Beginning in December 2019, a new type of coronavirus spread worldwide, becoming a pandemic. To limit the spread almost all universities stopped classroom teaching and moved to online teaching (Telles-Langdon, 2020). However, the virus continues to spread, online teaching is still ongoing, and universities now have to consider how to deal with exams, asking whether they want to postpone, eliminate them, or conduct them online (Crawford et al., 2020).As online exams are at risk of facilitating student dishonesty by cheating, one possible solution is to proctor students while they are taking exams (Fask et al., 2014). As proctoring is an expensive solution to minimize the risk of student cheating and also questionable for reasons of data protection, Cluskey et al. ( 2011) published recommendations on how proctoring can function without a machine or a person who watches over the students during the entire exam period and is also cost-effective at the same time. The recommendations consisted of eight control procedures (Cluskey et al., 2011).The control procedures describe how to present the exams to students digitally and how to display their questions (Cluskey et al., 2011). However, Cluskey et al. did not specify their recommendations on how time pressure should be implemented. To provide further insights into this issue, this paper reports on the results of an experiment comparing different types of time pressure; namely, the same amount of time for each question, different amounts of time for each question, a total amount of time for the entire exam, rather than breaking time periods down per question. This study aimed to ascertain the CONTACT Matthias Stadler