Do SEI surveys measure success or perpetrate bias?

At the end of the term, emails flood into student inboxes with the reminder to fill out teaching evaluations. This is a chance for students to grade their professors’ performance as an educator.

Student Experience of Instruction (SEI) surveys help departments understand students’ academic experiences and are used to grant teaching awards, holding significant influence over instruction.

Student evaluations also impact reappointment, promotion and tenure decisions.

However, scholars have questioned the accuracy of these surveys, and how they might mask gender, racial and ethnic biases.

These implications are particularly important at UBC, where only 22 per cent of senior management — deans, associate vice-presidents and others — identify as BIPOC according to the 2022 Employment Equity Interim Report. Currently, 38.5 per cent of UBC Vancouver faculty identifies as racialized.

Faculty’s perspective

The UBC Faculty Association (UBCFA) has been vocal on their opinions concerning SEIs.

In a 2019 public statement, it proposed the surveys not to be used when determining promotions of faculty.

“The invalidity of these instruments has been known for a long time,” wrote UBCFA, citing a 2018 arbitration between Toronto Metropolitan University and its faculty association. The arbitrator ordered the institution to stop using student evaluations to make promotion and tenure decisions, stating it had “little usefulness in measuring teaching effectiveness.”

The arbitration noted evidence from expert testimonies and peer-reviewed publications which revealed personal characteristics such as race, accent, gender, age and “attractiveness” could affect the results of student surveys. Other characteristics such as class size and quantitative versus humanities-focused courses could also impact results.

The UBCFA’s most recent statement about SEI surveys was in 2022, where it called them a “cheap candy of teaching metrics.” UBCFA wrote that these surveys are biased against marginalized groups and do not perform their intended purposes.

It entered a contract negotiation with UBC asking the institution to stop using these surveys as summative measures of teaching.

However, the ratification package for the 2022–25 period did not include a change in SEI practices.

The 2022 President’s Task Force on Anti-Racism and Inclusive Excellence Final Report found that “similar to other IBPOC folks, Indigenous graduate student teaching assistants report low teaching evaluations due to implicit bias of predominantly White students.”

The report also said BIPOC faculty members with accents may face negative course evaluations. Ultimately, the report called for student surveys to be removed for being discriminatory.

UBC’s commitment to improving representation

“Achieving and maintaining equity in employment is a systemic challenge facing many organizations and it’s one UBC takes very seriously,” wrote Associate VP Equity and Inclusion Arig al Shaibah in a statement to The Ubyssey.

She said the university is committed to improving representation across all equity occupational groups while focusing on improving experience and retention.

Currently, the university is “preparing to launch a guide that identifies a series of best practices to enhance equitable search and selection processes and to advance inclusive excellence in hiring.”

Senate working group and research

In Spring 2019, UBC formed the Student Evaluation of Teaching Working Group (SEoT) to re-examine the university’s approach to student evaluations.

As a collaboration between UBC Vancouver and UBC Okanagan, the group was chaired by 1 professor from each campus and composed of 15 faculty and student members.

After over a year of consultation, the group published its 16 recommendations in 2020. The group found no systematic differences in aggregate data when examining between female and male instructors, and recommended that the next step be conducting an analysis based on ethnicity.

Work on processing data based on instructor ethnicity is reported to have commenced in 2022.

Dr. Catherine Rawn, professor and associate head of undergraduate affairs in the department of psychology, was among the members of the 2019 SEoT working group.

She acknowledged the biases that arise when examining interactions of gender within disciplines. Marginalized faculty in fields traditionally dominated by white men are especially vulnerable to evaluation bias.

For instance, engineering stands as a traditionally male dominated field. Last year, the Faculty of Applied Science reported that 34 per cent of assistant professors identify as women across both university campuses. A 2019 report from Statistics Canada noted the largest gender gap in STEM enrolment was within engineering programs.

A 2023 study published by the UBC Sauder School of Business shows the business field may also be affected by unconsciously held beliefs about physical appearances and areas of expertise. The results also showed people associate the position of a finance leader with the description of a white male. Currently, 1 of the 6 Sauder finance branch full professors is female and 5 of the 18 full professors are female.

Similarly, al Shaibah pointed to the important role of disaggregating data to unveil the masked inequalities that may not be visible in the aggregate.

She wrote that a report following up on the interim 2022 Employment Equity Survey Report will include Indigenous, racialized and disabled employees. It will be released in the spring and will report on intersectional data.

However, challenges arise in balancing the act of protecting sensitive information while providing a comprehensive representation.

Rawn also highlighted a potential issue with generalizing the broadly construed literature on student surveys to UBC experiences.

“From a psychological perspective, when we administer the same survey, and when we want to compare across samples, we administer exactly the same survey in exactly the same way,” she said. This underscores the challenge of comparing across institutions with differing survey structures and methodologies.

As for discussions regarding the presence of SEIs in the realm of reappointments, promotions and tenures, Rawn said it is important to include the student voice in some way.

“I don’t think that the student voice can be the only indicator of teaching but I think it’s an important one,” she said.

“There’s work to be done about how to do that in a fair way that empowers people to identify bias and helps us to protect the faculty members who are most vulnerable.”