Videolesson: Interpretation of learning analytics

Watch the video lecture with examples of interpreting learning analytics from different information systems.

 Transcript

	Welcome to this video lesson on interpreting learning analytics. Learning analytics provides valuable insights, but only if we know how to properly interpret them. Teachers play a key role in interpretation and in implementing interventions based on insights from learning analytics. As an introduction to this topic, this lesson presents several specific examples. Learning analytics can be available within learning management systems, based on the data collected in those systems. One example is Moodle, which includes a range of learning analytics features from basic functionalities to plug-ins and integrations with external tools. In Moodle, for example, we can see activity reports, logs, activity completion, and other reports. When it comes to assessment, we can also view analyses of quiz results, which are covered in more detail in the reading materials. If we look look at the activity report for each activity in the course, we can see how many times it has been viewed and how many users have accessed it. If students are not using a particular resource, it is worth considering why. Perhaps it is not visible enough and needs highlighting, or perhaps it is too complex and requires additional explanation. However, low usage does not necessarily mean that a learning outcome has not been achieved. Just as high usage does not necessarily mean that it has. Assessment results are crucial here. This is why different types of data and analysis should always be considered together. We can also view student activity over time in a course. Increased activity during certain periods can often be linked to summative assessment such as midterms, exams, or project preparation. To better understand such patterns. It is useful to compare LMS analytics with learning design and the corresponding analyses in the BDP tool. For example, we can examine the distribution of activity types within a course. Based on this, we may decide to increase the proportion of activities that require more active student engagement. Alternatively, if the most prominent learning types are investigation and practice, as in this example, we may conclude that much of the learning takes place outside the LMS and is therefore not captured in activity reports. The BDP tool also provides a wide range of other useful learning design analytics. Particularly valuable are analysis related to student workload, the application of innovative pedagogies, constructive alignment and assessment. For instance, from the display shown on the screen, we can see that the planned student workload based on learning activities does not fully match the official workload assigned to the course. Taking this into account, the teacher can revise the learning design to better align activities with workload expectations. From the distribution of learning types. We can also determine whether the intended teaching and learning strategy has been achieved. For example, if the goal is problem-based learning analytics showing a high proportion of investigation and production activities suggests a meaningful design. However, if acquisition-type activities dominate, this signals a need to revise the learning design to better reflect innovative pedagogy and to increase activities that support active learning, such as problem-solving in team work. From analytics in the BDP tool, we can also see how learning outcomes are covered across course topics and assessment. If we see that a learning outcome is not fully covered or not covered at all in course topic, or is not accompanied by appropriate formative and summative assessment, this is a signal that learning design should be improved in line with the principle of constructive alignment. Based on learning analytics and learning design analytics, teachers gain insights that can help them design interventions for students, for example, further emphasizing certain activities or revise learning design, for example, strengthening the link between assessment and intended learning outcomes. For meaningful interpretation, it is essential that teachers develop data literacy, meaning the ability to understand, interpret, and use data. It is also important that they apply their pedagogical knowledge and critical thinking, understanding the educational context and individual context of each student, and to ultimately act in accordance with ethical principles when making decisions. Educational institutions play an important role in fostering a data informed culture and promoting the meaningful and ethical use of learning analytics.