Getting to Know Coursera: Assessments
by Katie McEwen, graduate assistant
So far in our discussion of common features in Coursera, we’ve focused on two methods of presenting content online: video lectures and video discussions. Certainly, students (whether online or in the traditional classroom) benefit from coherent, clearly sequenced, thoughtful and thought-provoking discussions and lectures. But, just as certainly, assessing comprehension, application, or mastery online poses a number of tricky questions for both students and faculty. How do faculty ensure rigor? And how do students work to demonstrate successful completion? In this post, we’ll turn our attention to typical methods of assessment in Coursera.
Again, though, the question arises: is there be such a thing as “typical” in this setting? And, once again, the answer is yes. With now over 200 courses across a wide spectrum of disciplines, the Coursera consortium offers a global audience of learners access to a dizzying array of possibilities. Yet, as we’ve already seen, these courses often share more in common than might be immediately apparent from a course title.
Much of this commonality is engineered by the Coursera platform, which supports — on the level of technology, to say nothing of the level of ideology — certain formats, while discouraging others. These formats, in turn, reflect a certain approach to teaching and learning (often, as we’ve noted rather teacher centered). Despite this, however, Coursera does also hold the potential to facilitate innovation, not only online, but also in the on-campus classroom.
Assessments follow this common thread, in that Coursera supports a limited number of assessment types: quizzes, programming assignments, and peer assessments.
Stand-alone quizzes are the most frequently used method of assessment in Coursera courses. Across almost all disciplines, weekly (or sometimes bi-weekly) automatically graded quizzes provide students with a quick check-in to test knowledge, get feedback, and further explanation of material. Quizzes in Coursera support multiple-choice questions (either one right answer or “select all that apply”), fill-in-the-blank questions (short numerical or free-form text answers). Immediate feedback lets students know what they missed and why; multiple attempts (when allowed by the instructor) emphasize mastery of the material.
Here, it soon becomes clear that writing smart, challenging multiple-choice questions is not nearly as easy as it might at first seem. In fact, with basic test-taking skills, I’ve managed to score surprisingly high on a number of quizzes in disciplines in which I have absolutely no knowledge or expertise. Given my background in German, I was quite proud of the 50% I earned on a Machine Learning quiz.
Like the stand-alone quizzes, programming assignments are automatically graded to generate both a score and feedback for students. Unlike quizzes, these assignments ask students to actively use their knowledge to create code or analyze data, implementing or applying material learned in lecture. Right now, programming assignments are most frequently found in Computer Science. But by enabling students to submit work in a number of formats — for example, programming languages and R, Mathlab or Octave, or even excel spreadsheets — these assignments are flexible enough to evaluate formatted output in disciplines as various as Statistics, engineering, or finance. Programming assignments challenge students to apply knowledge and work through complex problems, while still providing the immediate feedback only possible from a computer. And, at the discretion of the instructor, programming assignments may be revised and submitted multiple times, inviting students to gain mastery instead of merely receiving a grade.
Peer assessments similarly allow students the opportunity to create a wide range of material, from design projects to essays to videos or even music recordings. Unlike the other forms of assessment in Coursera which can be automatically graded, peer assessments require the input of fellow students. These kinds of assignments thus depend upon a good-faith effort on the part of each student not only to submit original work in the proper format and the proper language (largely still English), but also to then anonymously evaluate the work of others attentively and constructively.
Obviously, rubrics are essential to this form of assessment, as students are asked to provide both quantitative and qualitative feedback to their fellow learners. Yet, as we know, grading effectively and fairly is a learned and reflective practice, an art unto itself.
In light of these growing pains, Coursera is actively working to modify and improve peer assessment on the platform, including revising rubrics. Yet, peer assessments are still both an exciting innovation to facilitate greater student interaction and engagement as well as a sometimes-challenging problem for both instructors and students.
Next time, we’ll continue our discussion of assessments in Coursera.