Are whole-word measurements of speech accuracy more sensitive to change than segmental measurements?

Background: Measurement of change occurring during and after intervention is a critical part of evidence based practice. Therefore, it is imperative to find measurements that are sensitive enough to show real change. In intervention for speech sound disorders (SSDs), therapists have traditionally used a range of measurements. Currently there seems to be no consensus on the most valid method of measuring changes in speech accuracy (Flipsen et al. 2005). This has negative consequences for evidence-based practice. Objectives: This project investigates whether a whole word measure of speech accuracy, Proportion of Whole Word Proximity (PWP), is more sensitive to change than the segmental measure, Percentage Consonants Correct (PCC), most commonly used in the literature. Methods: 12 monolingual children with SSDs aged between 3;0 and 4;11 took part in the study. Participants were assessed in single word production using the Diagnostic Evaluation of Articulation and Phonology (DEAP) (Dodd et al. 2002), and in continuous speech using the Renfrew Action Picture Test (RAPT) (Renfrew 1997) at initial assessment and again 8 weeks later. Results: Analysis showed a high correlation between PCC and PWP. For the whole group, there was no statistically significant difference between PCC and PWP in measuring change. However, this may have been due to the fact that no significant change occurred at group level. Analysis of individual cases where significant change did occur showed that the different methods were differentially sensitive to different types of change. Conclusions: In this sample, PCC appeared to be a more sensitive measure of change in features, while PWP appeared to be more sensitive to changes in structure. Clinical implications are discussed and recommendations are made.