Phonological assessment: a systematic comparison of conversation and picture naming

The purpose of the present study was to systematically compare two methods of speech elicitation for phonological assessment: conversation and picture naming. Subjects were 13 male, phonologically impaired children (aged 4.2 to 5.11 years). All were English speaking and had received no prior speech-language therapy. The children's performance was assessed on a conversational speech task (CST) and a 162-item picture naming task (PNT) using three levels of phonological analysis. The CST and PNT generally yielded similar sound error patterns, and severity measures on the two tasks were highly correlated. However, the PNT yielded more phonological errors. It is suggested that both methods of speech elicitation are useful clinical tools for assessment. However, to obtain a thorough sample of a child's speech output, an extensive, well-designed PNT may tap the child's phonological system more deeply, may be more efficient in some cases, and may represent a good index of phonological ability.