My group members, Jason Barclay, Jennifer Ellis and I conducted a research study comparing the effects of online (asynchronous) and face-to-face (synchronous) discussion on student learning outcomes in a middle school mathematics environment.

To begin this study we reviewed the relevant literature about learning in synchronous and asynchronous discussions. Based on this information we chose to use a Moodle forum for the online discussion, and a fishbowl-style Socratic seminar for the face-to-face discussion. We also used the literature review to select the Interaction Analysis Model (IAM) discussion-scoring rubric.

Due to convenience, we chose eighth grade students in my math classes as the subjects of our study. We pre- and post-tested students so that we could measure the effect of a discussion intervention on learning outcomes (or lack thereof). Finally, we analyzed the testing data and categorized discussion responses in order to draw conclusions about the effectiveness of each type of discussion.

In our study, mean scores actually dropped from the pre- to the post-test. Using a paired t-statistic, our results were shown to be significant. This seemed counter-intuitive because the scored-discussion responses showed that most students were able to demonstrate some amount of learning. This led us to conclude that the timing of our study may have been a large factor in the lower post-test scores. We conducted this study over a two-week period at the end of the 2008-2009 school year. Apathy among students was a major problem. While they were excited to take part in the beginning of the study, they were more concerned about final exams and graduation at the end. This was evident in their discussion and post-test responses, and it led to lower post-test scores for both groups.

While we weren’t able to conduct this study under ideal conditions, we did collect a mountain of data to analyze, and it taught me a lot about how to organize and use data once it has been collected. Even basic statistic functions in Microsoft Excel allowed us to aggregate the data and get an idea of the ‘big picture’ of our study, especially when results were represented graphically.

In hindsight, comparing both types of discussion was too broad of a focus for our first research project. It would have been easier to study the effects of one type of discussion on student learning outcomes, especially given the amount of qualitative data collected. Through the course of this project I learned that I am good at analyzing data once it is in front of me, but I don’t have a passion for the research process as a whole. I also learned the value of data and meaningful ways to analyze it. More than anything, that is what I have taken with me and applied to other projects, both for classes and at work.