With all of the furore generated by the BC government and the BC Teacher’s Federation (Sherlock & Shaw 2014), the issue of class sizes is once again in the public consciousness. Teachers allege that they are unable to adequately furnish students with enough attention in large class sizes while the government bemoans the cost of additional teachers. With this in mind, how do class sizes affect student performance in an advanced setting such as university?
Class size, as Becker (2001) notes, is a useful piece of data because it is both easily observed and manipulable by university administrators. They can condense courses into larger sections with fewer lecturers or a multitude of sections based on budgetary and staffing concerns. Moreover, the rise of new lecture and course methods such as massive open online courses (MOOCs) like Coursera have pioneered new initiatives to overhaul course design and education delivery. UBC itself already participates in several initiatives, such as the Carl Weiman Science Education initiative (CWSEI, 2011). In any of these initiatives, class size is an important starting point that dictates the importance of class discussion, learning activities, and of course the level of attention that each student will receive from the professor.
Previous research in this area has been inconclusive with Johnson (2010) and Becker (2001) noting that, due to the number of variables involved, previous researchers (such as Raimondo, Espsosito, and Gershenberg (1990)) have drawn conflicting conclusions. However, the bulk of research done in the past 15 years indicates a negative relationship between class size and academic performance. Specifically, Becker and Powers (2001) report that they find a negative and significant correlation between beginning class size and academic performance.
However, Bandiera, Larcinese, and Rasul (2010), note that class size impacts performance disproportionately, with the effect only becoming significant in both small and large class sizes, as intermediate size classes showed no significant impact on performance. In accordance with Bandiera et al.’s findings, we will only be looking at small and large class sizes, represented in our sampling as classes of 50 students and under and classes of 100 students and over.
Our study will perform a general analysis of UBC’s courses in order to see if there is a correlation between class size and academic performance. The results could be a useful piece of evidence for researchers attempting to redesign teaching strategies. In an age where technology enables a lecturer to have up to thousands of students, it is vital for an innovative institution like UBC to re-evaluate its current course structure, especially in terms of class size.