In today's digital age, tests are increasingly being delivered on computers. Many of these computer-based tests (CBTs) have been adapted from paper-based tests (PBTs). However, this change in mode of test administration has the potential to introduce construct-irrelevant variance, affecting the validity of score interpretations. Because of this, when scores from a CBT are to be interpreted in the same way as a PBT, evidence is needed to support the reliability and validity these scores (AERA et al. 2014). Numerous studies have investigated the impact of changing the mode of test delivery from paper to computer, not only in terms of their psychometric properties, but also with regard to possible sources of construct-irrelevant variance. This article summarizes the main lessons learned from mode effects studies in education over the past 30 years and discusses some of the questions remaining.