Section-specific gains in foundation-year laboratory reports after criterion-referenced self-assessment and progressive authorship
While self-assessment against marking criteria is widely advocated, its effects on different components of student work remain unclear. This study evaluates a redesign of a UK foundation-year engineering laboratory module incorporating a structured self-assessment coversheet and a progressive authorship scheme. Across six cohorts (n = 134), improvements concentrated in the sections that require evaluation and argument --- Discussion (d = 0.65) and Conclusions (d = 0.69) --- while procedural sections showed smaller, less consistent gains. Coding of 993 matched student--instructor section assessments for depth of engagement with criteria revealed that deeper engagement correlated with more accurate self-assessment at the between-student level, primarily in evaluative sections. However, the same section-by-section calibration was not observed within individual students, leaving the causal direction unresolved. All coding-based findings were validated through independent automated re-coding, including an intervention-blind large language model. The findings suggest that self-assessment against criteria may develop evaluative writing skills more effectively than procedural ones, though the engagement--accuracy link remains associational rather than causal at the individual level.