Predictability of speech performance at age 2 for speech ability at age 3 in Cantonese-speaking Children

Abstract

This study investigated the predictability of children’s early speech profiles at age 2 in terms of (1) the Percent Consonant Correct (PCC), (2) the total number of atypical errors, and (3) the total number of errors for the speech performance at age 3 in 38 Cantonese-speaking children. Children’s speech sound production ability was assessed using a standardized test at ages 2 and 3. Candidacy for speech sound intervention at age 3 was the outcome measure. Hierarchical multiple regression analyses showed that the PCC, the total number of atypical errors and the total number of errors demonstrated high specificity but low sensitivity after controlling the individual contribution of gender, maternal education, and language performance at age 2. These three measures at age 2 correctly identified 84% to 88% of the typical children without intervention needs but only 31% to 46% of children with intervention needs at age 3. In conclusion, speech screening at early years solely based on the speech profiles at age 2 may not be an effective way in identifying children who are at risk of speech sound disorders at age 3. Other factors may play a role in shaping the trajectory of speech sound development.