One more grade raises digital learning as much as science at age 15
Year: 2026
Type: preprint
Study design: Observational Study
Abstract: Education systems worldwide increasingly expect schools to cultivate computational problem solving and self-regulated learning in digital environments. Yet a prominent view holds that adolescents acquire digital competence informally through everyday device use rather than formal classroom instruction. Here we test whether an additional grade of schooling builds digital learning competence at age 15 using PISA 2025, the first cross-national assessment of Learning in the Digital World. Exploiting school-entry birth-month cutoffs across 13 education systems and 130,975 students in a preregistered fuzzy regression-discontinuity design, we compare same-age adolescents separated by one grade of schooling. One additional grade raises digital learning competence by 0.15 standard deviations, matching the gains in science, mathematics, and reading of 0.16, 0.16, and 0.16 standard deviations. Within students assigned booklets pairing both assessments, the digital-learning gain equals the science gain with a difference of −0.002 standard deviations, establishing statistical equivalence within ±0.10 standard deviations across leave-one-system-out and specification checks. Simultaneously, being one grade ahead places students among the youngest in their classroom, lowering self-rated cognitive ability and perseverance while increasing schoolwork artificial-intelligence use. Formal schooling builds computational learning capacity just as strongly as directly taught core subjects.
Author: Liu, Kang
Institution: Shandong Normal University
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W7234731769
