I was reading a recent Will Thallheimer post about conversational writing where he referenced a great post by Kathy Sierra that says exactly what I’ve been feeling for a long time:
If you want people to learn and remember what you write, say it conversationally.
She actually references a 2000 study from the Journal of Educational Psychology that showed:
“In five out of five studies, students who learned with personalized text performed better on subsequent transfer tests than students who learned with formal text. Overall, participants in the personalized group produced between 20 to 46 percent more solutions to transfer problems than the formal group.”
This has been something that I have believed in for many years, but because I don’t have a formal Instructional Design background, I never said it very loud. When I worked on developing learning content we preferred doing scenario based learning, which placed the learner in a situation that was relevant to their jobs. Because this was usually done through a story, it would take on a conversational tone. The goal was to involve the learner and make them understand how the information was relevant to them, and what difference it would make on the job. Now at least I have some solid data to defend my position.
This is one of the reasons I like reading Will’s blog, because he focuses on research that provides actual data to back up his thoughts and opinions. While I was looking around his Work-Learning Research site I was horrified to find his article on the bogus learning chart that has been circulating, and realized that… I had used it myself!
I do believe in the basic concept of the graph, but now that I think about it, it is really odd that the numbers come out evenly. That should have been a tip off. I can’t even find the talk where I used it, but I know I was part of the problem.
So, to sum up: Informal Writing is Good, Informal Data is Bad.
Using an informal or conversational tone is effective for learning, but becoming so informal that you never bother to validate your data or sources needs to be avoided.







