Autism Spectrum Disorder (ASD) is a developmental disability that occurs among people across different socio-demographic groups. According to the Centers for Disease Control and Prevention (CDC) estimate, nationally 1.9% of children in the USA were diagnosed with ASD in 2016. However, the diagnostic and identification of ASD vary greatly across states. Differences across states are likely to impact the found prevalence for children with ASD in those states, which can cause the potential number of missed ASD cases to vary. The purpose of this study is to develop a potentially missed ASD case metric from available school data and investigate the relationship between the missed case metrics and relevant county-level socio-demographic covariates. The study focuses on the relationship between potentially missed ASD and ten variables include states (Idaho, Mississippi, and California), primary care physicians per 100,000 (2017), mental health providers per 100,000 (2019), children in poverty per 100,000 (2018), uninsured children per 100,000 (2017), residential segregation rate between black and white (2014-2018), high school graduation per 100,000 (2016-2017), median household income (2018), child mortality per 100,000 (2015-2018), and the percent of rural population based on Census Population Estimates (2010). Results: Simple correlation and regression models displayed significant relationships between potential missing ASD and most predictors. However, after including states, many predictors are not significant anymore, suggesting that individual states are an important source of variance to consider for analysis of missed ASD. By adding interactions between continuous predictors and states into the multiple linear regression models, uninsured children and percent of rural populations show significant differences between states with predictors of the relationship to missing ASD cases. Future studies should consider linear mixed models for the analysis of missing ASD.