Beyond problem solving: Evaluating introductory physics courses through the hidden curriculum

Abstract

A large number of innovative approaches have been developed based on Physics Education Research (PER) to address student difficulties introductory physics instruction. Yet, there are currently few widely accepted assessment methods for determining the effectiveness of these methods. This dissertation compares the effectiveness of traditional calculus-based instruction with University of Washington's Tutorials, University of Minnesota's Group Problem Solving & Problem Solving Labs, and Dickinson College's Workshop Physics. Implementation of these curricula were studied at ten undergraduate institutions. The research methods used include the Force Concept Inventory (FCI), the Maryland Physics Expectation (MPEX) survey, specially designed exam problems, and interviews with student volunteers. The MPEX survey is a new diagnostic instrument developed specifically for this study. Instructors often have learning goals for their students that go beyond having them demonstrate mastery of physics through typical end-of-chapter problems on exams and homeworks. Because these goals are often not stated explicitly nor adequately reinforced through grading and testing, we refer to this kind of learning goal as part of the course's hidden curriculum. In this study, we evaluate two aspects of student learning from this hidden curriculum in the introductory physics sequence: conceptual understanding and expectations (cognitive beliefs that affect how students think about and learn physics). We find two main results. First, the exam problems and the pre/post FCI results on students conceptual understanding showed that the three research-based curricula were more effective than traditional instruction for helping students learn velocity graphs, Newtonian concepts of force and motion, harmonic oscillator motion, and interference. Second, although the distribution of students' expectations vary for different student populations, the overall distributions differ considerably from what expert physics instructors would like them to have and differ even more by the end of the first year. Only students from two of the research-based sequences showed any improvement in their expectations.