Does transcription instruction make writing intervention more effective? A meta-analysis

Multi-component writing interventions that target text generation and/or self-regulation skills differ in the extent to which they also emphasize transcription instruction, which focuses on spelling, handwriting, or keyboarding skills. To determine how transcription instruction contributes to the effects of multi-component writing intervention, we conducted a meta-analysis of group experimental writing intervention studies for K-12 students. A total of 54 effect sizes from 33 studies for writing quantity and 105 effect sizes from 64 studies for writing quality were included. Results showed that multi-component writing intervention, either with or without a transcription component, yielded stronger effects on writing quantity and quality than transcription intervention alone (quantity: g = 0.40, 95 % CI [-0.14, 0.93]; quality: g = 0.15, 95 % CI [-0.20, 0.51]). However, multi-component writing intervention with transcription components (quantity: g = 0.50, 95 % CI [-0.01, 1.01], quality: g = 0.55, 95 % CI [0.10, 1.00]) tended to have smaller effects than those without a transcription component (quantity: g = 0.66, 95 % CI [0.30, 1.03]; quality: g = 0.85, 95 % CI [0.58, 1.13]). Moderator analyses showed that students at secondary grade levels tended to have larger writing effects than students at elementary grade levels, regardless of the type of writing intervention. Also, students with academic difficulties gained larger effects on writing quantity than writing quality outcomes. These findings supported the theoretical and practical importance of multi-component writing intervention on higher-level writing skills, encouraging its implementation even for young and/or struggling writers, for whom transcription intervention is often emphasized.