Thay đổi tư duy về cách đánh giá học sinh có thể cải thiện sức khỏe tinh thần của các em

When you were in school, what stressed you out the most? For many students, classroom assessments — graded tests, quizzes, group work, presentations, labs and the like — would be near the top of the list.

Like decades of research before it, the research my students and I have been doing continues to show that classroom assessment is a persistent source of social pressure and general unhappiness for post–secondary students.

This isn’t to say that instructors and institutions aren’t trying to address the problem. Two solutions seem to be common.

First, there have been many efforts to support student mental health on Canadian campuses, including national standards, institutional plans, counselling centres and crisis lines, to name a few. While helpful, these solutions often don’t have the capacity to handle the problem.

Second, when assessment is targeted more directly, the response often involves calls for particular forms of assessment, debates about ungrading or recommendations to minimize grades altogether. As a case in point, headlines have been abuzz with news that the University of Michigan’s plans to make first–semester courses credit/non-credit for freshmen, with the hope that it will alleviate stress.

But many of these decisions are outside the scope of most individual instructors. My research with my colleagues shows that even decisions instructors do control, like which format of assessment to use, may not matter very much to student well–being.

In other words, getting rid of specific types of assessment, or even grades, doesn’t seem to guarantee improved student well–being.

A different approach to well–being

As a social psychologist who specializes in education, I decided to ask a different question. Knowing that students are completing assessments that compromise their well–being right now, and that instructors have finite resources, what solutions might be borrowed from social psychology?

One option was clear to me. Instead of looking for a magic form of assessment that all students would agree supports their well–being, we could turn to robust psychological theories of well–being.

Over the last five years, I have been using self–determination theory to guide the redesign of classroom assessments for post–secondary students.

Self–determination theory examines human flourishing and well–being. Although it is best known for its distinction between intrinsic and extrinsic forms of motivation, it also explains that all humans have three basic psychological needs: autonomy, competence and relatedness. When these needs are satisfied, they produce a sense of well–being.

Instructors can satisfy students’ basic psychological needs by taking their perspectives into account, supporting their autonomy, making room for negative emotions, providing rationales for their decisions and showing patience.

Satisfying basic psychological needs is a precursor to reliably increasing a number of indicators of well–being, including enjoyment, effort, persistence, creativity and satisfaction. Until our program of research, however, no one had tested whether these sorts of practices would bring about similar effects when applied to classroom assessment.

Making assessments more supportive

Through a series of studies, my colleagues and I tested the effects of adding what we called “need–supportive features” directly to assessments. Need–supportive features are specific assessment practices that convey psychological information to students that helps fulfil their basic psychological needs.

The four features were: using a test blueprint, grouping items to match the blueprint, providing brief supportive messages at the start and end, and offering students an opportunity to provide feedback.

While these features may appear simple, it’s crucial to remember they are based on principles of self–determination theory and their reliable connection to well–being. These changes are not a magic solution to student stress: they are evidence–based choices.

We started with multiple–choice tests, a particularly hated form of assessment.

The results of two studies showed that adding four features improved the extent to which students felt the test supported their basic psychological needs. Students’ performance also increased.

Next, we partnered closely with one post–secondary instructor to revise her full assessment plan. This meant adapting and elaborating the multiple–choice, written assignments and the overall assessment plan for her undergraduate literacy course.

The pattern of results was similar, with students reporting higher basic psychological need satisfaction, perceived quality and fairness, and less anger, hopelessness and anxiety when assessments contained the need–supportive features.

This partnership also allowed us to document improvements in the instructor’s well–being. She told us that assessment felt less like a burden, that she experienced less anxiety and that she had more time for meaningful connections with her students.

Fairness still matters

There is one important caveat. In addition to need–supportive features, we ensured that all assessments were designed according to best practices themselves. We have been developing new ways to think about what makes high–quality multiple–choice questions in our ongoing research.

It is not sufficient to add need–supportive features to poor–quality assessment. In some of our early research, it was clear that fair assessment is foundational to basic psychological need satisfaction, and therefore to well–being.

Students and instructors dedicate many hours each academic year to designing and completing assessments. Too often, those assessments are associated with a sense of unhappiness for both groups.

As a promising alternative, our research shows that when assessments are designed to intentionally support basic psychological needs, measurable improvements can be realized. Everyone wins when assessments suck even just a little bit less.

Lia M Daniels receives funding from the Social Science and Humanities Research Council of Canada.