Abstract
Though often conflated with face validity, ecological validity refers to the degree that a test or measure predicts real-world behavior/functioning. The current study leveraged two independent samples to provide a critical evaluation of the extent to which clinic-based actigraphy demonstrates ecological validity evidence relative to parent- and teacher-reported hyperactivity ratings. Further, across both samples we evaluated the extent to which the ecological validity evidence for these mechanical measures of hyperactivity varies as a function of the task children are completing while their movement is assessed objectively (low vs. high cognitive demands). Across two independent samples comprising clinically-evaluated children with and without ADHD (Ns=88, 184; M (ages) =9.2, 10.4; 6%, 33% girls; 68%, 70% White Non-Hispanic), latent path models indicated that clinic-based actigraphy during visuospatial working memory testing (high cognitive demands) demonstrated significant associations with both parent- and teacher-rated hyperactivity that were indistinguishable (p>.05) from parent and teacher ratings's associations with each other in both sample 1 (r=.57) and sample 2 (r=.35; all p<.001). Actigraphy during baseline (low cognitive demand) conditions also uniquely predicted hyperactivity at home and school in both samples (all p<.001), albeit with a less consistent yet robust pattern relative to parent/teacher associations. In both samples, actigraphy showed strong test-retest reliability over 2-4 weeks across clinic-based tasks with high cognitive demands (r=.61-.93) and high concurrent validity across tasks with high vs. low cognitive demands (r=.35-.61; all p<.007). This pattern supports the ecological validity of clinic-based actigraphy during working memory testing, which predicts real-world behavior at home and school just as well as parent perceptions of hyperactivity at home predict teacher perceptions of hyperactivity at school (and vice versa).