It’s pay day. You decide to treat yourself and visit your favorite shopping app. Even before you search for what you want, you are presented with “styles you’d like,” curated just for you. You feel special — finally there’s someone who knows what you want. The same happens when you use a food delivery app to order dinner. These platforms make our lives simpler by remembering our preferences and offering them back to us.

Welcome to the age of big data. You tap a feature, and it is bookmarked. You tap many more, and you have a personalized set of information aligned to your preferences.

What if learning were this personalized and easy? What if your learning management system (LMS) understood each learner’s preferences and then curated his or her curriculum accordingly?

Learning Analytics for the Win

The term “learning analytics” refers to measuring, collecting, analyzing and reporting data about learners, their learning experiences and their learning preferences in order to improve training. For example, if learners refer to one content source more often, or if they have difficulty answering one type of question, the system can provide targeted support and courses accordingly.

Imagine you are back in school, and there’s a teacher whose class you don’t follow well. One day, he asks the whole class for feedback. He asks what your expectations are from the class, which topics you most enjoy learning about and which teaching methods you struggle to engage with. The following morning, the teacher acts on the feedback. He changes his teaching method, and your comprehension level improves dramatically.

To extend this example further, imagine that he does not seek your feedback directly but, instead, observes you closely and learns when you’re attentive and when you aren’t, which assignments you enjoy the most, and which ones you struggle with. Then, he continuously improves and fine-tunes his performance accordingly.

Analytics can help make this learning experience a reality at work: The technology observes employees’ learning patterns and curates training accordingly. In other words, while employees learn, the learning platform is learning about them!

The late Belgian computer scientist Erik Duval wrote, “Learning Analytics is about collecting traces that learners leave behind and using those traces to improve learning.” Let’s explore how to analyze these traces and use them in training.

Social Network Analysis

Social network analysis (SNA) involves “eavesdropping” on social networks to collect the information, data and artefacts learners share with each other. This process helps us understand how people communicate and, consequently, to design content in a way that facilitates collaborative learning. SNA considers relationships among people, artifacts and the settings in which they learn and maps these relationships to understand how people communicate and exchange information.

Attention Metadata

With increasing numbers of learning resources that are available online and offline, attention and focus on each resource are reduced. Attention metadata focuses on the interactions that learners have with the objects presented to them. Metadata could include clicking a “help” button, using a share or chat feature, or even switching tabs. This data represents how often the learner drifts away from the course, when the learner needs help, and how long he or she stays in the training program.

Discourse Analysis

Discourse analysis dissects the language, both verbal and non-verbal, learners use to communicate, including participating in discussion forums and sharing content during or after a training session.

Early Warning: Risk Identification

Learning analytics can help us predict under-achievers. If a learner is struggling with some topics, the system can pick up on their struggle and offer additional guidance, supplementary learning materials or online peer learning to help them.

Why Learning Analytics?

Now that you understand types of learning analytics, here’s how they can help you design a better curriculum:

  • Monitoring individual and group progress.
  • Identifying when learners are ready to move on to the next topic.
  • Discovering patterns: What mistakes are learners making frequently? How many times does a learner refer to a resource to answer a question?
  • Finding early indicators of success or failure to provide support.
  • Assessing the usefulness of learning materials.

In addition, learning analytics can help you answer questions such as these:

  • What are learners’ current skills?
  • What skills do they need to excel in the training?
  • Which elements of the course are learners struggling with?
  • Which sections of the course engage them the most and the least?
  • What prompts them to ask questions?
  • How are they navigating the interface and engaging with the resources?
  • What mistakes have they made?
  • Are there any issues related to comprehension, decision-making, accessibility, etc.?
  • If your training program is blended, which intervention do learners prefer?

Data Privacy

Finally, a word about privacy: Learning analytics works by gathering detailed data about learners, storing them in a database and then analyzing the data to make predictions. Considering the noble intention of this activity, few should have problems with learning analytics.

However, there is a fine line between collecting harmless learning data and collecting confidential information about the learner. As data privacy becomes a hot topic, discussed and debated across the world, and as governments enact legislation to protect individuals’ right to their data, learning analytics becomes a controversial issue.

Therefore, it is important to consider what the law of your learners’ land says about collecting, storing and processing employee data. It is good practice to inform learners of the system’s intention to collect their learning data and to explain how you will use the data to enhance their learning experience. Always obtain their consent before proceeding.

The Need of the Hour

Technology plays an important role in learning analytics. Analytics and big data can help you engage learners and improve comprehension — but only if you have the technology to support them.