Spending Time On Design: Does It Hurt Physics Learning? Eugenia Etkina, Alan Van Heuvelen, Anna Karelina, Maria Ruibal-Villasenor, and David Rosengrant Graduate School of Education, Rutgers University, New Brunswick, NJ 08901 Abstract. This paper is the first in a series of three describing a controlled study “Transfer of scientific abilities”. The study was conducted in a large enrollment student introductory physics course taught via Investigative Science Learning Environment. Its goal was to find whether designing their own experiments in labs affects students’ approaches to experimental problem solving in new areas of physics and in biology, and their learning of physics concepts. The theoretical framework for the design of the study was based on transfer theories such as “preparation for future learning”, “actor-oriented transfer”, “transfer of situated learning’’ and “coordination classes”. In this paper we describe the design of the study and present data concerning the performance of experimental and control groups on multiple-choice and open-ended exam questions and on the lab exams that assess student understanding of the physics and the reasoning processes used in the lab experiments. We found that the experimental group outperformed the control on lab-based and traditional exams and the difference increased as the year progressed. The project was supported by NSF grant DRL 0241078. Keywords: Reformed instruction, design labs. PACS: 01.40Fk; 0140.gb; 01.50Qb. INTRODUCTION This manuscript is the first of three papers that describe an experimental design study in an introductory course whose goal was to investigate the effects of design labs on student learning of physics and their acquisition and transfer of scientific abilities. Scientific abilities are approaches and procedures that scientists use when engaging in the construction of knowledge or in solving complex problems [1]. In this study all students enrolled in the course attended the same large room meetings and recitations that followed the ISLE curriculum [2]. However, they were randomly split into two groups in the labs. The experimental group designed their own experiments. Their scaffolding included questions concerning the scientific abilities needed for the ISLE design labs [3] and on self-assessment rubrics [1]. In the control group students performed the same experiments but with the design provided in the write-up and supported by conceptual questions that helped students work through the physics. Throughout the semester the groups were compared on their physics learning. At the end all students performed two lab transfer tasks: one to design an experiment to investigate a physics problem in new area of physics and the other an experiment in biology. This paper describes the part of the study related to the comparison of the groups on the paper-and-pencil exam problems and lab-based problems. The research question in this paper is: if students in the labs focus on designing their own experiments without having the “right answer” and on the elements of the scientific investigation instead of on solving physics problems, do they learn less physics than those who have a good experimental design provided for them and more opportunities to engage in physics problem solving? Two other submitted papers describe the aspects of the project related to transfer in the physics and biology content. MOTIVATION There are two big motivations for this study: (1) Recent reports concerning science and engineering education encourage student acquisition of conceptual and quantitative understanding of physics principles and also the acquisition of abilities to: design their own experiments, reason from the data, construct explanatory models, solve complex problems, work with other people, and communicate [4-7]. Should we spend time on the development of these latter abilities 88Downloaded 07 Nov 2008 to 128.138.145.135. Redistribution subject to ASCE license or copyright; see http://proceedings.aip.org/proceedings/cpcr.jspCREDIT LINE (BELOW) TO BE INSERTED ON THE FIRST PAGE OF EACH PAPERCP951, 2007 Physics Education Research Conference, edited by L. Hsu, C. Henderson, and L. McCullough© 2007 American Institute of Physics 978-0-7354-0465-6/07/$23.00or this will harm students’ acquisition of physics conceptual learning and ability to solve traditional problems? (2) Many experiments indicate that the ability to transfer what is learned in physics to other unstudied physics areas, to other academic disciplines, and to work after academia is lacking. Can students transfer what they learn in our physics design labs to other unstudied areas of physics and to other academic disciplines—the subject of our other two papers? THEORETICAL FOUNDATION Transfer As this study is a part of a larger “transfer” study, we briefly describe the theoretical perspectives that informed the design of the whole project. When designing the learning environment for the experimental group, we carefully followed the recommendations of the literature on how to create a learning environment that promotes transfer. As mentioned above, the purpose of the whole project was to determine if students in design labs are able to transfer the scientific abilities that they learned during one semester into new physics content and into biology. In other words, to find whether they apply the habits of mind learned in the labs when they face a new problem for which they do not have content knowledge or experimental skills. Transfer refers to the ability to apply knowledge, skills, and representations to new contexts and problems [8-10]. Research shows that achieving transfer is difficult [11]. However, new work of Lobato shows that transfer occurs often and the problem is in its recognition by researchers, not its existence [12]. There are several theoretical models of transfer [13, 14, 12]. The most relevant to this study are direct applications transfer,