We’ve all seen some sort of numbers or graphics depiction about how we retain what we learn. The story goes that we retain 5 percent of what we see/hear, 10 percent of what we read, 20 percent with a visual, 30 percent with a demonstration, and so on. The problem is – this is a fabrication. The root source of this information is attributed to Edgar Dale and while the cone of learning – where the hierarchy is covered — is his, it didn’t have percentages on it – and he cautioned about overly generalizing its use.
So why do we continue to see these set of numbers? Most likely the problem is really that there hasn’t been good research studies on the effectiveness of different delivery modes in education. There are some good reasons for this since changing the delivery mode means redoing the instructional design, and in doing so doubling the work. Further, instructional designers will have more experience with some modes than others leading to greater effectiveness at some forms of instructional design – and ultimately delivery. That means that the instructional designer themselves may bias the results.
The other aspect is that the materials change so there’s no good way to do a direct comparison of effectiveness between two different modes – even if both modes are created by the same instructional designer. That’s bad news when you’re trying to create a reliable study of how things differ because you have to eliminate as many extraneous variables as possible.
Of course, you have to deal with the fact that different kinds of content are more conducive to some delivery modes than others – try teaching someone how to ride a bike by writing it in text only. Try to teach someone how to do math without written text. So the type of message being conveyed must be isolated.
The other key challenge is the students themselves. Students are necessarily unique in their experiences and their ability to consume information via different delivery modes. Some students learn perfectly well with an abstract conversation and no pictures. Other students require the ability to “play” with the problem via models, pictures and simulations. A study to quantify the differences in average effectiveness of the medium would require a very large sample set to eliminate all of the biases in the instruction, the student, and other factors not considered here. (I’ve not fully thought through all the potential biases.)
The problem is that since we have no basis for knowing what modes of training work better, we have very little refinement we can do on our educational approach. When you’re looking to understand learning retention rates as it pertains to different modes of delivery, you’re stuck with guesses – or creating multiple versions of the same content and doing your own testing. Doing your own testing of which mechanisms are most effective isn’t something that many training groups have the luxury of doing.
Help from the Dark Side
The pollution of marketing messages into sources of perceived editorial content has made the marketing groups at organizations seem to be the enemy of training. Marketing is often disguised as training – like a wolf in sheep’s clothing – in order to increase its effectiveness. However, despite the natural and understandable aversion to marketing, the goals of marketing and training are strikingly similar.
In training, we’re trying to change behavior or to create new habits. We’re trying to enable better decisions or reduce mistakes. In short, we’re trying to influence how others behave. (Influence sounds better than what we’re really trying to do, which is exercise control.) Marketing has an almost identical goal. They’re less concerned with habits and more concerned about an immediate behavior – a behavior to buy – but ultimately it’s much the same thing.
The good news here is that there’s a very strong financial incentive to create good marketing. If we can generalize some of what marketers know about how they can drive prospects to action then we should be able to adapt those to help drive behavior change in students.
Marketing organizations long ago began to realize that personalization was necessary to get better effectiveness with their ads. Books like The One-to-One Future called out the need for personalization in marketing back in 1996. Recent studies have shown that personalization is an effective way to increase response rate. In fact, the retailer Target was recently identified as using analytics to identify potentially expectant mothers and sending them targeted coupons – before their families were aware. That level of awareness and personalization pays off.
This is consistent with what we know of Andragogy (see Training Search to be Your Adult Learning Hero) where we’re encouraged to consider what students bring to the learning experience and adapt the methods used to match their strengths and weaknesses.
Another strategy in wide use by marketers today is multi-channel marketing. That is where messages are spread across different communications channels to improve the response rate when compared to any individual channel. So too should we consider training to not just be an ILT course or a CBT course, but rather an entire environment that encourages and supports training. This should include tools to allow the employee to self-identify their need for information and for the internal systems and tools to help point the student to their best source of the information.
Finally, marketers know that they’re much more effective at driving action when the consumer is “at the moment” of being able to make the decision. The more disconnected in time between the target seeing an ad and their ability to take action dramatically decreases the response rate. So too we need to look for strategies that enable students to immediately use the information that they’ve learned. A study of “The Relationship of Time and Learning Retention” by Kamunch and Ledman showed that the correlation between performance in courses fell as time increased – leading to the conclusion that there’s a non-linear decline in information retention the further away the courses were from one another.
New Learning Retention
Although there are precious few studies that have focused on learning retention rates when the delivery mode was changed, there are some things that we do know that influences learning retention rates. We know that the time between the person learning the information and using it is important – thus why CBT can offer an advantage over scheduled ILT. By borrowing from marketing studies we know that we can drive more behavior change the more personalized we make the instruction and that we can get better response rates by utilizing multiple models for delivery. It’s not a question of either-or – it’s a question of either-and.