Theory of Physics Education - a workshop for new physics TA's.

Goals for this workshop:

Workshop outline:

0) Meet the presenter, and each other

1) Introduction: Physics Education Research.

2) Focus is on student learning (not teaching!)

3) "Interactive Engagement" (construction of knowledge) vs "transmission of knowledge"

4) Student preconceptions - learning about them, and working with them.

5) Knowledge, practice, and understanding: Using research to guide classroom techniques

a) Frequent, informal assessment, meant to guide both teacher and student. Not for grades, but to help everyone understand where they are and what they need. Example: concept tests.

b) Group work. Benefits of social structure, motivation and engagement. The sum is greater than the parts. Example: tutorial worksheets.

c) "Context-rich" problems. Complex, realistic, open-ended, vague, motivated!

d) Classroom atmospheres that encourage exploration and discovery, motivating students. (Students have different learning styles.)

True or False: Good teachers are born, not made?

Although we all have different skills and talents, there is evidence that anyone can become a better teacher by learning about the research and experiences of other teachers. It's perhaps a bit like learning to play a piano - not everyone can become a virtuoso, but anyone can get better with a well-targeted effort, and learning a little theory will improve most people's skills faster than "just messing around".

1. PER: "Physics Education Research". Teaching can (and should) be a scholarly activity: there exists research on both theory and practice of teaching, and one can make use of it to improve skills and effectiveness as a teacher. Education experiments and theory are not as crisp or clear as research in pure physics (by a long shot!), but the best research is converging on basic concepts which influence good teaching. For example:

A. There is quantitative and qualitative evidence that "interactive engagement" techniques (involving student construction of knowledge, and active participation in the classrooms) have demonstrable advantages over pure "teaching by telling". See hake/

This is especially important at the intro level, when students haven't yet "learned how to learn".



(The figure shows a histogram of over 5000 students' learing gain "<g>". The red bars are from "traditional courses", including those with award winning popular teachers. The green bars are from "interactive engagement" courses. These results span a wide selection of student ages, educational background, course level, teacher quaility, etc. The test used is a "force concept inventory" which has been heavily used and studied for over a decade. Figure is from Hake, AJP 66: p. 64. (1998) Available at: )


B. The importance of students' starting state can't be overemphasized - they are not blank slates! Telling them "the truth" can fail at many levels - they may reject what you say, or compartmentalize it into "what I need for physics exams" rather than connecting it with their understanding of how the world works, One must be aware of preconceptions in order to effectively challenge, modify, and clarify student beliefs. This comes in part from learning what others have already discovered, but you can learn it for yourself by talking to students, asking them what they believe and why, both individually and through group questions, homeworks, and class discussions. The goal is not to "purge wrong ideas", but rather to help students connect what they DO know and understand to a more rigorous way of thinking about the world.


2. Focus on student learning, not on "teaching" per se:

Set your learning goals. Write them down. Tell your students! This is harder than it seems. Here are some things I have thought about in an introductory physics course. (You might have completely different goals for your students!) You might want them to build simple knowledge - factual information, and memorized formulas. You might want them to know the limits and applicability of those formulas. You might be interested in their grasp of fundamental concepts of physics. Can they connect the concepts to real world applications? Or perhaps it's skills you're after - solving algebra, translating word problems into math, doing logic puzzles. It might be "affect" - social values, attitudes about science, interest in the topic and a desire to learn more... You might want to exercise their creativity. Perhaps you want them to be able to extract and approximate a simple solution from complex/vague problems. Many physicists and engineers need improvement in their verbal or social skills. You might want to improve their critical thinking, or logical thinking, or ability to translate skills from one context to another. You might be after metacognitive skills - teaching them to "think about their thinking" - how do they know when they've got an answer right? How can they check that they understand a concept? How does anyone know if a theory is good? You might want them to practice and learn the scientific method - developing lab skills, making models, predicting, understanding errors. Or perhaps, all you want is for YOUR recitation section to get the best average scores on their midterm exams...

Align your instruction with your learning goals. How and what you teach will depend on what your goals are. Unless your sole goal is factual and procedural memorization, it is unlikely that 100% lecturing is appropriate. (And even then!)

There is no "best way": I cannot tell you how to teach, or what to do. Do what you like, students respond to enthusiasm! But, try to be informed about how you teach - just doing what worked for you is practically guaranteed not to be optimal. Remember, you are an exceptional individual, dedicated both to school and physics, with natural talents in math and logic and interests in learning not held (at least not in the same way) by many of your students. Almost none of the students in your class are physics majors, and fewer still will go on to grad school! They are (in general) not stupid, or lazy - but they may think and learn differently from how you do. You may have received and processed spoken information with ease, but many students function much better when they are required to make explicit sense of material under the guidance of a teaacher. Generally, "less is more" - most learning goals are achieved with depth of understanding, not breadth of coverage. (I'd rather "uncover" material than "cover" it!)

3. Interactive Engagement:

Student construction of knowledge has demonstrable advantages over pure "teaching by telling". (C.f. Hake data, shown earlier.) This is a subtle point. We are not asking our students to "discover" Newton's laws entirely on their own - they'd have to all be Newtons to do that! Your job is to guide them, to provide an environment in which they can question their own understanding, seek consistency, make sense of how the world works. If they can figure out some piece by themselves, they will understand and retain it far better than if you tell them. Interactive engagement is a philosophy of teaching, not a single technique. We will talk later about many different ways you can engage your students, make them think rather than be passive, get them interested and motivated. But this should always be in the back of your mind when you consider alternative teaching approaches.


4. Preconceptions, or "Alternative concepts":

One needs an acute awareness of the  students' starting state - one must be aware of preconceptions in order to effectively challenge and modify their beliefs. This is an essential ingredient in teaching introductory students!

(From Wandersee et al, 1994)

1) Students have a diverse set of beliefs coming into class

2) These beliefs cut across age, abilitiy, gender, and cutural boundaries.

3) They are highly resistant to change

4) They often match older beliefs of scientists and philosphers

5) They arise from observation, culture, language, and previous schooling.

6) They interact and interfere with "knowledge" presented in our classes, often resulting in unintended outcomes

7) Knowing about these concepts can help you address them, but you cannot eradicate them by "saying so"

8) You may have some "alternative concepts" yourself!

One posssible approach: (from the University of Washington group) Elicit students opinions to force them out into the open. Diagnose and clarify their conceptions. Then, confront them with the consequences of their beliefs. Create dissatisfaction, cognitive dissonance. Don't tell them it's wrong, think of ways to make them see it. Discussion and articulation is important here - let them argue about their ideas, just as scientists at the forefront of research do! It's important we don't leave them thinking they have terrible intuitions about the world - (they don't!) What you then need is the crucial last step - they need to resolve the conflicts on their own. They need to generate sense making, plausibility, connection to experiment, and connection to their old belief systems. Any time you tell them "the answer", they will happily file the information into a small compartment, but most students will then shut down - no more need to think, or make sense, since they "know" the answer. But will they be able to generalize it? Apply it in an even slightly different context? The answer is generally no, unless they constructed their own understanding.

5. Using research to guide classroom techniques:

A. Formative assessment: This means frequent low stakes questioning of the students. It also means that you take advantage of it, in the short term, to modify your instruction. It's used to find out what your students know, and to help THEM understand where they stand, what they need to work on. It's "formative" in that it helps form their education, and inform them about how they're doing. (It's not summative assessment like exams, where it's too late for them to do anything about it if they can't answer it!) It can be questions asked on homeworks, or quizzes, that count very little (or nothing), or on which they are allowed to seek help, talk, use their books. It can be the feedback YOU write on their assignments.

We use CONCEPT TESTS in the large lectures: multiple choice questions that focus on conceptual ideas - what depends on what? How does one variable scale with another? (If you double the speed of a car, how does its kinetic energy change? You'd be surprised how hard it is for some students to answer this - many don't see formulas as describing relationships, they merely see them as templates for calculator work) Concept tests should not involve number crunching or fancy algebra, in general - start by asking the most basic questions you think of! If they get it, make them more sophisticated. If not, it's a great clue to you about where to procede next. You can find concept tests at the end of the chapter of any standard freshman physics book - they'll have a ton of them! Or, go online on any of the old web pages of our large classes (e.g. /phys1110, or 1120, or 2010... You can find old ones by knowing the directory structure looks like e.g. .../physics/phys1110/phys1110_fa03 (fa for fall, sp for spring, sm for summer) , or Eric Mazur's web site ( galileo), or find books (like "Thinking Physics", by Epstein.) Ask instructors, or fellow grad students. Your time is limited - don't reinvent the tire (especially flat ones) - make use of other peoples teaching resources!

B. Collaborative learning: Why would we want students to work in teams? There are a huge number of reasons! It's one of the most significant results from education research, and has been intensively researched. Groups stimulate activity, humans are social animals. Articulating ideas helps clarify them: it's demonstrably true that one learns material the best when you have to teach it to someone else. (Use this last argument when a bright student tries to tell you they can do it on their own, and don't want to be "slowed down" by the group) "Talking physics" is an important skill not developed by numerical problem solving. Listening and comparing ideas gives students new/alternative perspectives. Conceptual understanding (often focussed on, in group activities) always assists in problem solving. Criticizing ideas helps solidify understanding, students must deal with and resolve inconsistencies. Groups can generally solve problems harder than any individual can (!) Small groups are much less intimidating and engaging than whole-class discussions, especially with larger groups. Group work (including leadership and organizational issues) is an essential life and business skill. There are many more arguments and evidence for the advantages of using group/collaborative learning, can you think of some?

Research argues that groups of 3-4 are best for complex problem solving. Pairs may not provide enough "physics knowledge". More than 4 means someone will often be left out. Diverse groups bring the most benefit - stronger students benefit by teaching, weaker students gain better support. Try to mix genders, but perhaps avoid 1 woman with 3 men. If the groups are disfunctional, assigning roles can help: "Manager" - in charge, keeps group on task, decides plan of action. Makes sure all members contribute. Watches time. "Recorder" - responsible for writing out what the group has found. Must ensure all members agree, and that everyone can also explain. Paraphrase others. "Skeptic" - responsible for understanding and questioning all group ideas. Explores other possibilities if needed/possible.

Groups need accountability - all members must be responsible for "reporting out". Otherwise, you start getting "hitchhikers" who do no work. Start off calling on enthusiastic students, but later you must call on all students, especially those who are not dominant. Alternatively, you can require everyone to write up their results on their own. Groups need training and support - it's hard to work together effectively! Encourage them, tell them some of the reasons why the group work is being used in class. Monitor groups - do not lecture! Avoid explaining anything until you've observed all groups at least briefly. Encourage participation, explain why groups help them learn. When you lead class discussions, keep it student centered. Let them argue, try to avoid "telling the answer" which shuts down thinking. (Ultimately, scientists need to decide for themselves what is right or wrong, no authority tells us!) Compare and contrast answers, do not criticize wrong ideas, but get them to focus on differences, and consistency. Avoid making anyone or any group feel stupid or resentful. "How are the different answers different? Are the differences important? What do the rest of you think about this?" If you can get them to convince themselves without relying on you as the final authority, they are learning well! You may need to ask students (orally or in writing, individually or collectively) what difficulties they have in working as a group, and how they could interact better in the future.


C. Realistic /"Context-rich" Problems: (Stolen from )

They must be just complex enough to require sophisticated strategies - we want to avoid "plug-n-chug" questions where the group would be a hindrence rather than a help. A good problem requires a systematic and structured approach (with diagramming, conceptual summary, "analysis of the problem", a procedure ("choose subproblems, collect basic relations") and in the end, checks and sense-making) It must still be simple enough that the solution, when found, can be understood and appreciated by the group. Each individual must be accountable for all the planning and "monitoring" skills practiced by the groups.

Real-world problems incorporate motivation, approximations, elimination of superfluous data. They may be open ended or ambiguous. They may involve multiple physics concepts. There may be need (or cause) for initial discussion. But you don't want all of these in any one problem - it's easy to make group problems too complex. Here are some guidelines to think about: Tick off any of the elements in the table below - if you have more than a half dozen of them checked off, the problem may be too challenging for an in-class problem.



Mathematical Solution

1. Cues Lacking

___ A. No target variable stated

___ B. Unfamiliar context (e.g. neutron star, quarks, lasers, ..)

2. Agility with Principles

___A. There is a choice of principles.

___B. Two or more principles are needed

___C. Principle are very abstract (e.g. flux, potential)

3. Non-standard application

___A. Atypical situation (not what you see in texts)

___B. Unusual choice of target variable. (e.g. what material should you pick?)

4. Excess or missing info.

___A. Excess data given

___B. Numbers are needed that aren't given

___C. Assumptions needed but not stated

5. Seemingly missing info

___A. Vague statement

___B. Special constraints (students must invent or deduce constraints)

___C. No diagram given, or students must use a diagram to extract information.

6. Additional Complexity

___A. >2 sub-parts

___B More than 5 variables

___C. Requires vectors

7. Algebra required

___A. No numbers given

___B. Unknown(s) cancel

___C. Simultaneous eqns

8. Targets math difficulty

___A. Calculus or vector alegbra needed

___B. Lengthy algebra (e.g. messy quadratic equation)

(Table stolen from the Minnesota group, http://groups.physics.umn. edu/physed)

When you make up problems like this, let yourself be guided by textbook problem. Then think about a motivation - can you change the problem to make it personal? Can you start it with "you..." (e.g. "You have a summer job doing..." , "you are watching.... and wonder", "Because you've had a physics class, a friend asks you to help them ...", "You are writing a story for your English class about ... and need to figure out...", "You have been hired as a technical advisor for a movie to make sure the science is correct. In the script.... but is that correct"? ) Do not choose problems that are solvable in one step, or that have tons of algebra, or can be solved best by a "trick".

But as usual - don't spend too much of your time reinventing education research! A great source of problems, organized by topic, can be found at Minnesota's physics web page, lineArchive/ola.html

(Their main page is VERY useful, and mostly aimed at graduate TA's. It's at http://groups.physics.umn. edu/physed/)


Example: You are flying into DIA when the pilot tells you that the plane cannot land immediately because of airport delays, and you will have to circle the airport. This is standard operating procedure. She also tells you that the plane will maintain a cruising speed of 400 mph at an altitude of 15,000 feet wile traveling in a horizontal circule around DIA. To pass the time, you decide to figure out how far you are from the airport. You notice that to circle, the pilot "banks" the plane so that the wings are oriented roughly 10 degrees from horizontal. An article in your in-flight magazine explains that an airplane can fly because the air exerts a force, called "lift", on the wings. The lift is always perpendicular to the wing surface. The magazine article also gives the weight of the 727 you're in as 100*10^3 pounds, and the length of each wing as 150 feet. It gives no information on the thrust from the engines or the drag on the airplane.

Questions you might ask yourself as a TA about this question: Does the question seem realistic? In what ways is this better than a typical end-of-chapter question? Does it encourage organized, logical problem solving strategy? Why? Is it tedious? Is it a "trick"? What concepts of physics does it involve? What procedures are required? What physics principles does it involve? Is there a way of "dividing the labor"? Is there any information missing? Anything superfluous? Is the "unknown" target variable explicitly named?

D. Classroom atmosphere: This is partly a very personal item - your own personality will deeply impact how you interact with students. But it's essential that you never talk down to students, or discourage anyone. No question is stupid!!! (Even the stupid ones :-) You may want to encourage mistakes - as long as they generate thoughtful discussions, they can be invaluable learning tools. Don't be afraid to make them yourself! NEVER bluff or try to pass yourself off as an expert on something you don't understand - think out loud, ask for help, or tell them you need to think about it some more. (Or ask them to think about it and explain it to you!) Every student in your class comes in with different ideas, beliefs, and attitudes. Respect this diversity, try to make use of it. It's helpful for everyone to hear different perspectives, and to learn to argue as physicists do (on substance, using logic, without personal attacks or emotional responses) That's right - it's a learnable skill! Different students have different learning styles - some are visual, some auditory, some tactile. Some like math, others prefer graphs, some want stories, some need analogies. What worked/works for you is NOT generally what they will need - so always try to begin interactions by asking questions and listening carefully. Try to get in their head - what do they need from you? Very rarely is it "the answer" - what they need is a push in the right direction, right for them, so that they can make sense of the problem at hand.

I asked at the start if teachers are made or born? The answer is complicated, and individual. You WILL get better as time goes by! You will make mistakes, have good and bad classes.

Enjoy the process - teaching can be the most rewarding and stimulating activity you can imagine (and also a frustrating, time consuming, and challenging one!) Bear in mind that it's all about the students, and their learning - not about you or your teaching! Follow your own instinct, while bearing in mind the experiences and research of other teachers. Have fun. You will learn more teaching than you ever did as a student!

Some references for further reading:

Start with our grad TA education page, http://www.

Then check out the CU Education web page, http://www.

For literature, you might start with:

L.C. McDermott, "How we teach and how students learn - a mismatch", AJP 61 (4), 295-98, (1993) homepage/jcannon/ejse/mcdermott.html

Hake: "Interactive-engagement vs. traditional methods: A 6000-student survey of mechanics test data for introductory physics courses," Am. J. Phys. 66, 64- 74 (1998) ( hake/, and / )

Redish: Excerpts from "The Physics Suite" (, and redish/Book/)

Mazur: Peer Instruction: A User's Manual (just read the starting chapter!)