Predictors of phonological change following intervention.

To date, predictor variables strongly associated with phonological change as a result of intervention have not been identified. The purpose of this study was to determine the best predictor or combination of predictors of change in percentage of consonants correct (PCC; L. D. Shriberg and J. Kwiatkowski, 1982) as a result of speech-language intervention for a group of 20 participants and to replicate this procedure with a second group of 20. Participants were preschool children, ages 3;0 (years;months) to 5;11, with impairments in phonology and morphosyntax who received intervention focused on both phonology and morphosyntax in different goal attack configurations. The relationship of predictor variables chronological age, inventory size, error consistency, and expressive language to the criterion variable, change in PCC, was investigated. In both the initial study and the replication, the mean change in PCC following a 24-week intervention period was 13.1%. In the initial study, error consistency and a finite morpheme composite (FMC; L. M. Bedore and L. B. Leonard, 1998) accounted for 52% of the variance for the criterion variable. Error consistency at the first step in the regression accounted for 31.6% of the variance. In the replication, error consistency was the only variable related to PCC change, again accounting for 31% of the variance. Further research examining overall error consistency is warranted.