Supplemental Instruction (SI) is a well-known academic support model to address retention and student performance in higher education. However, in studies reporting the effect of SI, the number of attendees at SI sessions are seldom mentioned or reflected upon.

This study investigates whether there is a lower, optimal, and upper number of SI attendees for SI sessions with viable learning conditions. A literature review of 135 publications on studies of SI programmes was conducted along with a survey of 44 SI Leaders and 176 SI attendees at Lund University in Sweden.

The literature review shows that there is no consensus regarding minimum, optimum, or maximum numbers of SI session size for viable learning conditions.

In the survey, the number of attendees for optimal learning conditions was estimated to be 11–12 by both leaders and attendees. These respondents also estimated that if the number of attendees is below five or above 16 students, the learning conditions are likely to suffer. In the former case, this is attributed to too little collective knowledge, too few viewpoints, and a risk of the SI Leader being too prominent (less active participants). In the latter case, attendees are likely to find the conditions noisy and feel that they do not get seen, while the SI Leader may have difficulty structuring the session as well as getting an overview of the different group discussions.

The results hint at the importance of reporting attendance numbers at an SI session. Otherwise, it is impossible for an outsider to determine whether the conditions were favourable for small group learning and thus makes it hard to judge SI’s effectiveness.