Training can’t solve everything.

Guild Master and consultant Julie Dirksen of Usable Learning wants instructional designers, managers, and others engaged with creating and using eLearning and other corporate training and performance support to internalize that essential message.

When managers want to see behavior change—whether it’s getting learners to start doing something new, stop an unwanted behavior, or apply a new procedure to an existing task—they often turn to training. But when behavior doesn’t change? “A lot of times it isn’t necessarily a training problem,” Dirksen said. She’s specifically referring to situations where people know what to do or how to do it, whatever the new or desired behavior is—whether it’s following hand-washing protocols in a medical office or providing effective feedback to direct reports—but they are either not doing it or doing it ineffectively.

“These are complex human behaviors; they need a whole lot of analysis—and there are some really good models out there for how to structure that analysis so that you can feel that you’re being thorough in looking at the problem and breaking it down.”

Dirksen mentions the behavior change wheel, based on the COM-B model, out of University College London, as a good diagnostic model. COM-B looks at capability, opportunity, and motivation; behavior is an interaction of these elements; behavior change requires a change in at least one of these elements. 

If it’s not a training problem, what is it? According to Dirksen, the problem could be:

For example, in the case of hand-washing protocols: “One of the things when they teach hand-washing classes is, they tell you to talk to other people when you see mistakes being made. They tell you to talk to your co-workers if you see that they’re not washing their hands,” Dirksen said. “Nobody wants to do that!”

A protocol like that, fraught in any setting, can become impossible where “there is a power differential like nurses talking to doctors,” and a lot depends on the culture of the organization, she said. It’s also a question of what the training covered. If people are taught the procedure and told to follow it—and they are told to have those conversations with colleagues who didn’t follow the procedure, information-based training is not enough.

“That’s not something that you’re going to get people to do just by saying, ‘Hey, it’s really important for you to do this.’ The approach that you would have for that would be really different. You would do something like actually practice and role-play having those scenarios so that people get a little bit less scared by the idea of talking to somebody about their hand-washing. They would have a set of gentle and non-judgmental explanations and a way of doing it that they feel comfortable with,” Dirksen said. “That’s a case where I’m pretty sure that a knowledge-based approach of just telling people it’s important isn’t going to make a significant dent in the behavior. You’re going to need to actually give people good tool sets for those—and they do exist. There are different kinds of models for having different difficult conversations.”

Effectively diagnosing the problem helps instructional designers understand what can—and cannot—be fixed by training. This, in turn, helps set appropriate expectations for eLearning and can turn the focus to solving the actual problem. In this case, the real issue was learners’ anxiety or discomfort over calling out co-workers for their noncompliant behavior.

Other times, the problem might be misaligned incentives or motivation. Dirksen cites another example: “I was working with an insurer; we were working on multinational policies. Insurance policies that have to cover several different countries are really complicated because you have to deal with the laws in all the different countries,” she explained. While she was working on the training, she was asked by the client, “Can we do something that really emphasizes the importance of entering the information accurately? I don’t think our data entry people are always as careful as they should be about entering the data accurately.” That was a clue. “I said, ‘Sure, we can do that, but before we do, why don’t you tell me a little bit about how they get feedback, how their incentive systems work?’”

It turned out that the incentive system rewarded the data entry staff based on how many applications they entered per hour. Just as Wells Fargo’s emphasis on tellers opening numerous accounts for each customer led to tellers’ and managers’ deplorable—even criminal—behavior, incentives that rewarded data entry volume rather than quality led to sloppy work. “That’s not a training problem. That’s very emphatically not a training problem,” Dirksen said. “That’s a case where it really is the misaligned incentives. But I think we often, in training and development, get handed some of these kinds of things with the expectation that we’ll magically fix it through training.”

Learn more

Julie Dirksen will present a Pre-Conference Certificate Workshop on March 21, “Design for Behavior Change,” and two conference sessions: “Diagnosing Behavior Change Problems” and a panel, “What’s Wrong with Evaluation?” at Learning Solutions 2017 Conference & Expo, March 22 – 24 in Orlando, Florida.