The Effect of Sampling Context on Preschoolers' Finite Verb Morphology Composite Scores.

Abstract

PURPOSE The finite verb morphology composite (FVMC) is a valid measure for charting children's tense development and for differentiating children with and without language impairment during preschool and early elementary years. However, it is unclear whether FVMC scores vary as a function of language sample elicitation contexts. The current study evaluated the performance on FVMC in preschool-aged children across different language sampling contexts. METHOD Participants were 38 English-speaking children who were between the ages of 3 and 4 years and below the mastery level of tense usage in three language sampling contexts, including conversation (free-play), picture description, and narratives (story retell). FVMC from each sampling context was computed by calculating the overall accuracy of copula be, auxiliary be, third-person singular present -s, and past tense -ed combined. A linear mixed-effects model comparison was carried out to determine the effect of sampling context on FVMC scores. RESULTS After controlling for child age, mean length of utterance, and the number of obligatory contexts for FVMC scoring, FVMC scores were significantly higher in conversation than in picture description and narratives. In addition, FVMC scores across the three sampling contexts were significantly correlated (rs ≥ .62, p < .001). CONCLUSIONS Although children's performance on FVMC relative to each other was quite stable across sampling contexts, FVMC scores may vary with sampling contexts. As compared to picture description and narratives, conversation may not adequately capture the limitation in preschoolers' tense development that is important for therapeutic planning.