Please Be Loud: Testing in Progress

Think back to when you were in school. You probably remember learning about the Egyptians, the Greeks, and the Romans. You might recall proving that a triangle is congruent. Maybe you learned the sine and cosine of different angles. But it might be harder to remember more than three rulers of ancient civilizations or how to properly apply side-angle-side or what cos(60o) is. More than likely, you were tested on this at some point and probably passed. This highlights a problem with current methods of assessment, where students are taught how to apply a procedure and regurgitate that method on tests, often forgetting what they remembered after only a few days. Some students learn well this way, while some are successful as students but end up failing in the real world.

I have been thinking a lot about teaching these past few months. I might be teaching chemistry labs next year, and so I took a mandatory teaching assistant training through the chemistry department. I developed a short teaching philosophy after participating in discussions and attending short seminars on teaching labs. I also co-taught the Plastic Fantastic course through the School for Scientific Thought. In this endeavor, we had education specialists helping us develop our courses. I went into the School for Scientific Thought thinking about awesome science experiments that I could teach high school students: how to make silly putty, how to build a polymer solar cell and test it, or how plastics stretch and bend. I finished that program with a new appreciation for a backward design approach; my original thinking had good intentions but does not create a conducive learning environment for everyone.

School for Scientific Thought

Instead of designing our course based on mixing chemicals and making electronic devices (which most science-oriented high school students would find interesting), we needed to make sure that all of our students learned something by articulating a desired outcome, determining what we considered acceptable evidence that they achieved that outcome, and developing our instructional approach based on that. To further explain, we hoped to show how changing some atoms in the molecular structure could make polymers behave different (e.g., Teflon is like the same material that plastic milk jugs are made of except that it has fluorine instead of hydrogen). The one thing that we needed was a way to make sure our students were learning: a form of assessment.

In the School for Scientific Thought, there is no homework or tests. These are hard-working high schoolers who are sacrificing Saturdays to learn science for two hours; they don’t need any extra work on top of their school load. But there are other forms of assessment: questions during class and worksheets that can ensure students are following along. One problem that we observed with asking questions during class is that there is invariably an entire spectrum of learning abilities, and there are always a couple of students who want to answer every question and a few students who won’t answer any questions. There were workarounds. We had them form groups and answer questions as a group, teasing out answers from those who like to think slowly before answering. We tried asking for a “show of hands” or to “hold up one or two fingers” instead of answering, hoping to hear from shyer students this way. Many of these methods appeared to be effective in accessing ideas from all of our students and can be modified to fit various classroom sizes and teaching goals.

Just a Test

But what about tests? Even though we didn’t have tests in the School for Scientific Thought, if I wanted to be a professor, it’s practically inevitable that I will have tests in the future. Tests are great tools for motivating students, holding students and instructors accountable, and providing feedback to students and instructors. I went to a talk today; Eric Mazur, professor of physics and applied physics at Harvard University, visited the campus to present a short lecture: Assessment: The Silent Killer of Learning. Now, I wouldn’t go as far to say that assessment is murdering the learning experience, and that was far from Professor Mazur’s point. However, as Mazur pointed out, current assessment methods are merely means to rank and classify students on how well they can study for a test, rather than a means of preparing students for the workplace by focusing on 21st century skills.

Mazur mentioned the aforementioned point that students are merely remembering and regurgitating. In the real world, there are problems presented, but we typically have a vague idea of the outcome. For example, if my car breaks down on the highway, I have a problem. I know the outcome: my car needs to be fixed. I have to understand the problem, analyze possible options, and create a pathway to the outcome. In school, students are taught a procedure, given a problem, and find an answer.

Mazur recalled a physics professor that wanted to make his teaching more accessible to the students. He thought that baseball would be a great way to teach physics. There’s friction as players slide into the plate, velocity and acceleration of a pitch, collisions of a bat and a ball, and parabolic trajectories of a home run. Through the entire course, examples in class were about baseball, homework was about baseball, and the midterm was about baseball. When the professor was writing the final, he realized he ran out of baseball problems, and so he threw in a football problem. The students were thrown for a loop; they couldn’t answer the football problem, despite the physics being exactly the same. This was because they had only learned to apply equations in a single context without really knowing what was going on.

Knowledge Taxonomy

Knowledge Taxonomy (From

Knowledge Taxonomy (From

The knowledge taxonomy pyramid is important to consider when designing assessments. The most basic form of knowledge is remembering, but teachers should strive to have students creating, synthesizing, and analyzing. The students in the above anecdote were only remembering. If a problem said that baseball players were running around the bases, a synapse would trigger and they’d plug in the numbers to calculate the velocity. Mazur further elaborated on knowledge taxonomy. He recalled entering a busy car lot and circling around and around, finally seeing a spot on the other side vacate, only to have it filled by someone else. Frustrated, he thought it would be better if he just waited at the end of the lot and watch the twenty parking spots. He estimated that people spend an average of two hours shopping and since there were twenty spots, he’d probably wait about three minutes since some shoppers might have just left their cars while some may have left 2 hours ago. He found his quick Fermi problem estimate to be exactly correct.

If this were a problem on a test, it would read something like this:

“You are sitting at the end of a parking lot waiting for a parking space. There are twenty filled spaces that you can see. How long will you have to wait to find a space?”

This problem requires complex thinking, brushing the top of the knowledge taxonomy pyramid. It requires an assumption, which most students in today’s teaching environments won’t feel comfortable making. So, they’ll complain, and teachers will change the problem.

“You are sitting at the end of a parking lot waiting for a parking space. There are twenty filled spaces that you can see. How long will you have to wait to find a space? Assume that the average person shops for 2 hours.”

Now, the assessment is not as deep in the taxonomy of knowledge. Students only need to evaluate rather than create and analyze. Still, students will be unhappy. The students will require that the model is presented for them.

“You are sitting at the end of a parking lot waiting for a parking space. There are twenty filled spaces that you can see. How long will you have to wait to find a space if people leave at regularly spaced intervals and shop for 2 hours?”

But students will still complain given their current experience in education. They’ve never seen a problem like this before. So, after giving them the assumption, making estimates for them, and developing the model, there’s a final revision we can implement.

“You are sitting at the end of a parking lot waiting for a parking space. There are twenty filled spaces that you can see. How long will you have to wait to find a space if people leave at regularly spaced intervals and shop for 2 hours?”

It looks identical (because it is), but the difference here is that in their textbook, students have seen this equation:

twait = ½ x Tshop/Nspaces

So now they know exactly what to do: grab a calculator and plug some numbers in. It’s only remembering, and with access to Google literally at our fingertips, remembering should be much lower priority in modern education systems.

To the Future

Of course, it would be unfair if Mazur (and myself) griped about current learning environments without offering any remedies. The first improvement that Mazur mentioned is to mimic real life. The testing environment is isolated. A single student sits in a small desk, suffering from anxiety and half-asleep, half-awake after an all-nighter studying and a few cups of coffee in the morning. In most workplaces, employees aren’t restricted like this. You can work with who you want and use whatever internet-ready devices you need. Mazur showed his students taking tests with access to books, notes, and even Google. This, of course, is a challenge to teachers; your assessments should not be Google-able, and if they are, they need improvement.

Finally, the last method that he showed was a multiple choice exam. Yes, progressive teachers, I can picture your best looks of cringing and repulsion. The multiple choice exam, often touted for its efficient throughput while condemned for its inability to provide students and instructors adequate feedback, can be an effective tool for assessment. Mazur showed a video of students independently answering twenty multiple choice questions. Then, the students formed teams and were given the same multiple choice test and a scratcher card. Students can now work together to find the answer and scratch it off on the card. If a star is revealed, the students get four points. If not, the students get another try but can only get half of the points.

In this case, students must justify the answer to their peers and teach. Everyone participates, everyone argues, and everyone learns. This and similar methods should be considered in classrooms for improving assessment. The current model, while maybe not a killer, is definitely lacking. Perhaps young teachers of today can flip the “Quiet! Testing in Progress” sign on its head to improve the future of education in America.


Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s