Learning analytics

Learning analytics

Table of Contents

    What are learning analytics, and why do they matter? Learning analytics are assembled through a process of collecting, analyzing, and reporting data about learners and their activities. This data helps organizations gain insights into the effectiveness of training programs, student learning, and outcomes. By leveraging the insights derived from learning analytics, organizations can identify areas for improvement, track performance, and optimize learning experiences for better outcomes.

    This comprehensive guide will:

    • Answer the question “What are learning analytics?”
    • ExploreDiscuss the uses and benefits of these analytics
    • Explain the process, key metrics, and levels of learning analysis
    • Identify how to define and track learning analytics for your organization

    What are learning analytics used for?

    Learning analytics support different industries and jobs, from higher education and healthcare to business and technology, for both in-person and online learning. This big data industry provides all the information and measurements you need to determine how your corporate training program is going and how it should be improved. Educational data mining is being used by organizations all over the world to help them reach their goals more quickly and efficiently and stay ahead of the competition. By tracking key metrics such as user engagement, time spent on tasks, and successful completion of courses, organizations can make adjustments to ensure their training programs are performing at their best. 

    What are learning analytics going to do for your business?

    • Improve learner experiences: By understanding how their learners interact with the training material, companies can tailor their content and activities to the needs of their learners. For example, a company may find that certain sections of a course are receiving less engagement than others and make adjustments accordingly, like creating more interesting content or incorporating more real-world examples.
    • Enhance efficiency: With real-time insight into course progression, companies can identify any bottlenecks or other issues slowing down completion times. Additionally, they can use the data to assign resources more effectively to help ensure learners understand expectations for completion.
    • Effective allocation of resources: So, what are learning analytics’ effects on your budget? Companies can use learning analytics to make data-driven decisions about allocating resources. By analyzing data points such as budget, enrollment rates, and completion rates, companies can see what approaches are most effective for their team and make informed decisions about how to prioritize resources.
    • Optimized performance outcomes: With access to data points such as course progress, engagement levels, and assessment scores, organizations can use learning analytics tools to assess performance outcomes and identify potential areas of improvement. For example, suppose learners are performing well on quizzes that focus on information recall but struggling to complete activities that focus on information application. This insight can be used to create more real-world examples and provide actionable steps. 

    The learning analytics process

    Assembling learning analytics can be pretty simple and easily automated with the right educational technology in place. The process consists of three steps: 

    1. Collecting data: This involves gathering information from the learning process, such as user engagement, assessments, activity outcomes, and other feedback data. This data usually comes from different sources, such as a learning management system (LMS), surveys, and other measurements and trackers.
    2. Analyzing data: This step analyzes the data you collected in various contexts. This includes looking for patterns in the data, examining trends, and identifying correlations between different measurements. It also involves interpreting the data to draw meaningful insights. For example, say you launch a series of live virtual training sessions for employees throughout the U.S. These live sessions take place at 9am on the west coast and 12pm on the east coast. You notice that east coast participants are much less engaged and have a higher absence rate than west coast participants. This type of trend reveals that early morning is a better time for engagement and midday may be problematic for many people’s schedules. If you want more engagement in the live virtual training, providing separate morning sessions to accommodate the differing time zones would probably be a smart adjustment. 
    3. Reporting on data: The final step in the process is reporting key findings in a digestible and informative way. These reports should be shared with stakeholders and provide recommendations based on the data analysis. These can be used to improve learning outcomes and inform future decision-making regarding methodology.

    Tracking data with key metrics gives you a better understanding of the overall performance of your program and allows you to make data-driven decisions on how to improve the program and incorporate learning sciences. 

    Key learning analytics metrics to track

    Learner retention rate

    This metric is the measure of how many learners are continuing with the learning program over time. This rate directly relates to how engaging and useful your course is for the learner. For example, if a course has 1,000 learners enrolled at the beginning and 200 learners remain enrolled by the end, then the retention rate is 20%.

    Learner engagement rate

    This metric measures how actively learners are engaging with their learning environment and content. This can be tracked through such metrics as the amount of time spent on a particular activity or module, the completion rates for activities or modules, and any comments posted by learners on discussion boards. For example, if you have a discussion forum or an interactive component to your course and 50 out of 100 participants regularly contribute to the discussion forum or complete the interactive activities, then the learner engagement (or active participation) rate is 50%.

    Completion rate

    This metric measures student progress and how many learners have completed the course or learning activities. This is usually measured as a percentage of learners who have completed all the required activities or modules within a certain period. While the learner retention rate measures how many people are dropping out of the course completely, this metric measures how many learners are jumping around and not completing all elements of the course, to what extent, and when. 

    Achievement rate

    This metric measures how many learners have achieved the intended objectives or goals of the learning program. This can be measured as a percentage of learners who have passed a quiz or completed an assignment with a certain grade or score. For example, if your course has 1,000 learners and 800 achieve a passing grade of 80%, then your achievement rate would be 80%. 

    Knowledge retention rate

    This metric measures how much of the learning content was retained by learners after completing a course or program. This can be measured by running tests after a program to measure how much the learner retained and remembered. For example, you could send out an email test a week or so after completing an e-learning course to measure the extent to which participants retained and remembered the learning material.

    Time to complete

    This metric measures how long it takes for learners to complete a course, program, or its various elements. This is usually measured in hours or days and is helpful with ensuring the training material is fitting for the learners’ schedules and not too extensive. 

    Four levels of analytics

    Within all the metrics, measurements, and data you can collect, there are various depths that can be attained. 

    1. Descriptive: Descriptive data analytics answers the question “What happened?” by looking at raw data and summarizing it. It’s a collection of techniques and processes that give meaning to collected data, such as identifying patterns, trends, and correlations. This level of analytics examines the present.
    2. Diagnostic: Diagnostic analytics dives deeper than descriptive analytics to answer the question “Why did it happen?” It looks into past data to determine what caused certain events. Through this subset of data science, organizations can uncover hidden relationships between different variables and gain insight into performance. This can help identify problems that need to be addressed and gain insight into user behavior.
    3. Predictive: Predictive analytics uses algorithms and statistical predictive models to look into the future. It examines current and historical datasets to make predictions, such as forecasting sales, predicting user behavior, or anticipating customer needs. This type of analytics helps organizations plan for the future and make more informed decisions.
    4. Prescriptive: Prescriptive analytics takes predictive analytics one step further by providing actionable recommendations on how to act upon those predictions. It evaluates multiple scenarios and suggests the best course of action. This type of analytics can help organizations stay ahead of the curve and optimize their performance.

    Let’s apply these levels to a real-world scenario using a completion rate example:

    1. Descriptive: A report states that the average course completion rate for a specific online course over the last year was 75%.
    2. Diagnostic: Upon noting that the course completion rate is 75%, diagnostic analytics reveal that most of the participants who didn’t complete the course struggled with the same difficult module, or that non-completers tended to fall behind early in the course timeline.
    3. Predictive: Given the course completion rate and the identified patterns, predictive analytics estimate that the next iteration of the course will have a similar 75% completion rate if no changes are made.
    4. Prescriptive: Prescriptive analytics recommend that the difficult module should be broken down into smaller, more manageable sections, or that early intervention strategies should be implemented to help learners who fall behind early in the course timeline. These actions aim to improve the course completion rate in the future.

    Define and track your organization’s learning analytics

    Without tracking and analyzing data, you can’t know if your efforts are achieving the desired results. To ensure you’re making the most of your learning investment, it’s important to define and track your organization’s learning analytics. Here are a few easy ways to start tracking your learning analytics:

    1. Track engagement metrics for your online courses.  Monitor digital learning assessment results. Use surveys to gain learner feedback. 

    These are just a few easy ways to track your organization’s learning analytics and generate learning analytics research. However, if you want to get the most out of your learning programs, using a platform like Docebo Learn Data can be highly effective in helping you understand and interpret the datasets from your courses. 

    See more with Docebo Learn Data

    What are learning analytics like with Docebo? Docebo Learn Data provides powerful reports and visualizations to help you quickly identify areas of improvement. This easy-to-use platform makes measuring, analyzing, and acting on learning data simple. With Docebo’s Learn Data, organizations can understand the impact of their learning initiatives across all departments and areas of business. Leveraging advanced artificial intelligence for data collection and analysis, Docebo’s Learning Analytics platform turns learned data into actionable insights on clear dashboards that drive organizational goals and strategies. Learning analytics are an invaluable tool for organizations to understand how their employees are learning.