We had a visit from Melissa Dancy today, to discuss her research, with Charles Henderson, on how and why faculty adopt new teaching practices.
We’ve put in a lot of magnificent effort, she said, to develop innovative teaching techniques that have proven effects on student learning. Education researchers get frustrated, trying to tell faculty that their methods of teaching aren’t working, and showing the data to prove it. Yet most instructors are still teaching by traditional lecture format. Why?
Well, while we’ve done a lot of research on what constitutes good teaching practices, we’ve done very little research on effective dissemination techniques. The intuitive technique, which doesn’t work very well, goes something like this:
- We demonstrate that traditional lecture is ineffective
- We show alternative approaches and data that shows that they’re more effective
- We publish articles and give workshops to get the word out
- We wait for change to happen.
But, it doesn’t. Why not?
We often blame faculty — say they don’t care about their teaching. Well, surveys show that’s not really true. But faculty generally adopt new teaching strategies based on both personal and structural factors. They modify them according to their tastes and situation.
Melissa, along with Charles Henderson and Chandra Turpen did surveys and interviews of 22 faculty to identify what it is that they want to do in their class. The good news is, education researchers’ dissemination efforts have indeed been effective. Most faculty know about research based instruction methods:
- 87% know about at least one of the strategies on their list
- 50% know 6 or more strategies
- 48% use at least one
- 70% want to use more research based strategies
However, they aren’t always using these methods in ways that are consistent with what we know causes learning gains. For example, 41% of peer instruction users change something about the method, and only 28% of instructors have students discuss the answer during the question — a key feature of peer instruction!
Not surprisingly, most people cited lack of time as a reason for not using more research-based techniques. The factors that were most predictive of using research based techniques were:
- Attending new faculty workshop
- Attended workshops related to teaching
- Satisfaction with instructional goals
- Reading journals related to teaching
- Being female (!)
But, people who had high reearch productivity or large class size were just as likely to use these research-based techniques, which flies in the face of some of our assumptions about hard-core researchers being less interested in teaching.
Let’s take peer instruction as an example. Where did people find out about it? They would usually find out about peer instruction from a colleague, and decide to try it. They would then read materials to find out more. However, we usually put out all these materials in the hopes that people will read them and decide to try it based on those materials, but it seems that most people go to the written materials as a second step. Similarly, they use the effectiveness data as a second step — to justify to others why they’re doing what they’re doing, and to bolster their own confidence that they’re making the right choice. They don’t use that data to convince themselves, however.
So, we might conclude:
- Dissemination should focus on methods that involve personal contact with faculty
- Written materials should be geared towards helping faculty learn more, rather than convincing them
- Data should be presented to bolster their decision, rather than to convince them
So, dissemination is a poor model of creating change. We can’t just lecture people about making more interactive teaching!
We have to move towards a more effective model of change:
- Building communities to create social connections. For example, the Modeling curriculum has been very effective, in part because they built a community first — they give ownership to the teachers, who teach the modeling workshops.
- Provide materials in a modifiable form, and support instructors in making effective modifications
- Support instructors as they implement it, rather than just giving it to them at the start
- Do some research on the barriers that instructors face in implementation
- Get into policy to create institutional opportunities for change