Steedle, J. T., Cho, Y. W., Wang, S., Arthur, A. M., & Li, D. (2022). Mode effects in college admissions testing and differential speededness as a possible explanation . Educational Measurement: Issues and Practice , 41 (3), 14–25. https://doi.org/10.1111/emip.12484
Accommodations were not specified; instead, the differences in college admissions testing when the exam was taken in paper form versus taken online were examined.
U.S. high schools that had sufficient Chromebooks for students to effectively participate were included. A total of 1807 and 1776 individuals participated in the paper and online October ACT exams, respectively. For the December ACT exam, 3147 participants took the paper exam, and 3205 took the online exam. February's ACT exams had 3297 participants take the paper exam and 3348 participants take the online exam. Participants were offered free testing to participate.
Random assignment was used to assign students to paper or online testing conditions. Participant performance was examined in order to identify the impact of testing in the different conditions.
Online scores were higher than paper scores, particularly on the reading, English, and writing tests. Furthermore, participants who took the exam online were more likely to respond correctly to test items, particularly near the end of the English and reading tests.