A Psychometric Assessment of Pilot Language and Communication Items on the 2018 and 2019 National Survey of Children’s Health

• In Event: Poster Session 09
  In Poster Session: PS 09 Topic - Language, Communication

Abstract

Mastery of language and communication during early childhood is among the strongest predictors of later academic achievement and intellectual ability (Walker, Greenwood, Hart, & Carta, 1994; Wilks, Gerber, & Erdie-Lalena, 2010; Young et al., 2002). Failure to meet language development milestones during infancy and early childhood can be an early indication of developmental disorder or learning disability (Wilks et al., 2010). Until recently, nationally representative data on language and communication development among children in the United States (US) has not been available to contribute normative data for critical efforts to promote language development. This study represents the first psychometric analysis of nationally representative data derived from a National Survey of Children’s Health (NSCH) pilot measure of language and communication development among children ages 1–5 years. We analyzed 14,573 parent responses to language and communication items on the 2018 and 2019 NSCH to evaluate whether together the newly added 11 items represent a single latent trait for language and communication, and to determine the normative age of success for each milestone.
We analyzed 2018 and 2019 data from NSCH, a nationally representative, parent-report survey funded and directed by Health Resources and Services Administration’s Maternal Child Health Bureau and conducted by the United States Census Bureau. Our study sample included children ages 1-5 years, with parent responses to all eleven language and communication items (n=14,573). All items were dichotomous, with response options “yes” (for reported success) and “no.” We calculated unweighted frequencies of reported child success on each item as well as weighted prevalence estimates and Clopper Pearson 95% Confidence Intervals (CI), stratified by age. We used weighted exploratory factor analysis and unweighted correlation matrices to confirm overall unidimensionality of the measure. We applied weighted one-parameter Item Response Theory (IRT) to rate and cluster items by difficulty relative to the latent trait of developmental language ability. We employed weighted logistic regression methods to examine each item for potential Differential Item Functioning (DIF) by race/ethnicity, child sex, household language, and highest educational attainment of parents/guardians.
Exploratory factor analysis and correlations indicated overall unidimensionality. Item Characteristic Curves (Figure 1) and corresponding weighted difficulty parameter estimates ordered items by relative difficulty, indicating groupings were overall concordant with proposed age milestones and representative approximate 90% success cut-points by child age (Table 1). Indicated normative age cut-points for three items differed slightly from proposed milestone ages. Potential for DIF by demographic groups assessed was not indicated for most items.
This study represents the first psychometric analysis of pilot language and communication items on the 2018 and 2019 NSCH using data representative of non-institutionalized US children ages 1-5 years. Using weighted IRT methods accounting for complex survey design, we established that normative age cut-points were overall concordant with previously established age milestones, with minor adjustments indicated for three items. These findings have the potential to inform future research on risk and protective factors for language and communication development among US children, enhance utilization of developmental surveillance tools, and guide critical prevention efforts at a population level.

Authors

• Helena Jean Hutchins, Centers for Disease Control and Prevention
  Presenting Author

• Lara R Robinson, Centers for Disease Control and Prevention
  Non-Presenting Author

• Sana Charania, Centers for Disease Control and Prevention
  Non-Presenting Author

• Melissa Danielson, Centers for Disease Control and Prevention
  Non-Presenting Author

• Reem Ghandour, Maternal and Child Health Bureau, Health Resources and Services Administration
  Non-Presenting Author

• Jennifer Zubler, Centers for Disease Control and Prevention
  Non-Presenting Author

• Kathy A Hirsh-Pasek, Temple University
  Non-Presenting Author