home - login - register

PERC 2019 Abstract Detail Page

Previous Page  |  New Search  |  Browse All

Abstract Title: Understanding and assessing problem-solving in introductory physics
Abstract: Expert-like critical thinking and problem-solving skills are often lauded as outcomes of an undergraduate STEM education, and are highly sought-after by potential employers of physics graduates. While many heuristics for problem-solving and some research on expert problem-solving exists, there is little evidence showing that we are training students to become good problem-solvers or critical thinkers. Moreover, previous work on expert versus novice problem-solving has two important limitations: 1) The experts used are often graduate students, and may not be fully-developed experts in their field. 2) The "problems" used in these studies are not authentic--for the true expert these are merely exercises that one might find in a normal introductory textbook. In this session we will discuss recent work to advance our knowledge of how experts solve problems, of how to construct authentic problems and improved assessments of problem solving in students, of how students interact with complex problems, and of instructional practices to help students improve their problem-solving skills. Current work identifying how experts--including STEM faculty and industry experts--think and solve problems they encounter in their work has shown that experts across disciplines make a remarkably consistent set of decisions when solving problems. Based on these findings, we discuss work on designing assessments of the many skills used when solving complex problems, and current work on how students interact with complex problems. The ultimate goal of these assessments is to improve instruction by providing feedback to instructors and undergraduate programs about whether their students are learning expert-like critical thinking and problem-solving skills. We will also discuss methods to teach expert-like problem solving skills in the classroom, which emphasize students practicing making the same decisions that experts make when solving problems along with formative feedback from the instructors.
Abstract Type: Talk Symposium
Session Time: Parallel Sessions Cluster III
Room: Cascade A

Author/Organizer Information

Primary Contact: Eric Burkholder
Standford University
and Co-Presenter(s)
Argenta Price

Symposium Specific Information

Presentation 1 Title: Identifying expert problem solving decisions
Presentation 1 Authors: Argenta Price
Presentation 1 Abstract: Multiple frameworks, from the "scientific method" to Chi's work on problem solving by experts vs. novices, have been used to describe how scientists solve problems. We add to this work by developing a framework, based on empirical evidence from experts solving authentic problems in their work, of detailed problem solving decisions that experts make when solving problems. We developed an initial list of decisions from interviews with faculty members and industry experts, then conducted a second, structured, set of interviews to test and refine the list. In the structured interviews we asked scientists, engineers, and doctors to think about a problem they solved in their work and walk us through the detailed decisions they made during the solution process. We were struck by how, at a high level, the decisions made by experts across STEMM fields were remarkably similar. From these interviews, we generated a list of around 40 decisions that most STEMM experts make when solving authentic problems. This list of decisions has implications for teaching and assessment, because in order to become expert-like problem solvers, students will need to practice making the decisions that experts make.
Presentation 2 Title: Assessing adaptive expertise in undergraduate engineering curricula
Presentation 2 Authors: Eric Burkholder
Presentation 2 Abstract: Engineering programs frequently claim that they teach undergraduate students how to be good problem solvers; however, there has been no research to-date that demonstrates this, in no small part due to the fact that measuring problem-solving is quite difficult. We characterize problem-solving skills exhibited by experts as elements of "adaptive expertise," which includes critical thinking, problem-solving in novel contexts, and the ability to learn and adapt to new situations. We develop an instrument in the context of chemical process design that aims to assess elements of adaptive expertise, and how closely student thinking aligns with this framework. Preliminary investigations reveal that some students lack particular problem-solving skills; most salient is the apparent absence of a predictive framework when solving a problem. This lack of a framework leads students to make fatal errors in problem-solving, including some that they receive explicit instruction in during introductory courses. This research shows that problem-solving may not be a guaranteed byproduct of an undergraduate STEM education.
Presentation 3 Title: Development of a Practical Problem Solving Assessment Tool
Presentation 3 Authors: Wendy K. Adams and Adria Brown
Presentation 3 Abstract: Motivated by an interest to better understand student's strengths and weaknesses in problem solving and the desire to compare instruction across institution, we developed the CAPSS – the Colorado Assessment of Problem Solving. While developing the instrument, we empirically identified over 40 distinct sub-skills that affect a person's ability to solve complex problems in many different contexts. The original CAPSS is an open-ended instrument that requires either a two hour interview or, if completed as a paper and pencil assessment, an hour to grade by a trained evaluator. Therefore it is not a practical tool for an instructor to use in their teaching.  Over the past year, we have been developing a forced-answer version of CAPSS with the goal of efficiently scoring as many sub-skills as possible. We will share progress to date as well as future plans for further development.
Presentation 4 Title: Building up to complexity: synthesizing multiple concepts to solve problems
Presentation 4 Authors: Andrew Heckler
Presentation 4 Abstract: Since complex, authentic problems inevitably involve multiple concepts, it can be productive to investigate student solution processes for relatively simple "synthesis problems" that involve a small number of concepts. For example, we find that even simple synthesis problems require qualitatively different solution methods compared to single-concept problems that are commonly used in physics courses. Through a series of studies, our team has found  a number of important factors that influence the problem solving process and can interfere with recognition of the need for multiple concepts and hinder their joint application once the relevant concept are identified. These factors can include the mathematical complexity of the solution process, whether the relevant physical phenomena are temporally simultaneous or sequential, and whether there are large differences in the cognitive availability of the relevant concepts. Further, we find that employing carefully designed worked examples and guiding students through self-explanations and analogical comparisons of the structure can significantly improve performance.  Finally, we hypothesize how these insights might be used as tools to build skills for solving complex, real world problems.
Presentation 5 Title: Improving Problem-solving Through Reflective Training
Presentation 5 Authors: Shima Salehi
Presentation 5 Abstract: Effective science and engineering education goes beyond teaching content knowledge, and encompasses training problem-solvers who can use their knowledge in practice to solve complex problems. While over the past decade, the science and engineering education
community (NRC 2012, NGSS) has acknowledged the significance of training good scientific
problem-solvers, there remain essential questions to be addressed: "What are the characteristics of good scientific problem-solving?", and "how should we teach scientific problem-solving?" In a series of studies, we have addressed these two questions. In our previous works, we have identified major problem-solving practices used in solving complex problems. Based on the identified practices, we have designed a problem-solving training program, called reflective problem-solving. In this talk, we will describe the reflective training, and show that: 1) this training indeed improved students' problem-solving; 2) this improvement transcended specific content area, and students' problem-solving improved in a content area different than the content area of their training; and 3) the improvement resulted from the reflective training exceeded the improvement resulted from the standard educational setting of attempting multiple problems and receiving feedback only about the accuracy of their answers.