Purpose: Children with neurodevelopmental disorders historically exhibit lower and more variable nonverbal intelligence (NVIQ) scores compared to their typically developing peers. We hypothesize that the intrinsic characteristics of the tests themselves, particularly the cognitive constructs they assess, may account for both the lower scores and variability across tests and over time. Using a qualitative content analysis approach, we examined the extent to which key cognitive constructs are engaged in NVIQ tests and how these constructs compare across different tests. Method: Current editions of seven NVIQ tests were selected based on their relevance in clinical and research settings. Qualitative coding of constructs was developed iteratively by speech-language pathologists and researchers. The codes focused on cognitive domains most affected in highly prevalent neurodevelopmental conditions, including attention, receptive language, statistical learning, and working memory. Results: We identified multiple subfeatures for our constructs of interest. Using this coding framework, we found that NVIQ tests qualitatively differ in the extent to which these four constructs influence test performance. Conclusions: Our findings suggest that understanding the impact of cognitive constructs on NVIQ tests can help explain why children with neurodevelopmental disorders exhibit lower and more unstable NVIQ scores compared to their peers. We provide recommendations for the use of NVIQ tests with neurodevelopmental disorder populations and encourage researchers and clinicians in speech and hearing sciences and psychology to use our results to inform test interpretation and selection.