Reliability and validity evidence for the English and Spanish preschool narrative language measures-listening

For young Spanish-speaking children entering U. S. schools, it is imperative that educators foster growth in the home language and in the language of instruction to the fullest extent possible. Monitoring language development over time is crucial for promoting language development because it allows educators to individualize student instruction. The Narrative Language Measures-Listening (NLM-Listening) subtest of the CUBED assessment was designed as a curriculum-based benchmark and progress monitoring tool for eliciting narrative language samples from preschool children in an authentic manner. In both English and Spanish, NLM-Listening includes 25 preschool alternate forms equated on story structure and language complexity. Standardized administration and real time scoring procedures enhance the NLM-Listening's reliability and feasibility. This study offers a rigorous examination of the reliability and validity evidence for using the Spanish and English NLM-Listening with preschool children. Using a Latin-square design to randomize order of administration, 126 three- to four-year-old children attending Head Start preschool received up to 25 forms of the Spanish (N = 61) or English (N = 65) versions, and two criterion measures of language in Spanish or English, respectively, within a few weeks of each other. Results indicated that the NLM-Listening has strong alternate-form reliability and strong evidence of validity based on unidimensional factor analyses and moderately high correlations with criterion measures of language. The NLM-Listening is a promising tool for monitoring preschoolers’ language development in English and Spanish.