After the height of the COVID-19 pandemic, even as institutions had adapted to online learning, students belonging to underrepresented ethnic minority groups struggled to bounce back academically as compared with their non-minority classmates.
Researchers in the Cornell Ann S. Bowers College of Computing and Information Science looked at student performance before, during and after the pandemic caused a transition to online learning. While grades on average returned to normal in 2021, the achievement gap between ethnic minority and non-minority groups grew, likely due to ongoing pandemic-related challenges. The researchers suggest that instructors monitor not only achievement but also engagement, to help identify struggling students.
“If we really want to provide equal opportunities for all students, and we want underrepresented minority students to do well, we really need to pay attention to their experiences and their perspectives and give them the tools to do better,” said first author Lilach Alon, formerly a postdoctoral researcher at Cornell and now a researcher and lecturer at Tel Hai College in Israel.
The new study, “From Emergency to Sustainable Online Learning: Changes and Disparities in Undergraduate Course Grades and Experiences in the Context of COVID-19,” was published in the October issue of Computers & Education.
The forced transition to online learning offered an opportunity for educators to innovate new teaching methods. In 2019, Alon and her co-authors had already been collecting data on student achievement and engagement in four large lecture courses at a large research university. When the pandemic struck and all classes moved online, they continued collecting data, first during the “emergency online learning” phase in spring 2020 and then in the “sustainable online learning” phase in 2021.
Pre-pandemic, students belonging to ethnic minority groups showed a small achievement gap in the courses, earning about a quarter of a grade lower, on average, than non-minority students. During lockdown in spring 2020, all types of students received higher grades due to a switch to pass/fail grading and institutional policies encouraging leniency and accommodation. After instructors had the chance to refine their online courses in 2021, students generally found the courses to be more active and engaging. However, a larger achievement gap emerged – students belonging to ethnic minority groups averaged almost three-quarters of a grade lower than non-minority classmates, which amounts to the difference between an A- and a B.
While the exact reasons for this gap are unknown, previous studies suggest it is likely due to lack of resources – such as a quiet space and high-speed internet access – or that students belonging to ethnic minority groups more often work or care for family members in addition to their coursework.
“It’s easy to think that the pandemic is over and we’re back to normal. But the research shows this is not necessarily true for all groups equally,” said senior author René Kizilcec, assistant professor of information science.
From the study, the researchers drew several lessons that can be applied to improve the student experience in online courses. Alon and her colleagues recommend that instructors give students multiple, smaller tasks during the semester to monitor achievement, instead of one big final, and that they give surveys to identify students who may need help. They also suggest structuring courses to include small group assignments and synchronous activities, where students are learning in real-time, to help them feel more connected.
Additionally, they suggested institutions offer more support and resources to instructors developing online courses.
“In the academic literature, people say that hybrid online learning is the new reality of higher education,” Alon said. “If that’s the case – and it seems to be the case in the U.S. – we need to really rethink pedagogy and student engagement.”
SeoYoon Sung, a former postdoctoral researcher at Cornell, and Ji Yong Cho, a doctoral student in the field of information science also contributed to the study.
Patricia Waldron is a writer for the Cornell Ann S. Bowers College of Computing and Information Science.