“Learning by doing”: Open science tutorials for students, by students

For Dr. Joey Bernhardt, teaching science goes beyond passing on knowledge to the next generation. It also includes passing on the skills to do open science.

“Open science” refers to the sharing of data to promote collaboration and transparency, and the re-use of data over time to foster new discoveries.  It represents a growing paradigm shift in how science is done, but science education has not necessarily kept pace with this shift.

“Open science practices are an essential part of the scientific process, and students need to learn the technical skills necessary to do open science,” explains Bernhardt, an assistant professor in the Department of Integrative Biology.

In collaboration with a team of graduate students from across Canada and colleague Dr. Jennifer Sunday at McGill University, Bernhardt developed a series of interactive, online tutorials that integrate fundamental knowledge in ecology and evolution with open science concepts and skills.  Specific skills that relate to open science can include, for example, proficiency in data and code sharing platforms, programming language, and version control systems. 

The tutorials, now published in the journal Ecology and Evolution, were designed with a scaffolding approach that can be thought of as Lego blocks. First, you build the foundation of your tower with knowledge of the subject area. Then, you build the first few floors by doing simple data manipulations. Finally, as you keep building, you learn the coding that goes into developing data visualizations and analyses.

Mastering fundamental ecological concepts while simultaneously learning open science technical skills would normally be too overwhelming for an undergraduate class.  However, the scaffolded approach allows students to engage with the technical skills as much or as little as they would like, based on their own comfort level, and everyone gains the same content knowledge.

Importantly, the tutorials were created using "rescued" historical data sets that might otherwise be lost or remain underused. For example, one of the tutorials uses more than three decades worth of tree swallow data from Long-Point, Ont., to teach data exploration techniques. Another tutorial is based on legacy tree data from the Canadian boreal forest to teach biodiversity, dendrochronology and data visualization. Students who complete the modules gain an appreciation for the types of data that are readily available data through open science.

In turn, the graduate students who helped to develop the tutorials found it to be an incredibly valuable learning experience and hope to continue to build these skills going forward.

“The entire process of working with this team of graduate students was inspiring,” says Bernhardt. “The students did not even know each other before this project, but it turned into a highly positive and collaborative workspace.”

The team behind the tutorials sees many potential applications. For example, tutorials can be widely incorporated into the undergraduate curriculum to teach disciplinary knowledge and provide hands-on experience in data manipulation and analysis. At the graduate level, students can enhance their programming and other technical skills as they develop new tutorials.

In short, it’s a win-win situation for all, and represents an innovative step forward in how universities can prepare their students with the skills they need for impactful careers in the sciences.

This study was part of the Living Data Project, which is funded by a Collaborative Research and Training Experience (CREATE) from the National Science and Engineering Research Council (NSERC) and the Canadian Institute of Ecology and Evolution (CIEE)

Read the full study in the journal Ecology and Evolution. Visit the Living Data Tutorials website. 

Read about other CBS Research Highlights.