793 - To Trust or Not to Trust: The Relationship Between Parental Education/Multilingual Status and Parent Concerns about Language

Publication Number: 793.6804

Megan Choi, Northwestern University, Chicago, IL, United States; Janedelie Romero, Endeavor Health, Deerfield, IL, United States; Matthew Pellerite, University of Chicago Pritzker School of Medicine, Evanston, IL, United States; Adriana Weisleder, Northwestern University, Evanston, IL, United States

Background: Multilingual children are under-referred for early intervention services. Parental concerns are an important resource for early identification of developmental delays. Previous research found that sociodemographic variables such as parental education and socioeconomic status did not affect the accuracy of parental concerns. We revisit this question by examining parental concerns about language development in a sample with a high proportion of multilingual households.

Objective: Investigate the links between multilingual status/parent education and the frequency and accuracy of parent concerns about language development.

Design/Methods: Parents (n=180) completed an intake when their child was 14-18 months that elicited concerns about their child’s language development. Of these, 155 also completed the MacArthur-Bates Communicative Development Inventory (MCDI). Toddlers scoring ≤10th percentile for words produced on the MCDI were classified as language delayed. Multivariate models examined associations between (a) parent education (< BA, BA, post-grad) and (b) language background (monolingual vs multilingual), and parent language concerns, adjusting for age.

Results: Fifty (28%) of parents raised concerns about their child’s language development; this was higher among those with a post-grad degree (35%) than those with a < bachelor’s or bachelor's degree (19% and 23%; p <.05). There were no differences in the frequency of concerns between monolingual and multilingual households (p=.75). Parents who raised concerns were more likely to have a child in the bottom 10th percentile on the MCDI than parents who did not raise concerns (p <.001). Overall, parental concerns had 57% sensitivity and 79% specificity in identifying children with language delays. Among parents who raised a concern, those with a < bachelor’s or bachelor's degree were more likely to have a child with an actual language delay (43% and 50%) than those with a post-grad degree (14.3%, p<.05). Household multilingualism was not associated with the accuracy of parent concerns.

Conclusion(s): In both monolingual and multilingual households, parent concerns were moderately accurate in identifying language delay. Parents with lower education were less likely to raise concerns than those with higher education, and those concerns were more likely to identify a real delay. Clinicians serving linguistically diverse populations can use parent concerns about their child's language development as a useful indicator of language delay. Eliciting concerns from parents with lower education levels may be especially useful and lead to earlier interventions.

Participant Demographics


Percent children with language delay among parents with a language concern