…But Modeling Instruction Works Better

In my last post, I suggested that one of the ways to approach the subject of curricular and pedagogical change with colleagues, especially as it pertains to modeling instruction, is to share one’s own personal story of transformation. So here’s mine.

I used to be an awesome teacher. Seriously, that’s what everyone used to tell me and I had started to believe them. For the first six years of my career, I did my very best to make my classes dynamic and engaging by incorporating various types of activities. During lecture, I was enthusiastic and worked hard to distill an idea down to its essence when explaining it. My reviews from students were always glowing and the administration loved what I was doing. And the awards! In those six years, I taught at a rural public school, was a TA at a state university and spent the final year of the first six at my current school. In that time, I won three teaching awards and was nominated for two others. By every measure I had available, I was a great teacher. And then, the FCI came along.

The FCI (or Force Concept Inventory) is 30-question, conceptual multiple choice test meant to determine how well the students in your class understand the Newtonian concept of forces and the results of their application. No calculations are involved. The FCI came to me through Dr. Kathy Harper at OSU. She had presented a short, one-day workshop on modeling instruction locally and had followed up with flyers inviting the participants to the summer workshop. I signed up and as part of the process, I was asked to give my students the FCI before coming to the workshop. Looking over the test, I thought my students would make short work of it. After all, we were ending the year on magnetic induction, so surely they understood Newton’s laws of motion. We’d been using them all year. With the first year at my new school ending, I administered the FCI to my students. When I reached the summer workshop, I eagerly sought out the results.

College Prep Physics – Mean = 6/30 (20%)

Honors Physics – Mean = 9/30 (30%)

AP Physics C:Mechanics – Mean = 13/30 (43%)

WTF?! How were those scores possible? I spent a year teaching those students physics. The FCI covers some of the most fundamental ideas I told them about. They got As and Bs! The AP scores were all 4s and 5s! While I didn’t expect perfect scores, I did expect results significantly better than guessing. Those FCI scores were like a swift, spiritual kick to the head that altered my reality forever. I had to figure out what was going on. I began by doing what every normal person would do in this case – I blamed the test.

When I started the workshop, my skepticism for modeling instruction was high – very high. Of course, kids taught with modeling do better on the FCI – it’s written by the people that developed modeling! How does it relate to other measures of physics learning? (It correlates very strongly.) How many students do we have data on? (At this point, n> 10,000.) Students still learn by lecture. Why bother changing? (Because more of your students will learn more material using modeling.) All of my questions had answers. The research on modeling instruction and the tools used to evaluate it is extensive. (Here’s a start.) As a science teacher, if I’m going to talk the talk, I need to walk the walk. Having been confronted with what I found to be convincing evidence, I was obligated to investigate this pedagogy and see if it had an effect on how well my students understood physics. After the workshop, I returned home, prepared for the coming year and did my best to implement modeling in the following months.

It was difficult. I felt like a first year teacher again. Rushing home to review and prepare for the coming day, I was forced to truly change the way students engaged with the material and forego much of what I had developed over the past six years. Too many times, I’d start to revert to simply explaining some idea rather than forcing the students to provide the explanation. As the year progressed, it became even more difficult as we started covering content that the workshop never got to. I’d never been good at writing assessments and my old ones weren’t addressing the things I was now asking about. It was a mess. But there was a noticeable shift in how my students were acting in class. They began to enjoy the content rather than the show I used to put on. Their confidence in their knowledge was much greater than it had been in the past. But, to my critical eye, those things, while very important, were not measures of how much physics they’d learned. Was this admittedly chaotic year more productive than the way things had been in the past? For that comparison, I needed the FCI.

At the end of that first year of modeling, I only gave it to my honors physics class. I hadn’t taught college prep that year and hadn’t used modeling in AP. Nervously, I scored the tests and tabulated the results:

Honors Physics – Mean = 18/30 (60%)

I stared at that result for a long time. Even in the face of the frankly terrible job I felt I had done that year, my students had shown incredible learning gains. The next two years showed even greater improvements on the FCI as I refined and adapted modeling instruction to my own style of teaching. The pre- and post- scores remained consistent with the national averages. Additionally, enrollment in the 2nd year AP Physics class has grown from 4 to 13. AP scores remained high. And the unmeasureable qualities I mentioned earlier continued to grow in both intensity and frequency. From those results, I had to conclude that modeling instruction was more effective than lectures at producing lasting learning and getting students to adopt a Newtonian view of the universe. That seems like the only logical conclusion to draw. It would be irresponsible of me to continue to teach the way I had been when confronted with this information. So, I started using modeling and have not gone back. That’s my story.

I’ll leave you with this analogy. Newton’s law of universal gravitation works. We used it to discover Neptune and Pluto, both invisible to the naked eye. It predicts the paths of comets and dates of eclipses. We could continue to use it, but then we’d miss out on solving the mystery of Mercury’s perihelion, the bending of starlight, GPS, black holes, the origin of the universe and more. If you know about general relativity, but refuse to use it, then all of those problems remain unsolved. I refuse to leave any tool unused that will demonstrably improve the knowledge and experience of my students.



  1. Hi there, one of my concerns with the modeling curriculum is how much time it takes. I buy into the claims the students learn and understand more from the material covered using modeling, but I’m really concerned about running out of time for topics that certain students find relevant and useful (waves, thermo, fluids, nuclear). I feel like I’d rather have my kids learn/understand a little less in the mechanics realm in order to get to experience the other branches of Physics of introductory Physics. Those kids who are scientifically inclined will have many other opportunities to delve further into mechanics in second year courses and beyond, but for many kids this is the only exposure they’ll ever get, and I’d hate for them not to learn what causes a rainbow. Your thoughts?

    1. Hi Kate,

      Your concerns are valid. Modeling instruction definitely takes more time than transmissional teaching methods. I struggled with this at first and kept trying to cover all the content I used to. After all, it wasn’t boxes on inclined planes that got me interested in physics. That honor belongs to relativity, black holes, Schrodinger’s cat and other weird ideas that you might get to at the end of the year. I wrestled with the breadth vs. depth question for a while and I’m still looking for ways to include modern physics. Obviously, my conclusion was to emphasize fewer topics to help students develop the skills needed to be able to later learn any topic in physics that interested them. This is a personal decision that each teacher has to decide for themselves as it is greatly dependent not only on personal preferences and teaching style, but also what type of community the class is serving, future testing, and college preparedness. But, if you want to deemphasize mechanics to make room for other subfields of physics, there are a few things that can help.

      First, I have noticed that as students become familiar with the concept of what a model is and the structure of the modeling cycle, the pace of the class picks up. Early units seem to take an inordinate amount of time, while later units move along at a good clip. By November, you’re no longer reminding them to whiteboard last night’s homework at the beginning of class, because they’re already doing it when you walk in the room. My first units run 3 to 3 1/2 weeks, but by the end of the year, a unit is barely taking two weeks. I’m planning for 12 units next year in my honors physics class with the option to add on at the end if there is time.

      Additionally, modeling instruction isn’t a curriculum. It’s a structure to learning that you can hang a curriculum on. I’ve read about modelers that start the year with geometric optics rather than mechanics. If you want to emphasize other ideas in physics, I believe you could easily rearrange the content of your course or even drop some mechanics units such as projectiles or uniform circular motion if you want room for nuclear physics and fluids. (I’m sure some physics teachers will cry “Blasphemy!” for what I’ve just suggested.) Ultimately, the decision on how to use modeling instruction is just your answer to the question, “What’s best for the students in my class?”.

      Hope the above has helped. Thanks for the thoughtful comments.

    1. Penny,

      Not explicitly, but I have been incorporating more and more modeling into the AP course. Part of my work this year was to identify the models that an AP Physics C:Mechanics class would be building. For this year, I came up with the following:

      Momentum Transfer Model (review)
      Energy Transfer Model (review)
      Kinematics and Force Models (review)
      Rigid Rotating Body
      Oscillating Particle Model
      Gravitational Field Model
      Drag Force Model

      Overall, I was pretty happy with the split of the models for this year, but I still don’t feel like they jelled into a coherent model by the end. In particular, I just think that the Rigid Rotating Body model incorporates too much information into that single model. I may look at splitting it apart or reframing it as an extension of the earlier models.

      1. Thanks for the response! I am a new physics teacher next year after teaching 9th grade physical science for 10 years. I will have 2 sections of regular physics and 1 section of ap physics c, just mechanics. I tried modeling for the first time last year using physics first material in my accelerated physical science classes. There were lots of glitches, but i absolutely fell in love with modeling. So, what i will do in physics is obvious! Spending a lot of time this summer reviewing physics and trying to figure out how to approach the ap course. Thanks for identifying relevant models and how it has worked for you.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s