A key issue in educational research is whether the outcome of an assessment is affected by the
modality via which it is administered. This large-scale study of existing data investigates whether the mode of administration of a state university’s writing proficiency examination (paper-based or computer-based) affects the quality of submitted essays, as indicated by scores produced by trained human readers. We sought to determine if measurable differences in mean scores exist across modes of administration for various demographic groups, particularly between “digital natives” and “digital immigrants” (Prensky, 2001), as well as across the “digital divide” (Sidney Howland, 1998). Prior studies investigating comparability between paper- and computer-based assessments have produced conflicting results, as have studies that focused on demographic impacts across modes. This study strives to provide substantial empirical evidence of test mode-related impacts on mean essay scores via statistical analysis of a large corpus of assessment data, including essay scores and associated demographic information. Evidence of age-related effects regarding computer-based testing was inconclusive. Significant evidence of the effect of the digital divide on subjects of higher and lower socioeconomic status was found.