Factors relating to the prevalence of autism

Abstract

This study examined the relationships between socioeconomic, cultural, and ethnic variables on the rates of autism reported by the states. Each state and The District of Columbia was included in the sample $(N=51).$ A latent structural relations statistical analysis was conducted to analyze the data for goodness-of-fit to the theorized model.

The variables studied as influences on the rate of autism were childhood poverty, ethnicity, access to psychological/psychiatric services, and access to advocacy, specifically geographic access to Autism of America members. The theory tested whether these variables affected the rates of autism reported by the states in their annual reports to Congress. The variables associated with the increased rate of autism included Latino ethnicity as well as access to autism advocacy groups. Using goodness-of-fit statistics to analyze the model, 49 states had a good fit with the proposed model. A modification of the model also found a relationship between the 1994 rate of autism and membership in the Autism Society of America. The perfect fit of the data (GFI = 1.000) was in a modification which included all the states and proposed a direct relationship between Latino proportionality in the population and the rate of autism in 1994. In that model, when the proportion of Latino children is proportionately large, the rate of autism reported by the states is higher $(p<.05).$</p>

These findings have implications for further research and study. The influence of socioeconomic and sociocultural factors on the labeling and education of children should be monitored closely. There is ample evidence to suggest that the disability categories of IDEA are replete with over- and under-representation difficulties. This study would suggest that those same difficulties, particularly ethnicity, may well be influences in the labeling of children as autistic.