Gwen Mitchell and Kenneth Locke have an article in
Autism titled "Lay Beliefs about Autism Apectrum Aisorder among the General Public and Childcare Providers."
The abstract:
We conducted a survey of beliefs about autism among the general public in the United States and Canada (n = 823) and among individuals working in childcare facilities in the state of Idaho (n = 176). Results included the following. Almost all respondents correctly believed that autism’s primary causes are genetic and neurological (not parenting, drugs, or current diet), that it can be identified in early childhood, and that helpful interventions exist. Respondents generally distinguished diagnostic from non-diagnostic traits, but approximately half incorrectly labeled constant squirming as diagnostic and difficulties in making friends as non-diagnostic. College graduates and childcare workers were more likely to have learned about autism in professional/academic settings and to correctly recognize diagnostic traits. Of concern, 10% of respondents considered vaccinations to be among the two main causes of autism. Accurate public understanding of autism spectrum disorders can facilitate early identification and effective intervention; our results suggest that efficient channels for conveying accurate information include broadcast and online media (from which the general public, especially members of ethnic minority groups, were most likely to learn about autism), and professional development courses for childcare providers.
The article contains an important caveat:
We obtained our general public sample through MTurk, an online crowdsourcing site that has been shown to generally yield diverse and representative samples (Buhrmester et al., 2011). For example, an investigation by US political scientists Berinsky et al. (2012) concluded that “the demographic characteristics of domestic MTurk users are more representative and diverse than the corresponding student and convenience samples typically used in experimental political science studies” (p. 352). Although their conclusion applies to the current sample as well, we should caution that compared to 2012 US population census data, our sample contained a larger percentage of college graduates and a smaller percentage of African Americans and Latinos, which is one reason we reported the findings separately for Whites and non-Whites and for individuals who had versus had not graduated from college. On the other hand, because our analyses showed that most beliefs about autism were not influenced by ethnicity or education, there is generally no reason to believe our sample produced biased estimates of lay beliefs about autism.
