Table 1 Main findings of quantitative research (see text for details)

• Gender: Female students tend to provide more positive ratings (2 studies).

• Procedural aspects of teaching: Course organisation, effective communication of learning objectives and high staff responsiveness are associated with higher overall ratings (1 study).

• Satisfaction with examinations: Students who are more satisfied with end-of-course examinations tend to provide more favourable course ratings (2 studies).

• Timing of data collection: Course ratings provided retrospectively (i.e., up to one year after a course) can be less favourable (2 studies) or slightly more favourable (1 study) than ratings provided at the end of a course.

• Initial interest: Students who are more interested in course content tend to provide more positive ratings (2 studies).

• Didactic methods: Provision of high quality feedback predicts overall ratings (1 study).

• Blueprint availability: Availability of an examination blueprint improves overall course ratings (1 study).

• Data collection tool: As opposed to paper/pencil evaluations, online evaluations yield lower response rates (1 study) but slightly more favourable ratings (1 study).

• Performance level: High-performing students tend to provide more positive ratings (2 studies).

• Presentation format: Live lectures receive more favourable ratings than identical, videotaped lectures (1 study).

• Response rate / selection bias: Data obtained in a mandatory evaluation procedure are no different from data obtained in a voluntary setting (1 study). High-achievers might (1 study) or might not (1 study) be over-represented in student samples self-selecting to participate.

• Attendance: Mandatory seminars receive more positive overall ratings than lectures with voluntary attendance (1 study). Students voluntarily attending lectures tend to provide more positive ratings than non-attendees (1 study).

• Point of reference: Asking students to predict how their peers would rate a course produces the same results as obtaining individual ratings but requires fewer participants to get stable results (2 studies).

• Teacher attitudes: Negative teacher attitudes towards a course negatively influence student ratings (1 study).

• Design of rating scales: Positively phrased items and scales with the positive anchor on the left and no labels on intermediate scale options yield the most favourable ratings (3 studies).