Design Tip: Productive Failure

Most traditional learning experiences end with some kind ofassessment. But what about beginningyour next module with a test? If it sounds like you’re setting your learners upto fail, you are. And, believe it or not, that’s a good thing.

Recent research has found that struggle during learning—called“productive failure”—has significant long-term benefits.

Failing forward

A recent study from UCLA explored what happens when learnersare thrown into the deep end. The researchers conducted the study in an introductorypsychology course. They began some classes with a pretest—a multiple-choice assessmenton the material students were aboutto learn. In other classes, they skipped the pretest and went straight to thelecture.

As you might expect, the students struggled on the pretests.But on the comprehensive final exam, students scored 10 percent higher on the pretestedmaterial.

A separate—and more extreme—study of learner failure, alsocarried out in a real classroom, reinforced these findings.

Math students were divided into two groups. One groupreceived traditional math instruction—a lecture with instructor-led examples,followed by homework. The second group was simply given math problems to solvewithout any instruction at all. If that wasn’t bad enough, the researchers alsowrote the problems in an intentionally confusing style.

The “sink or swim” students fumbled through the problemsweek after week until finally, right before the exam, they received a lecturethat explained all the key concepts.

How did they do on the exam? Again, they scored 10 percenthigher than the traditional instruction group.

What the researchers did next is even more interesting. As afollow-up, they conducted a second test—this one on an advanced math topic thatneither group encountered before. Half the productive failure group passed thetest, compared to only 21 percent of the traditional instruction group. The experienceof failure and struggle gave students something valuable.

Implications

It’s easy to assume that, when it comes to eLearning, failureis the last thing you’d want to intentionally design for. Failure could lead tolearner frustration, disengagement, or abandonment. But the research suggeststhat asking learners to experience a little struggle has clear benefits.

Researchers offer several explanations for why failureworks. First, it familiarizes learners with what they are about to learn. So,when trainees are exposed to the key concepts, their brains are already primedto receive it. Second, it reveals what learners don’t know, ensuring that theydon’t go into a learning experience with false confidence, thinking they’vealready “got it.” And third, struggle during learning may increase persistence,which is important when learning complex job skills—learners likely won’tperform them correctly the first time.

Application

Here are some suggestions on how to apply the research inyour eLearning design.

Start with a pretest. Just like the study, ask students toanswer a few multiple-choice questions upfront. If you’re worried about theirself-esteem, assure them that their grade won’t count. Pretests inspire deepthinking and problem solving that will benefit learners’ knowledgeretention. 

Start with a writing exercise. Ask learners to write abouttheir existing knowledge on the topic. This will prime them for the learningexperience to come, reveal their knowledge gaps, and cause them to evaluatetheir current behavior, which likely can improve.

Finish by asking learners how they will apply what theylearned. At the end of a module, employ an open text box or an audio or videoresponse to capture learners’ plans for applying the new concept. Thinkingabout the real-world application will help them to distill the key ideas and deeplyconsider the changes they’ll be making in their workplace behavior.

Sources

Carey,B. How We Learn: The Surprising Truth AboutWhen, Where, and Why It Happens. New York, NY: Random House, 2014.

Kapur,Manu, and Katerine Bielaczyc. “Designing for Productive Failure.” Journal of the Learning Sciences, 21(1).2012.

Kapur, Manu. “Productive Failure in MathematicalProblem Solving.” Instructional Science,38(6). 2010.