John Caranci, CTL Lecturer Ontario Institute of Education
john.caranci@utoronto.ca

As physics teachers, few of us consider the psychological effects of assessment on students during assessments. Content, content, content is the mantra of teachers when assessing (as, of, and for student learning). “We are assessing what individuals know” seems to be our Newtonian view. The phenomenon (learning) happens the same way if we assess or not, we believe.

I've spent 50 years in the physics classroom at the high school or university level. We have the basic cause-and-effect rules for assessment. If students study hard and/or you teach them well, they will succeed; but we work with humans. Humans are complex creatures, they do not always follow cause and effect.

Consider this scenario: You have taught Newton’s second law (F = ma) doing four examples on the board which included some manipulation of the formula before and after substitution. You reiterated “without an unbalanced force on a mass it will not accelerate”. You remember to emphasize the term unbalanced. Then you gave the students five homework questions which were equivalent to F = ma in various states of manipulation. You may even have done a few lab activities about the unbalanced force. The students did the homework questions relatively successfully even though a couple of students had to get assistance because they found that some of the questions were not grammatically clear to them. Then, nine days later you give a test on forces of which you give one multiple-choice question and one problem that involves Newton’s second law in terms of F = ma or its definition. As a teacher, you chose the questions, you chose how they were to be assessed, and you chose how many questions to ask. Do students truly know Newton’s Second Law if they answer those questions correctly?

Let us bring in some physics. We understand what happens in an interference activity involving the double-slit, we would see an interference pattern on the screen. We don't know which slit the particles/waves went through, but if we change the experiment a little by setting up an apparatus that will see which slit the particles/waves went through, then the interference pattern disappears… In this context measuring something changes that which is being measured. Is this true for assessment?

Consider another scenario: when an auto racer enters a corner poorly, they assess mentally what mistakes they made to correct it for the next lap. They find themselves incorrectly positioned for the successive three to five corners because that self-analysis puts the car and driver off rhythm. Self-reflection can be negative or positive during a test. This doesn't just mean reflecting on the question at hand, but also their metacognitive study of the material in that question.

During a test there are three component individuals: the person who created the test, the invigilator and the student taking the test. Each has an effect during the test. Students that self-identify even subconsciously as competent in the subject influence the specific outcomes of the assessment. It can even affect an individual question. Basically, these are self-fulfilling prophecies of their performance. The creator of the test may also influence the results from the questions chosen, where students may experience changes in mood and behaviour based on their ability to correctly anticipate the types of questions that ultimately appeared on the test. An invigilator can also have an effect during the test due to their body language (kinesics) and eye contact. If an invigilator focuses on an individual, the student can feel that they are suspect in some way. This influences the results of the test. This seems incidental, but it will still affect the results of the test.

Teachers often make several assumptions about the effects on students during physics test-taking. While these assumptions can vary depending on the educator's experience and training, some common ones include:

• Assumption of Uniform Test Anxiety — all students experience anxiety, for some it instils a challenge to their competency, but others revel in the rush of this anxiety
• Belief in Fixed Ability Mindset — a few educators believe that some students only reach a competency threshold determined by their innate ability
• Time Pressure Effects — in a sliding scale from time pressure being deleterious, to others requiring the time pressure to excel
• Expectation of Rational Problem-Solving Under Stress — some students lose rationality under stress conditions
• Underestimation of Stereotype Threat — many students accept their stereotypical role, for example at-risk students may accept their at-risk performance because the student, their teachers, and their parents, all only expect them to earn 50% in the course
• Assumption That Test Performance Reflects Understanding— one of the strongest assumptions in education today (by students, parents, teachers, administrators, and the general public)
• Neglect of Emotional and Motivational Factors — students have a life outside the class that effects tests done in class
• Belief That All Students Have Equal Test-Taking Strategies (or Should) — this assumption is used by some teachers not to teach test taking strategies (not review strategies)
• Overlooking the Impact of Prior Experiences — remembered negative emotions in previous tests
• Assumption That Feedback is Always Helpful — marking up a lab report full of red ink may seem like you are providing the student with tons of helpful advice, however many students would see all the red and think they did terribly, and simply stuff the lab report into their backpack without reading any of the feedback
• Overlooking Cognitive Load — how many tests are they going to do this week?
  

Recognizing these assumptions can help teachers’ implications for teaching. Try these:

• Differentiate support (e.g., offering anxiety-reduction strategies). Teachers should learn how to recognise when students are under stress by knowing kinesics and para-language.
• Foster a growth mindset by emphasizing effort and improvement over innate ability. Place the answers on the test so only the steps count. Give large number of practice questions where the student only does only the first step (this improves the confidence of starting a problem).
• Design assessments that minimize unnecessary stress by including skills (like measurement) as well as knowledge questions.
• Provide explicit test-taking strategies to help students perform their best, like how to plan doing multiple-choice or algorithmic problems. For the most nervous, cover the test with a blank sheet of paper only looking at one question at a time (some students will scan the test until they see the question they cannot do, raising their nervousness).
• Reduce cognitive load by creating an assessment timetable with other subjects and breaking up high cognitive load topics.
• Reduce anticipatory anxiety and choking under pressure by simplifying and systematising review of the kinds or types of questions.
  

Understanding these assumptions and solutions can help teachers create more supportive testing environments and develop strategies to mitigate the negative effects that arise. If you have any specific concerns or experiences related to these assumptions, feel free to share with me to the above email address. I will use any solutions you have developed for your classes in my classes.

Tags: Assessment, Pedagogy