Multimodal Czech online handwriting and cognitive data from children with and without handwriting disabilities

Handwriting is a fundamental skill that significantly influences academic achievement and cognitive development. Despite its importance, open-access datasets combining dynamic handwriting signals with comprehensive cognitive profiles are scarce, particularly for children with handwriting disabilities. We present a multimodal dataset consisting of online handwriting and cognitive assessment data from 276 Czech children (ages 8–12), including 161 individuals diagnosed with dysgraphia. The data acquisition utilized a Wacom Cintiq 16 tablet and a digital pen to record 16 graphomotor and writing tasks, capturing high-resolution x-y coordinates, pressure, and pen tilt at an average frequency of 167 Hz. Alongside these raw signals, the dataset includes scores from standardized assessments of cognitive abilities (WJ-IV), visuospatial skills (RCFT), phonological awareness (BACH), and self-reported handwriting proficiency (HPSQ-C) . We provide raw data in SVC format and integrated, preprocessed records in JSON and RDS formats, supported by specialized Python and R processing scripts. This resource is intended to facilitate research in graphonomics, machine learning, and educational psychology, providing a benchmark for the objective diagnosis of handwriting difficulties.书写是一项基础技能，对学业成就与认知发展具有显著影响。尽管其重要性不言而喻，但目前将动态书写信号与全面认知特征相结合的开放获取数据集仍较为匮乏，针对书写障碍儿童的此类数据集更是如此。本研究构建了一个多模态数据集，包含276名捷克8至12岁儿童的在线书写数据与认知评估数据，其中161名被诊断为书写障碍（dysgraphia）。数据采集采用Wacom Cintiq 16数位板与数位笔，完成16项书写运动与书写任务，以平均167Hz的采样率采集高分辨率的x-y坐标、笔尖压力与笔身倾斜角度。除上述原始信号外，本数据集还包含标准化认知能力评估（WJ-IV）、视空间技能评估（RCFT）、语音意识评估（BACH）以及自我报告式书写熟练度问卷（HPSQ-C）的得分数据。我们提供SVC格式的原始数据，以及JSON与RDS格式的整合预处理记录，并附带专用的Python与R处理脚本。本数据集旨在推动书写科学、机器学习与教育心理学领域的相关研究，为书写困难的客观诊断提供基准参照。