Analyze Learning Systems
Using administrative, instructional, and learning outcome data, we analyze learning systems in order to measure and analyze the effectiveness of current teaching practices, and to find leverage points for improvement within a system.
Is Curved Grading a Valid Measure of Student Learning?
"Proponents argue curving accounts for changes in the difficulty of exams, guards against grade inflation, and is a tool for ranking students and evaluating potential for graduate school. Critics argue that curving grades does not provide a valid measure of the degree of content mastery. Despite the contentious debate over whether curving student grades is a valid strategy, little empirical research has examined this practice." --(Guzman-Alvarez, Smith & Molinaro, 2015)
Understanding the Curve: Implications of Norm-referenced Grading in Large Introductory Science Courses presents research by CEE's Alberto Guzman-Alvarez, Amy Falk Smith and Marco Molinaro on the impact of curved grading. Guzman-Alvarez, Smith and Molinaro did a statistical analysis of more than 16,000 students in 49 sections of an introductory chemistry class over 2008-2013. Students' chemistry course grades were modeled as functions of individual- and class-level characteristics. The analysis contrasted performance in on- and off-cycle courses, when students tend to have lower aggregated prior achievement (as measured by SAT and chemistry placement exam scores).
Results indicate that students' grades were associated not only with their own prior achievement, but also with that of their classmates. The authors conclude that curved grading is inequitable in that it artificially inflates or deflates students' grades depending on whether they take the course with a lower or higher performing cohort.
The poster received the 2015 Pat Rogers Poster Prize from the Society for Teaching and Learning in Higher Education (STLHE).
Can Open-Access Wikitexts Effectively Replace Costly Commercial Textbooks?
UC Davis faculty initiated the STEMWiki Hyperlibrary containing wikis for six STEM fields, to reduce costs of education by providing an open-access alternative to costly textbooks. ChemWiki is the most widely accessed, reaching more than 4 million students per month worldwide as of January 2015, and is used either to supplement textbooks or in lieu of them.
A 2014 pilot tested the efficacy of using ChemWiki as the primary resource for a large general chemistry class, against a similar control class using the regular commercial text. The pilot and its results are described in Chemwiki: Evaluating the Effectiveness of the Open-Access ChemWiki Resource as a Replacement for Traditional General Chemistry Textbooks in Chemistry Education Research and Practice, and in a 2015 Educause white paper, Assessing the Impact and Efficacy of the Open-Access ChemWiki Textbook Project, by UC Davis professors Gregory Allen and Delmar Larsen, and Alberto Guzman-Alvarez and Marco Molinaro of the CEE (then iAMSTEM).
The report concludes that “no statistical differences existed in either class’s performance… Each class performed equally well on the in-class exams, had similar normalized learning gains, and showed equivalent changes in thinking like a professional. The similarity indicates that student learning (as measured) is not affected by using the ChemWiki rather than a traditional textbook for one quarter of general chemistry at UC Davis.” While further work is needed to assess the viability of using the ChemWiki for the entire general chemistry series, “ongoing work shows promising results that can lead to cost savings for students and more timely responses in improving the quality and relevance of their reading materials.”