Technical Skill Development and Content Knowledge

Data literacy & content knowledge – require prerequisite skills or build as you go?

Establishing clear learning objectives is an essential part of any program development. Learning objectives identify what students are expected to learn and be able to do by the end of the program. They guide curriculum development, the pacing of the program, the composition of administrative and mentor teams, and provide the vision to keep the entire program together and focused.

How we did it:

Learning outcomes for our program included:

Polar science: understanding the polar regions, with specific depth in the Arctic, as complex systems and global environmental regulators.
Data literacy and integrity: understanding how to find, use, collect, manage, assess and interact with various datasets.
Visualization and interpretation: creating digital visualizations for analytical, interpretive and communication purposes including mapping and storytelling.
Team science: experiences with the practices of cooperative and team-based learning, equitable and transparent processes, identifying and leveraging unique strengths of each team member, reflection and adaptation to optimize contributions, sharing workload and credit, and expressing multiple aspects of one’s identity while participating in science communities.
Science communities and communication: tell the story of research findings, engage peers within and beyond disciplines, connect with public audiences, and position findings in terms of social, ecological and policy contexts.

There were no strict prerequisites for participating in ARC-Learn, as research skill development was a main focus of the program. This meant some students entered the program with strong skills from courses or other research experiences. Other students entered with very little experience and learned the needed skills through the offerings of the program, while some students took classes or sought out other avenues of developing the needed skills during the two years of the program.

We met as a cohort every other week over the course of two years. These two-hour meetings were used to develop the skills needed to meet the learning outcomes listed above. Some content was provided in a standard classroom delivery, including PowerPoint presentations, interactive activities and assignments. Assignments were always specific to their research projects, and were designed to help lay the groundwork for the next steps in their project development.

Cohort meetings were also used as an opportunity to crowdsource key skills from mentors. Finding, cleaning, analyzing and interpreting data was a big challenge for most of our students, especially because most sophomores and juniors have very little experience using programs like Python or R. Sometimes, the specific knowledge or skill a student needed was not to be found with the mentors in their research team. We therefore held workshops to allow students to connect with mentors from other research teams. This approach helped multiple students get “unstuck” and move forward on their projects.

Guest speakers were invited to host workshops during cohort meetings as well, including how to prepare a scientific poster for the poster symposium, a panel of polar and polar adjacent scientists to discuss their career paths, and a workshop to reframe the skills they obtained through ARC-Learn to be most competitive when applying for graduate school or jobs.

Considerations for your program:

Identify clear learning objectives for your program, and build curriculum and program structures to meet those

Consider each element available (such as mentors, classroom sessions, team meetings, and online interactions) as individual resources that can be utilized in creative ways to help students meet your learning objectives.

Working with existing data sets is an entirely different task than working with data you collect yourself. This was a challenge for students – who had limited coursework to support their efforts and did not necessarily realize the amount of self-directed work they would need to put into finding, cleaning and analyzing data – as well as the mentors – who were supporting multiple students at once, each using different data sets and programming languages. Make it very clear at the beginning of the program what this process will look like, and provide necessary support for student skill development.

Provide focused, practical, hands-on work sessions for students to make progress on their research projects. Topics could include how to choose a coding platform, how to open and clean data sets as well as work sessions for each of the programming platforms students are using.

Conduct an inventory or assessment of student research skills at the beginning of the program to establish a baseline, and to inform how you will support each student in gaining the skills they need to be successful.

In a long-term, self-directed program, students are in a position where they need to identify their own research questions, find their own data, clean it, analyze it and interpret it. Mentors may or may not be familiar with the data set or the program the student is using to analyze it. Consider securing funds to support a data science coordinator to join the implementation team. This person could host the hands-on workshops described above, as well office hours to support students in the technical aspects of working with data. This would relieve much of the burden on volunteer mentors, and would help the students to be successful at finding and accessing the data they need.

Students in a self-directed research program need to have a foundational level of experience related to data analysis and subject matter knowledge, otherwise they — and their mentors — can find the experience quite frustrating. Consider requiring specific coursework as prerequisites to participating in the program, to be of junior or senior standing or whatever makes most sense for your institution and program, to assure they have the skills they need to be successful.