What science says about blended learning

31 March, 2022| Ger Driesen| 6 min read

In 2017, Dr. Will Thalheimer published his findings on the question: does e-learning work? Although his initial question was not about blended learning, his main conclusions were. In general, blended learning tends to outperform e-learning and classroom training; the combination of the two seems to be more successful.

It’s not that black-and-white, though. We’ll get into the nuances after we first take a step back and look at how the concept of blended learning began. Finally, we’ll provide five steps to help you choose the right blend for you.

How we ended up at blended learning

Once upon a time… the major approach to learning was learning in a classroom setting (‘face-to-face’). With the arrival of computers and then the Internet, a new concept called e-learning became increasingly popular. From there, other approaches to learning like ‘learning on the job’, the instruction and practice of skills and performance support, returned to the centre of attention.

With the pandemic, virtual learning took another leap, and now we’re looking at how to incorporate ‘offline learning’ again.

Every year, a new approach is ‘hot’: think about video, the 70:20:10 approach and micro-learning. It makes it hard to choose the best way to go. Attempts to integrate these approaches led to the concept of blended learning (or: hybrid learning).

The effectiveness and benefits of blended learning

Right, so we concluded that blended learning was the way to go.

But is there any proof that it’s actually an improvement going from classroom to blended learning? As mentioned, Dr. Will Thalheimer did the heavy lifting and dove into research on this topic.

Dr. Thalheimer reviewed:

  • five meta-analysis studies comparing e-learning to classroom instruction;

  • six research studies validating the effectiveness of e-learning;

  • six research studies comparing learning methods;

  • ten meta-analysis studies on the effectiveness of learning technologies related to e-learning.

His overall conclusions were:

01. Analysis first...
When learning methods are held constant between e-learning and classroom instruction, both produce equal results.

02. …but otherwise, e-learning performs better.
When no special efforts are made to hold learning methods constant, e-learning tends to outperform traditional classroom instruction.

03. There’s lots of variability, though.
A great deal of variability is evident in the research. E-learning can often produce better results, worse results and similar results to classroom instruction.

04. In the end, methods trump modality…
What matters, in terms of learning effectiveness, is NOT the learning modality (e-learning vs. classroom); it’s the learning methods that matter, including such factors as realistic practice, spaced repetition, real-world contexts and feedback.

05. …and that’s why blended is more effective.
Blended learning (using e-learning with classroom instruction) tends to outperform classroom learning by relatively large magnitudes, probably because the e-learning used in blended learning often uses more effective learning methods.

The best of both worlds

Science tells us that blended learning with effective learning methods is the best choice. But with all the learning interventions and approaches that exist today, it’s more of an art to create ‘the right blend’ of interventions that leads to the best results. Often, some type of e-learning is part of the blend, but not necessarily.

Some professionals suggest that all learning today is, or should be, a blend of different learning interventions to be effective, relevant and inspiring. From a professional perspective, the chosen blend should not be based on what is ‘hot’ at the moment; the personal preference of the learning designer or the design tool that is available.

The right blend should be a logical consequence of a good analysis of the performance issue to be solved and the characteristics and context of the learner.

Five starting points for using blended learning

Analysis first

When we focus on workplace learning, we should first understand the challenges that professionals have while getting their job done. For whom are we designing learning, what are the causes that are holding them back from getting their jobs done at the right level and what is the specific context in which they work? These questions are part of the analysis to start with. Some learning designers are very good at designing and love their job as designers. That is nice but it has an important pitfall.

An analysis is needed before design. A good analysis results in insights into what kinds of interventions to design, be it learning or something else. And if it is about learning, the analysis will give clear clues about which blend of learning interventions is needed to create the desired results.

The best approach available for a good analysis is the ‘Performance Analysis’ based on the work of Rummler and Brache. Another framework to do the analysis is ‘Start with who’, which is based on a design thinking approach.

Apply what works

The analysis gives a clear picture of what kind of (learning) results we need to design for. Often, good learning design will be a combination of elements. Professionals are most effective in their work when they combine the right mindset, skillset and toolset. So, learning is often a combination of learning new knowledge, new skills and new attitudes or behaviours. It is important to be aware of the difference because they all need a different learning approach to be effective.

A good analysis results in insights into what kinds of interventions to design, be it learning or something else.

Make use of innovations

New technologies will continuously offer new opportunities for effective learning approaches and the availability of new ‘ingredients’ will create new opportunities to create optimal blends. It is good to be curious as a learning designer and experiment with new tools available. Also be aware of the ‘bling, bling’ of those shiny new tools. Check their merits via experiments and see what they have to offer. In the end, the decision on if and how to use them in your blend should be based on your analysis: it must make sense and the intervention should deliver the desired result better, faster and cheaper than other interventions.

Blended reality, blended learning

Technology is now part of all aspects of our personal and professional lives. Think about the new technology, the new apps that you started using over the past year. Most people at work today do their jobs in a ‘blended reality’, with both an offline (face-to-face) and an online presence to get their stuff done. The expectations and needs of people change by the availability of all kinds of online tools and apps. They expect the useful tools and apps that are available in their private lives to also be available with the same kind of functionality in their work life and vice versa. It is a logical consequence that the learning situation reflects the work situation: both to be blended.


A good blend of offline and online interventions might save time and money. Think of costs related to travel. Think of costs for the rental of a venue or the costs of maintaining meeting rooms at the office. But also think of a blend that has the opportunity to serve professionals in a way that is ‘right here, right now, right enough’ to support them with their work. Easy access to learning materials and performance support via different devices can also support savings. In many situations, people like to have access to learning via their smartphone while commuting or travelling, at a desktop at work and to prepare for the next day in the evening on the couch via their tablet. Integrating smart technology into the blend to create adaptive learning can lead to impressive savings.

You can see an example of this in the following infographic, which shows the results of a blended course that was created by a Dutch training company on the aNewSpring platform.

Want to learn more about this topic?

Find out more about effective methods for learning. Be sure to read Will Thalheimer’s (PhD Work-Learning Research, Inc. https://WorkLearning.com) article on ‘The Decisive Dozen’.

Check out our blended learning page to continue reading about this learning concept.

[This is an updated version of the 2019 ‘Benefits of Blended Learning’ blogpost]

Learning trend catcher Loves to connect

Ger Driesen

Would like a direct brain-internet connection. Loves connecting people, ideas and inspiration in the global learning community.

Find me on