Dataset: Examining the medium effect with VET students: Do comprehension and decoding skills matter when reading on paper and screen?

Abstract:The present study investigated how authentic informational texts on different mediums impacts the comprehension of students at a Norwegian upper secondary School of Vocational Education and Training (VET; n = 106, M = 16.5), and whether the impact differed according to individual differences in comprehension performance and decoding skills. In a within-subjects experimental design with presentation order counterbalanced, students read two history texts on their own laptop computers and printed A4 booklets. Results indicated that the whole sample had lower comprehension on screen compared to on print. There was also an effect of decoding skills on overall comprehension, but no significant interaction between medium and decoding skills. However, when splitting into higher and lower comprehension groups, only lower performers experienced a significant medium effect. Given the prevalence of screens in upper secondary classrooms, these findings suggest that educators should be aware of challenges with, and teach strategies for, screen reading.Data description:The dataset includes the anonymous data of the participants who took the reading test. The dataset is presented in long format with 20 cases per ID (matching each question response per text). The outcome variable is the RC-score, coded 1 or 0 for correct and incorrect responses.The predictor variables are text (1 = text A, 2 = text B), text order (1 for first or 2 for second), Item (A1-A10 and B1-B10), and medium (1 = digital, 2 = print)Other variables are gender (1 = female, 2 = male) and the WCT_scores representing decoding skills (from 0-75).