Are My Tests Problematic?

The answer is – yes! But how?

Now that I have discussed my assumptions of assessments, it’s time to dissect one of mine so I can improve upon them in the future. Typically, in physics I model my end of unit assessments to be similar to the regents examinations, which are composed of a majority of multiple choice questions with multiple point open response questions following it. The regents exam is graded very objectively – there is a right and a wrong answer for each question, and work must be shown with units for the open response questions. The purpose of creating assessments with this format is to try to ensure predictive validity with the regents – if students are able to pass the unit assessments with the same structure, ideally, they will pass the regents at the end of the course as well. The most recent summative assessment given in class prior to the move towards remote learning has been uploaded here. 

After looking at this assessment, I am embarrassed at how much I have been only gearing the unit towards students who are the “good, classic test-takers” – my tests only celebrate those who are good at rote memorization and recitation. Instead, I am ignoring the divergent thinkers that may approach the concepts in physics in a different manner. In only using these types of summative assessments, I am using the assumptions that:

1. All students know the meaning of the vocabulary words used in the assessment and know how to apply them well (i.e. “gravitational field lines,” “newtons,” “elliptical,” “centripetal,” etc.).
   1. Through answering the questions, students are not given a place to justify their answers, so simply guessing correctly can give the same results as knowing the information and applying it.
2. The questions on the exam are appropriate for their grade level. They are based on old regents questions, so it is assumed regents questions are appropriate for the grade as well.
3. The test will be an accurate predictor of how students will perform on the actual regents exam at the end of the year.
4. All students who understand the material at a conceptual level will also perform well on this exam.

Comparing all of this to my last post, I am perpetuating my issues with assessments by assigning ones that are “not fair to those who are low in reading and written communication skills” and “geared towards student memorization of the material that is taught in class, rather than any real-life applications.” Because this is a summative assessment, it is difficult to use the results to guide instruction throughout the unit; it can, however, inform of what needs to be retaught if all students do not master one content. 

Also uploaded is a student reflection sheet provided after students get feedback on their assessment so that they can reflect upon their performance on the exam. Even with this, I am making assumptions that:

1. Students internalizing the mastery levels that they have achieved with the different topics and are using them to guide their future studying habits.
2. They know how to calculate percentages.
3. They are honest about their reflections and are not just completing the worksheet to get a grade.

Although I have criticized this assessment a great amount, most exams that I have seen throughout my experience from my coworkers are structured similarly. According to Shepard (2000), tests that valued recall and recitation through objective means have been used for the past century, but education theories have progressed a lot since then and now it is time for assessments to catch up. Even behaviorist views where feedback is given through the amount of wrong marks on an assessment, students are more likely to perform because of the overjustification effect, in which they achieve good grades for the external motivators and in turn decrease the amount of intrinsic motivation for learning content, preventing the development of learning information at more than a surface level. 

Another idea that resonated a lot with me from Shepard (2000) is the importance of transfer – “to support generalization and ensure transfer, that is, to support robust understandings, ‘good teaching constantly asks about old understandings in new ways, calls for new applications, and draws new connections (Shepard, 1997, p. 27). And good assessment does the same. We should not, for example, agree to a contract with our students which says that the only fair test is one with familiar and well-rehearsed problems.” All of the assignments that I give throughout a unit are very similar to those that are on the summative assessment, and this is in turn extremely detrimental to the transfer of the content knowledge. Instead, a variety of assessment methods should be used so that students become accustomed to expressing themselves through different means, rather than just the one expected of them on the regents.

Reference: Shepard, L. (2000). The role of assessment in a learning culture. Educational Researcher, 29(7), 4-14.