10 Questions to Will Thalheimer

Training design can be tough. In my world there are at least three components of modern training: substantive content, instructional design and technology. Today I would like to give you few tips in the field of instructional design and myths around it. Please welcome – the scientist, the myth debunker, the president of Work-Learning Research, Inc., and the very wise guy – Will Thalheimer!

Tomasz Jankowski:

Hi Will! I am so happy to have you here! I would like to introduce your work to L&D people in Poland. Please describe what do you do?

Will Thalheimer:

Tomasz, thank you for the invitation! For the past 18 years, I’ve been doing research-based consulting in the workplace learning field. I translate research from the scientific journals on learning, memory, and instruction and use what I learn to help my clients.

I help people build more effective learning interventions, measure them effectively, and take a strategic look at their learning-to-performance efforts.

Tomasz Jankowski:

Why L&D people, including trainers, learning designers, instructional designers, HR and managers, should listen to what you have to say?

Will Thalheimer:

Because I’m gorgeous and charming and a good family man—and nothing like my country’s political leader.

In short, I’ve spent years compiling research, looking for practical wisdom, and sharing this with all types of organizations, from multinational companies, government organizations, universities, and elearning companies.

Tomasz Jankowski:

Thank you for this answer :) Let me give you a short task, maybe even entertaining one for you! Please find a list of learning myths and nonsenses below and prioritise the list from the most harmful to the least harmful.

  1. Learning styles
  2. Dale’s Cone of Learning with % (people remember 10%…)
  3. Presentation of learning objectives to the learners
  4. People forget 90% of what they learned after 7 days
  5. 70:20:10 model and 10% formal learning
  6. We only use 10% of our brains…
  7. MBTI is valid
  8. Elearning is less expensive than classroom learning
  9. Feedback to learners should always be immediate

Will Thalheimer:

Tomasz, you are a tough taskmaster. Ranking these is too hard… I will comment instead — super briefly — on each one.

  1. Learning styles. 
    This is the idea that we can determine what kind of learning style a person has and then give them specific learning methods to help them learn. This idea has been debunked (shown to be false) by many reviews of the scientific literature. First, there are too many different ways to characterize people. Second, most learning-styles diagnostics are bogus. Third, using learning styles has not been found to be useful in supporting learning and can be harmful. Instead of focusing on learning styles we should focus on the many research-based learning factors that are proven effective.
     
  2. Dale’s Cone of Learning with % (people remember 10%…) 
    For years, this has been the most popular blog post I’ve written. Everyday dozens of people flock to my post to find out that it’s not true that people remember 10% of what they read, 20% of what they hear, 30% of what they see, etc. Indeed, recently three academic researchers and I wrote a series of articles about how this falsehood came to be. It has been passed down starting in—I forget the exact date, but it was something like 1913… and the internet has accelerated the dissemination of this falsehood.
     
  3. Presentation of learning objectives to the learners 
    This is a long discussion. I recommend my video to folks interested. Basically, we in the field completely misunderstand how learning objectives work when we present them to learners. If they’re well-written, they guide learner attention.
     
  4. People forget 90% of what they learned after 7 days 
    No. No. No. People forget at different rates depending on many factors. And rarely, when things are completely extraordinary, people may not forget a thing. For example, I’d bet that anyone reading these words will remember well into their old age that I am a gorgeous family man. Seriously, here’s what we should remember. People forget! We need to help prevent them from forgetting by using such learning factors as retrieval practice, spaced repetitions, and simulating workplace situations. Sometimes, we should forget about helping people learn and remember and instead provide them with job aids, performance support, or other mechanisms that directly prompt performance. People forget at different rates depending on many factors. See this research review.
     
  5. 70:20:10 model and 10% formal learning 
    These models, which emphasize the importance of on-the-job learning—in comparison with formal classroom training or elearning— send us a reasonably powerful message, but rely on completely bogus research. There simply is no good way to tell how much people learn while working on the job… and any attempts to ask them are just silly exercises. On the other hand, people certainly do learn on the job—and, if we can leverage that learning, we have an opportunity to drive performance improvement. The key question is, can we leverage on-the-job learning in such a way that the benefits outweigh the costs.
     
  6. We only use 10% of our brains… 
    Anyone who has read this far is clearly using only 10% of their brains. Everybody else uses 100% of their brains.
     
  7. MBTI is valid 
    The MBTI, DISC, and most other personality measures—unless they are based on the scientifically-vetted Big Five inventory—are typically NOT valid, and NOT reliable. There is no doubt in the scientific literature about this. The MBTI and DISC are NOT GOOD.
     
  8. Elearning is less expensive than classroom learning 
    eLearning can be less expensive than classroom learning and it can be more expensive. Obviously, when development costs are held reasonably constrained and these costs can be spread across thousands of learners, then savings will result. On the other hand, the research shows that overall, organizations don’t usually gain a cost savings—the research suggests that even when elearning saves money, that overall budgets do not go down.
     
  9. Feedback to learners should always be immediate 
    Sometimes feedback should be immediate. Sometimes it should be delayed. It’s complicated, but here’s a useful heuristic: When someone is learning something, they probably will benefit from immediate feedback. They’ll certainly need feedback before they make another attempt (so they don’t reinforce poor skills or incorrect knowledge). However, once learners fully comprehend something, then we probably want to delay their feedback because such spacing improves long-term remembering… Folks can read my general report or research report on feedback.
     

Tomasz Jankowski:

How do you debunk learning myths and what do you think about this: https://www.skepticalscience.com/docs/Debunking_Handbook.pdf?

Will Thalheimer:

Debunking myths is a hobby. Fun but doesn’t pay well. :) But seriously, speaking truth to those who spread untruths is important. Sometimes they don’t know they’re spreading misinformation. Sometimes they need to know that someone is watching.

There’s research to show that in some instances trying to persuade true-believers can backfire, making them more strongly believe in the untruths they espouse. But this research has its limitations. For example, it doesn’t tell us what happens when we debunk folks who aren’t strong believers. It also doesn’t tell us what happens when we use creative methods of debunking or we debunk continuously or multiple sources of debunking are delivered to someone. Cook and Lewandowsky’s handbook of debunking is very good, but it’s not the end of the story.

Tomasz Jankowski:

How should learning designers navigate through the ocean of learning theories? How to determine which ones are good for learning design?

Will Thalheimer:

There are 4,273,891 (plus or minus – :)) learning factors that have been suggested since 1900. And, like you say, there are many theories too. Learning designers should focus on the most important, most fundamental learning factors first and ignore the regular fads that hit our industry. Focus on things like repetition, retrieval practice, spaced repetitions, feedback, variety, etc. See for a review of 12 critical factors. Finally, forget the damn theories! Focus on learning factors.

Tomasz Jankowski:

What kind of commercial services do you provide to L&D market?

Will Thalheimer:

I do keynotes and speaking; workshops, consulting, learning audits (where I help organizations see how their learning initiatives are aligned with research-based best practices), and I do evaluation work based on the wisdom spelled out in my book Performance-Focused Smile Sheets (http://SmileSheets.com).

Tomasz Jankowski:

Please describe the most tough customer you have ever had. What was your biggest commercial challenge to face?

Will Thalheimer:

I had one jerk of a customer once. Maybe I’m lucky, but that was the only time I was mistreated. And funny story, this person eventually got fired and had the audacity or blindness to ask me for help in their job search.

You ask about commercial challenges. Well, I’ve successfully survived as a research-based learning consultant for 18 years on my own. It’s not always an easy journey, but it is what I want to do.

Tomasz Jankowski:

Which resources do you recommend as compulsory? I mean resources so important, that without understanding them no trainer or designer should professionally design learning for others?

Will Thalheimer:

If you’re in the learning field you ought to know how learning works. You don’t have to get a PhD, even though that can help, but you ought to have done exhaustive reading on human learning. To see a list of great books to read that are research-based go to The Debunker Club website and look down the right side of the page (http://Debunker.Club).

Tomasz Jankowski:

How neuroscience affects and will affect elearning in the nearest future? Therefore, what do you forecast about nearest future of L&D?

Will Thalheimer:

I’ve been invited to speak on neuroscience and learning many times over the last year. The bottom line is that neuroscience, though it has tremendous promise for the future—has almost nothing UNIQUE to teach us now, that we didn’t know from behavioral research before. See my blog post to learn more.

Tomasz Jankowski:

I hear your book, Performance-Focused Smile Sheets: A Radical Rethinking of a Dangerous Art Form, has just won an Award of Excellence from the International Society for Performance Improvement. What is your book about and why has it been deemed so important?

Will Thalheimer:

I’m grateful to ISPI for recognizing my book. The book provides a radically-improved method for getting feedback from learners. For years and years we’ve used questions that tell us very little about learning effectiveness. In fact, two meta-analyses (scientific articles that together compiled results from over 150 scientific studies) showed that our learner-feedback approaches (smile sheets, happy sheets, reaction forms, level 1’s) are virtually uncorrelated with learning results!

My book provides a method that gives us hope for improved evaluations. Indeed, in a recent pilot study 92% of learners like these new kinds of questions better than typical Likert-like questions or numerical-scale questions!

For those interested in taking a look at a free chapter, the book’s website is: http://SmileSheets.com.

Tomasz Jankowski:

Thank you Will. I really enjoyed your answers!

Will Thalheimer:

Thanks Tomasz!

The Summary:

The domain of instructional design evolves. The research results guides us on what is effective and what is not. Why not take it into account when you design a training? The knowledge of effective training strategies is out there, you just need to search, read and apply it. But be careful and avoid the bad myths! Be more critical while choosing a pattern, a model and a strategy. Choose wisely! And if you are responsible for training programs in your organization – do you consciously choose a training or an e-learning provider that uses reliable design methods?

Sources: