Foundations of Data Science (Data 8) typically starts the new semester in Zellerbach Hall to accommodate the class, which draws more than 1,000 students. Last spring, when UC Berkeley switched to remote learning to protect against COVID-19, Berkeley’s data science professors quickly transformed this large yet interactive course into a complete remote experience.
“Last semester when we were forced to move everything online we began to experiment with different ways to teach the material,” explained Assistant Professor Joey Gonzalez, Data 100 instructor. “One of the bigger changes we tried was to focus on organizing the lecture around the interactive Jupyter Notebooks that we would normally release with a traditional lecture. We recorded snippets of the lecture and then embedded the video in the notebook along with questions and interactive plots.”
Smaller Discussion Sections to Create Community
This fall, 1,350 students enrolled in Data 8. The instructors were concerned with creating opportunities for such a large remote class size to form connections with their classmates and support students from falling behind in the coursework. The summer semester allowed the team to strategize how to restructure Data 8 in a way that would be cohesive for remote learning.
The discussion sections are now smaller to help promote more interaction amongst the students. The hope is that the smaller discussion sections will help students find a partner to study with and support one another.
“We are adjusting Data 8 to emphasize community and do more learning in small groups. Previously our labs and discussions were in person in groups of about 30 students,” Professor David Wagner said. "We know that interaction in large groups on Zoom can be unwieldy and intimidating. We adjusted the discussion sections so each has just 6 students and one teaching assistant.”
Paired Learning in Labs
Traditionally in Data 8 students get hands-on experience with the curriculum in the labs. The labs are now designed for students to work in pairs so they can learn from each other. “Course staff will rotate through to help them,” said Wagner. “Our experience is that peer learning can be one of the most effective learning strategies, so we're hoping that this will help our students get a lot out of the lab. If they also meet someone new each week and get to make connections and build community, that's a bonus.”
Taking an online course can be an isolating experience for students especially in a large-sized course such as Data 8. The instructors were also concerned with ensuring that students can keep up with the coursework. To monitor student progress the instructors implemented three mini-quizzes per week.
“We monitor their answers and give them immediate feedback on what to study or what their misconception might be if they answer the quiz incorrectly,” explains Professor Wagner. “We'll continue our practice of having extra, optional weekly small-group tutoring for everyone who wants more help or more practice.”
Remote learning presents challenges and opportunities for all educators. Fortunately, Data 8 instructors have already created a format to teach to a large number of students but the remote format will continue to be a learning process throughout the semester.
“We're committed to providing an encouraging, supportive environment in Data 8 and building a sense of community where everyone helps each other learn. I am looking forward to working with students this semester!” said Professor Wagner.