesian techniques model the student responses. Outcomes consist of item parameters, test and sub-test characteristics, and predictive linkages among items. A stratified, nationally representative sample of 250 high school biology teachers field test the items with classrooms of students. Descriptive statistics are generated to establish the state of student knowledge, pre-and post-test performance by item and by standard, and teacher knowledge, including the fraction of items for which teachers have correctly identified the most popular wrong answer. Descriptive analyses are followed by hierarchical linear modeling (HLM) of students within classrooms to examine the relationships between student and teacher knowledge. The dependent variables in HLM are student gain scores. Independent variables include teachers' knowledge, and student performance on grade K-8 assessments.
The MOSART instruments have been a strong line of assessment tools that are based on a model of cognition with a strong research base in misconceptions in science education. That research base is only slowly being augmented with a more coherent framework on learning progressions in STEM education, and the MOSART instruments will have the potential for extensive use for the foreseeable future. The grades 9-12 life science instrument based on coupling core ideas with science and engineering practices addresses the gaps in the current MOSART system of assessments. Given the rich literature on misconceptions in life science and the ubiquity of life science as a course at the high school level, the instrument promises to be as useful as the one for K-8 developed with MSP RETA funding. The new instruments on cross-cutting concepts provides a much needed set of assessments for researchers and practitioners, particularly teacher professional development providers. The transition to coupling core content and sciences practices with both the life sciences and the cross-cutting concepts is an opportunity to expand and update the suite of instruments.
