What Else (Besides the Syllabus) Should Students Learn in Introductory Physics?
Recently our education research group has turned its attention from studying micro-learning (see relate.mit.edu) to issues that lie at the heart of any educational reform – even though they are often overshadowed by discussions of what the syllabus topics should include. The questions we’re addressing include:
This contribution is based on a survey we administered to ~ 600 teachers and students concerning the first question – what students should learn in introductory physics. Since our survey was designed to emphasize student habits, pedagogical objectives, and overall student skills, it will be relevant to faculty in all General Institute Requirement (GIR) courses.
This paper represents an attempt to shift the course reform discussion in introductory physics – and hopefully other subjects – to instructional goals rather than teaching techniques and syllabus topics.
This is imperative, because numerous learning goals have recently come under contemporary discussion in physics education research including cognitive abilities, scientific abilities, and habits of mind (e.g., demonstrating problem-solving skills by initially developing a qualitative description of the problem).
To elicit non-topical learning goals for introductory physics we used a Delphi study approach, starting by asking about 20 successful instructors – mostly from the Physics Education Research (PER) or American Association of Physics Teachers (AAPT) communities – to suggest such goals in their own words. From their responses we distilled 12 alternatives. Then we polled successive groups of instructors, using the question:
“Due to a change in the academic calendar, you have 20% more time to teach the calculus-based introductory course to non-physics majors, and the syllabus has not been expanded. What learning will you seek to add or emphasize with this extra time?”
The respondents were asked to vote for two of the 12 alternatives, which were grouped into four categories:
1. Course Content
2. Instructional Themes
3. Problem Solving
4. Relation to the Outside World
It should be noted that this approach leaves two major questions out of the discussion: Should we adopt a more modern approach (e.g., basing discussion of matter and interactions on an atomic viewpoint), and should we assess students more broadly than by their ability to solve problems (e.g., a term paper or project)?
We received 708 responses from instructors representing different groups: educators at AAPT meetings, atomic researchers at a Gordon Research Conference, and education researchers at a physics education research meeting. These three groups agreed on some topics, but also disagreed substantially on others. We also asked several groups of students what they wanted to learn, given the same alternatives, receiving a total of 562 responses. Students in different institutions were in reasonable accord, but their preferences generally anti-correlated with those of their instructors. The percentage of positive responses on each of the alternatives is presented in Figures 1 and 2. The dashed portion of the bars corresponds to the 95% confidence intervals for proportions calculated using a normal distribution. The scientific method and vocabulary of subject domain were unpopular (average under 2%) with the three groups of instructors and with students as well; therefore, they are not included in the figures.
Similarities and differences between instructor groups
The most striking fact about instructor preferences in Figure 1 is that there is no “must do” selection. Sense-making of an answer was the instructors’ top choice (17% of the votes representing ~34% of the teachers, since they could vote twice). All instructor groups showed about average preference for both laboratories (10%) and understanding science in news and society (10%), and a disdain for wider content (4%).
The most notable difference among instructors was on problem solving – the combination of vocabulary, concepts, and plan set-up. Educators selected problem solving (excluding sense-making) at 39%, more than atomic researchers (16%) and even more than education researchers (7%). For education researchers, epistemology (17%) generally applies to the construction of individual students’ knowledge (e.g., whether the student thinks problem solutions are obtained by applying memorized formulae rather than thinking about the concepts), and a good fraction of epistemological effort is aimed at better problem solving. Counting most of the “epistemology” responses as problem solving, responses puts Education Researchers near the average of all instructors in this category. Education researchers thought “scientific argument” (15%) was more important than the other two instructor groups (average of 5.5%). Atomic researchers rated “physics from a few ideas” (17%) as their top selection, while educators were less enthusiastic (6%).
Similarities and differences between instructors and students
The substantial difference between the preferences of students and instructors (average of educators, atomic researchers, and education researchers) is shown in Figure 2. Wider content was students' top preference but instructors’ lowest (19% vs. 4%). The relation of physics to everyday life/things was students’ second preference but the instructors’ second lowest (15% vs. 6%). On the other hand, students had no interest (3%) in sense-making, which was the instructors’ top selection (17%). Students had little interest in scientific argument (2%) whereas instructors thought it merited significant attention (9%). Students and instructors agreed on priorities of five of the 10 options – physics from a few ideas, epistemology, concepts, plan set-up of problems, and understanding of science in news and society. However, the differences on the other selections were so marked that the correlation between students’ and teachers’ preference is - 0.4. In other words, the students’ interests are more than orthogonal to their teachers’ – they are 115 degrees apart!
Implications for Course Reform at MIT
What do these findings imply about course reform for introductory physics? The most significant finding is that students’ preferences anti-correlate with ours. We instructors seem to be saying, “We are going to make you into expert physicists,” and the freshmen seem to be replying, “Before we commit to that much hard work, tell us how physics connects to the world around us and to society’s problems, and teach us new things we haven’t studied before.”
Freshmen don’t see the relevance of introductory physics to their lives, as evidenced by a survey we took in 8.02 where, by nearly 2:1, they characterized their prime motivation in 8.01 and 8.02 as “goal-oriented” [to pass the requirement] rather than “mastery learning” [of the subject]. In contrast, they indicated, also by ~ 2:1, that they’d choose to learn the subject matter of their major but receive no degree rather than get a degree that was accompanied by little learning.
Arguably, students don’t realize that introductory physics underlies a large majority of their likely majors. Many students view introductory physics primarily as a required hurdle and focus exclusively on how to do problems like those on exams, rather than focusing on any other aspect of the subject, such as epistemology, history, and relevance to their major or to their everyday observations or reading.
To get students’ engaged attention we should demonstrate the relevance and utility of physics to their lives and careers. We can do this without sacrificing a great deal of course time, simply by selecting illustrative examples involving “physics applied to everyday life/things” and “science in the news and society.” Our moving mass problems could involve topics such as air resistance and how airplanes fly – rather than artificial examples like a railroad car slowing because it is filling with rain – and our treatment of mechanical energy could mention the energy crisis. We could use examples that illustrate the application of basic physics in other disciplines at MIT as well, and in current research.
Demonstrating the relevance of introductory physics would address two of the students’ four top preferences, but not their top preference – new topics. Unfortunately, research literature and concept tests show that we already sacrifice basic conceptual understanding by covering too many topics. Thus a better approach might be to add sufficient real world and societal relevance to existing topics (certainly not part of what our students learned in high school) so our students find revitalized interest.
At MIT, teaching students to become critical thinkers is one of our professed general educational goals. A foundational skill for critical thinking is to “make sense of an answer (includes estimation) using units, special or limiting cases, symmetry, etc.” – the top instructor choice in our survey. In honesty, we (and most professors nationwide) do not regularly make sense of our example solutions; we employ a partial credit grading system that often awards significant credit for physically ridiculous answers, and only sometimes do we take a point off if the answer is dimensionally incorrect.
Both students and instructors want students to become more expert problem solvers, although students don’t see the value of making sense of their answers. (According to research, novice problem solvers put their faith in procedures rather than concepts.) Currently introductory physics is primarily oriented toward problem solving, especially in TEAL, and clearly achieves success – students exiting 8.01 show a learning effect in excess of two standard deviations on the problems requiring an analytic answer (versus just over one on conceptual questions). However, typical physics problems represent a narrow slice of possible assessment tasks. They give only necessary and sufficient information, use standard notation for the various physical quantities, give the approximations, and (at MIT) rarely involve numbers. A consideration of possible consequences, such as not assessing “thinking like a physicist,” encouraging novice problem solving techniques, or failing to give a quantitative view of the world is important in a discussion of course reform, but lies outside the scope of this article.
Our results, like any exploratory study, raise new questions. In addition to the professors and the students, the educational goals of the Institute at large, the Physics Department, and the instructors of courses requiring 8.01 and 8.02 have priorities that must be considered in any comprehensive discussion of what to teach. Also there might be other groups of instructors with different preferences; or student priorities may change over time. (Results from College of DePaul indicate that students lose interest in studying more topics and become much more interested in problem solving by the end of the semester. MIT seniors highly value 8.01 and 8.02 for teaching them problem solving.)
These refinements aside, our results show a robust mismatch between our teaching and what students are interested in learning; moreover they show that we physics teachers ignore our own top preference – making sense of the answer. We really should consider addressing such disparities when we reform our GIRs.
Future articles in this series will address what students actually learn in 8.01 and how much of it they still retain at graduation.
We are grateful to E. Cornell, C. Monroe, M. Sabella, and J. Thompson for giving us time to survey the attendees at their conferences, and to T. Carter and D. Demaree for administering the survey to their classes. We are grateful for support from NSF grant PHY-0757931.
Note: This work is part of a study for national publication that will hopefully affect physics discussions nationally, The views are those of the authors and not those of the Physics Department.