Episode 38 - A (pseudo) Randomized Control Experiment about Teaching Effective Learning Strategies

Learning Scientists

This episode was funded by the Chartered College of Teaching, and listeners like you. For more details on how to help support our podcast and gain access to exclusive content, please see our Patreon page.

Listening on the web? You can subscribe to our podcast to get new episodes each month! Go to our show on iTunes or wherever you get your podcasts.

RSS feed: http://www.learningscientists.org/learning-scientists-podcast/?format=rss

Show Notes:

In this episode, Megan and Cindy talk about their massive research project at Washburn University investigating the effectiveness of an intervention to teach first-year college students effective study strategies. The Project is huge, but asks an important question: how can we teach students to utilize strategies that we know are effective on their own to improve learning and academic success. We know these strategies work, we just don’t know a whole lot about the best way to help students learn about them and then transfer the use of the strategies to their own studying.

In a huge pseudo-randomized control trial at Washburn University, we taught half of the WU 101 students (a first-year seminar course) about effective learning strategies. In our first wave of data collection during the 2017-2018 school year, we found that first-generation college students actually had a lower GPA in the intervention group compared to the control, while non-first-generation college students had a higher GPA in the intervention group compared to the control.

NOTE: we have to say this with extreme caution! We need to replicate this finding before we decide to run with it. If we do replicate this during the 2018-2019 school year, future research will need to examine why this happened. Note, we’re not saying the strategies don’t improve learning in first-generation students. Instead, it’s possible that this interaction did something that led the students to perform slightly worse in courses! This could be because the intervention didn’t explain the strategies, or that first-generation students were overwhelmed, perhaps they didn’t know how to use the strategies strategically. There are a lot of potential reasons to investigate. Of course, this result could also be a fluke, and we need to replicate. (And honestly, we hope it is a fluke, but the data are the data!)

While this finding is disappointing, and we certainly don’t want to hurt any specific group of students, this highlights the importance of the research and control groups. It is best to test things out before implementing them widely!