Motivation quality and school implementation are associated with activity after high-stakes physical education testing in Chinese adolescents

Introduction High-stakes physical education examinations mobilize adolescents at scale, but whether examination-period activity is retained after the incentive is removed remians unclear. Methods Guided by self-determination theory, we followed a convenience sample of 3,493 adolescents from 50 Chinese schools across four waves spanning the physical education entrance examination. The design included a calendar-matched Grade 8 reference cohort and an ActiGraph GT3X+ accelerometer subsample; endpoint models used 336 Grade 9 students with valid device data at both post-examination waves. Results Grade 9 moderate-to-vigorous physical activity (MVPA) rose by 14.04 min/day before the examination, 95% CI [11.84, 16.24], then fell by 19.55 min/day more than the Grade 8 trajectory, 95% CI [ − 23.77 , − 15.32 ]. Autonomous motivation predicted subsequent non-compulsory MVPA, whereas controlled motivation did not; the within-model contrast was β = 0.157 , 95% CI [0.015, 0.299]. Baseline teacher autonomy support attenuated the association between controlling school implementation and need frustration, interaction β = − 0.065 , 95% CI [ − 0.097 , − 0.032 ]. More controlling implementation was associated with lower non-compulsory MVPA at follow-up, β = − 0.114 , 95% CI [ − 0.208 , − 0.019 ]. Discussion Activity outside required school periods is an operational proxy, not a direct measure of volition. Examination-period activity increased, while motivation quality and implementation style were associated with short-term post-examination activity.