Learning analytics

How is learning analytics useful?

Researchers and software developers are already using learning analytics to create bespoke learning solutions and tailor the content of their training to suit learners’ needs.

In this article, we look at some of the benefits of analyzing the learning process.

1. Adapt training to learner profiles

Learning analytics can be harnessed to determine a profile for each learner and identify potential failure scenarios for each profile category. The aim is to align any training support with learners’ needs. In this approach, unnecessary or redundant elements are removed and additional measures are taken to overcome sticking points.

2. Improve training content

Perhaps the most important step is the analysis phase, which makes it possible to challenge the course design and identify areas for improvement in order to optimize trainees’ learning journey. In addition, this analysis provides the basis for establish correlations between performance indicators such as course length and success rate.

More generally, learning analytics can be harnessed to identify the training that delivers the best results (in performance terms, for example), and conversely, the courses that provide the least added value for learners or are the most costly for the company or training center.

3. Consider ROI

Implementing precise analytics of training courses makes it possible to optimize course lengths, adjust training content to suit learners and obtain accurate results for each course or training module. Precise performance indicators can also be used to evaluating training costs and ultimately measure the return on investment of each item in the training catalog.

4. Anticipate problems

Analyzing training data yields insights that make it easier to anticipate potential problems and challenges. Some e-learning course designers are able to adapt their training on the fly. Based on feedback from learners during a training session (for example, in the light of a satisfaction survey), specific components of trainees’ personal learning journeys can be modified, adjusted or removed.

Examples of performance indicators analyzed in training contexts

Raw or descriptive analysis has little value for trainers, the company or learners. The most important thing is to identify the factors responsible for the results obtained and define areas for improvement in order to anticipate potential problems and improve the learning experience.

We have compiled a short list of performance indicators that can be applied and analyzed during a training session. Naturally, the list is not exhaustive and can be enriched as needed by companies or training centers:

• Participation rate: What percentage of employees/students/trainees took the training? It is important to also consider why this population attended the training. This indicator provides insights for appraising the benefits and appeal of a course.
• Dropout rate:What percentage of employees/students/trainees failed to complete the training? What were their reasons for dropping out: complexity, lack of interest, module length, etc.? This indicator is very useful for determining when learners are no longer able to follow a course, and identifying a suitable alternative learning format. For example, adding tutoring to an e-learning training session can raise the completion rate by multiple percentage points.
• Session length: On average, how long does the learner stay logged in to the platform? Is this average increasing or decreasing? This parameter can also be used to measure the course’s appeal and adapt each trainee’s personal learning journey.

These indicators help to understand what learners appreciate or dislike. The trainer can use these insights to rethink course formats, for example by including new, bite-sized activities (fast learning) or more fun activities.

What are the characteristic features of learning analytics?

Learning analytics are meaningful only when there has been a prior interaction with the learner.

For example, learning analytics can provide information relating to:

• The completion rate of a module,
• The time taken by the student to finish it,
• Mistakes made by the student,
• The number of views recorded for the module,
• The student’s digital environment

What are the pros and cons of learning analytics?

Learning analytics can be used to continuously improve the content of a training course. These tools are an excellent source of information relating to the execution of a training course, and provide an objective foundation on which to build. Some critics point to learners’ right to privacy. Learning analytics data is communicated exclusively to the trainer, who uses it to improve quality and learning methods. Permission to collect data must be sought at the start of the course.

In its 2021 white paper, Digital Learning in Figures, ISTF explains that “65% of training programs that include tutoring have a completion rate above 60%”. The figures for untutored training are abysmal. Half of all learners have only a 10% chance of completing the course. These figures highlight the importance of working with tutors. The statistics were obtained using learning analytics.