Building skills through online learning: 4 data-driven best practices

Teaching and learning online can feel daunting, especially at first. With the pandemic forcing millions of instructors and students abruptly into remote schooling, many have questioned the quality of online learning and its effectiveness — especially those new to the online setting. This new landscape inspired those of us at Coursera to explore what drives the most successful online courses and why.

Drawing on the satisfaction, skill development, and career outcomes of over 200 million course enrollments, the Drivers of Quality in Online Learning report showcases the power of online learning and provides actionable, data-driven insights for how instructors and learners can optimize their digital learning experience.

With eight years of data, we’ve found that, when done well, online learning can be extremely effective at helping students acquire and master new skills — including many that are in high demand in the current jobs market. In fact, 73% of our online learners report a positive career outcome within six months of completing a course. The percentages are more than 80% for those studying business, data science, information technology, and computer science.

Skill-focused learning has never been more important. In recent years, employers have struggled to find qualified workers for critical roles, and the COVID-19 pandemic has exacerbated the skills gap, according to a recent article in The New York Times.

No matter the subject area, there are specific techniques instructors can use to dramatically improve student engagement, satisfaction, and career outcomes. Here are four of the most effective ways we’ve found to build job-ready skills through online learning:

1.     Define specific learning objectives to help learners connect the content with their career goals.

You can drive career outcomes by defining specific and actionable objectives for each module and the whole course. For example, instead of “Learn basic Python,” a more helpful description would tell students that by the end of the course, they’ll be able to explain a “for-loop” and implement one to simplify their own code.

On the Coursera platform, students can see the concrete skills they’ll be learning in the “Key Concepts” section at the beginning of each week. When you make the learning objectives clear and provide examples of how that skill is used in an industry setting, your students get excited to dive into the content. Courses with specific objectives outlined for each week have a statistically significant increase in the rate of career outcomes.

Your learning objectives should shape every aspect of the course design. When creating a course, start with learning objectives and work backward. Where do you want your students to be at the end of their learning journey? Define those specific learning objectives, and then ask, “How are we going to assess if they’ve actually gained those skills?” Decide what projects will help students master those skills and how they can practice along the way. Finally, outline the instructional materials that will introduce key topics, provide concrete examples, and help students apply the skills to their own career aspirations.

2.     Test students’ knowledge to keep motivation and engagement high.

When students take frequent assessments as part of an online course, they’re more likely to complete the course — and also more likely to achieve a career outcome. Because some universities rely heavily on end-of-term exams, students often receive very little feedback until the end of the semester. As a result, they don’t have an accurate sense of how well they’re doing or the concepts they need to review again. Instead, help students master new skills and identify gaps in their understanding by providing multiple checkpoints throughout your course. These checkpoints should include a mix of graded and practice assessments.

Consider beginning your course with a longer module to scaffold learners into the content, and then test their understanding early and often. Feedback could come from peer-review assignments or automated feedback on quizzes. At Coursera, we’ve developed option-level feedback: depending on how a student answers a question, the system prompts them to review a specific piece of content, or — if they answered correctly — it reinforces the concept by reminding them why their response was correct. Including option-level feedback on quiz questions is a rare method for significantly increasing both satisfaction and completion rates for online learners.

In this way, an online course can provide scaffolding that builds up the learner’s mastery over time. Low-stakes or no-stakes assessments are an important part of that process. Ask students to take a practice assessment after watching videos, engaging with readings, or participating in a simulation to see if there are still gaps in their knowledge, and then direct them back to what they need to review. Including in-video questions and hands-on practice, each produces a 10% increase in average skill development for those who complete the course.

3.     Use hands-on projects, including peer review and programming assignments, to drive higher skill development.

Instructors can help their students connect skill-based learning objectives to career goals by providing industry examples and authentic, hands-on projects. Applied projects foster higher skill development by engaging students in the creation and evaluation forms of learning. In fact, including programming assignments in technical courses on Coursera increases average student skill development by 30%.

For example, in a data science course, instead of using a contrived dataset, you should use real data from an open-source organization or partner with a company that can provide relevant data for the project. In a business course, you could design a project centered on a real case study or ask students to apply the core skills to their own work contexts.

Have students put their new skills into action with reflection and peer reviews. Not every student feels comfortable asking a question in a lecture hall with hundreds of other students. But in an online setting, students can write a question in a forum, respond to a discussion prompt, or learn from a peer’s work.

For example, in the Applications of Everyday Leadership course created by the University of Illinois, discussion prompts and peer review assignments guide learners to reflect on key concepts and evaluate new business situations. In Google’s Technical Support Fundamentals course, students apply what they’ve just learned about creating effective technical documentation by writing their own and then reviewing their peers’ work.

4.     Break the content into bite-size chunks.

For higher engagement and larger skill gains, our data show that the optimal course length is 4 to 6 weeks, with about 4-6 hours of total learner engagement time per week. Avoid making students wade through a 90-minute lecture to find one specific concept by instead providing small chunks of video content with clear titles. Videos in the 3- to 9-minute range are ideal for continued engagement and satisfaction. In fact, by keeping lectures under 10 minutes in length, instructors can increase their course completion rate by 16%. This structure also makes it easy for learners to find and review specific concepts as needed.

The first module should be longer than average, to ground students in the subject matter. We recommend including the most videos in the first module and gradually transitioning into more project work, with most of the time in the final week spent on hands-on assignments. Students are more likely to complete a course when the total amount of learner engagement time per week stays relatively stable, but how you use that time should vary from week to week.

For example, the first course of the University of Michigan’s Applied Data Science with Python Specialization has more videos at the beginning to introduce students to key terms, definitions, and topics that will be used throughout the course. The instructor uses screen sharing to show his code and provides practice Jupyter notebooks, so students can follow along and run the same programs the instructor talks about in the videos. The first week has a light-weight assignment just to get students going, but by the end, they are building complicated data analysis programs on their own.

Related Posts