My Hybrid Course
Have I inadvertently designed a monster?
In the winter semester I offered a hybrid course, Extreme Weather (AOSS 102), at the University of Michigan (UM) that was simultaneously face-to-face, streamed live using Echo360 and captured. The course used LectureTools (as described earlier) so students could participate remotely to answer questions, ask questions, take notes and identify when they were confused. Hence my 168 students could actively engage in the course material regardless whether they physically came to class or participated remotely.
However, in analyzing the copious data that LectureTools affords I have discovered that students with higher incoming grade point averages (GPA’s) behaved remarkably different from those with lower GPA’s (Aside: At the University of Michigan – and many other American universities and colleges – student performance is measured on a 4.o grade point scale). The lower GPA students tended to answer a slightly lower number of questions, took significantly fewer notes and were far more likely to participate remotely than the higher GPA students. All this begs the question:
- Are hybrid courses creating an environment that incubates mediocre participation or
- Does the availability of streaming provide opportunities for lower GPA students to participate at some level when they otherwise might not have participated at all?
“Smarter” Students Behave Differently
First of all I learned, perhaps not surprisingly, that student outcomes in my course were related to the students’ incoming grade point average (GPA). The relationship between GPA and grades has been well defined by other, including UM’s Tim McKay (c.f. here). Figure 1 shows that the students who had better GPA’s tended in general to do better in my course, though there was still considerable scatter.
But as an instructor there’s nothing I can go about my students’ incoming GPA so I explored to what degree did students with differing incoming GPA’s behave differently. In other words what is it that “smart” students do that weaker students don’t?
The results show that higher GPA students behaved systematically different that their lower GPA counterparts in at least three ways:
1. Variations in Participation
First, when stratified by incoming GPA it appears that lower GPA students tend to respond to a slightly lower fraction of the questions posed in class (Figure 2), and when they do answer questions they tend to get fewer correct. The variation in participation across GPA is modest but fairly consistent.
Getting fewer questions correct in class is not necessarily surprising but the reduced level of effort (answering fewer questions) is problematic. The lower participation rate suggests a reduced sense of responsibility. That said, the level of questions attempted for the lowest GPA students was still around 70%.
2. Variations in Note-Taking
Second, and more significant, the number of words students type in notes and the number of slides to which they type notes increases dramatically with incoming GPA as shown in Figure 3. Students with higher GPA’s tend to type considerably more notes in class.
Again the reduced level of participation by lower GPA students is troubling. Does this provide evidence that the lower GPA students have lower GPA’s because of poorer study methods? Perhaps they do not know how to effectively participate in class or do not value note-taking as a mechanism for engaging with the material.
3. Weaker Students Avoid the Classroom
Third, and most surprising for me, the nature of how students chose to participate in this hybrid course was definitely different for higher GPA students than lower GPA students. Figure 4 shows that higher GPA students tend to participate in person in the classroom. Lower GPA students, on the other hand, both tended to participate remotely and/or tended to simply miss class more often.
These results are a tad discouraging for those offering hybrid courses. Is it possible that the very design of hybrid courses, even those that promote and encourage synchronous participation, may systematically provide weaker students with an environment where they can avoid taking active measures to participate? On the other hand, if the course were only face-to-face would the lower GPA students have come to class and participated at a higher level? I have no evidence that this would be true.
My walk away, and the one I will carry into the next semester, is that the students who have lower GPA’s coming into the course and who opt to participate remotely should be exposed to results such as these so they can understand the potential consequences of lower participation. Moreover as more is learned from the vast dataset afforded by systems like LectureTools it should be shared with academic advisors who presumably best know their advisees academic records and can share the findings with those who need it most.