Assessing phonology in young children.

We have presented a discussion of three important concepts regarding early phonological assessment. The first is that the child's language level must be considered in collecting and analyzing a sample and in interpreting the results. For a child with age-level language abilities, the sample should consist of both a single-word test and a running speech sample; both independent and relational analyses are appropriate. For a child with delayed expressive language and a small productive vocabulary, a sample comprised of spontaneous productions is more appropriate. In this case, the sample should be analyzed in terms of sound classes and syllable and word shapes that occur; a phonological process analysis is inappropriate. Lexical selection patterns should be noted. The results of the analyses should be interpreted in view of the expectations for the child's language level. A child with a normally developing language system is expected to have more advanced phonology than a child of the same age with delayed language. Thus, a child with a large vocabulary and word combinations is expected to have an expanding phonological system, with a full range of sound classes and syllable and word shapes. If a child is delayed in language and is still within the first 50-word stage, the expectation is that the phonological system will be more limited. Critical features for the phonology of early productive vocabulary have been identified. Lack of one or more of these features is indicative of atypical phonological development at any age and language level. The second concept is that the phonological system as a whole must be considered. In particular, the analyses and expectations should be based on the presence or absence of sound classes and syllable shapes rather than on sounds per se. Lack of an entire class or syllable structure would be cause for concern; lack of a particular sound, even though it has been shown to be acquired early, would not be. Thus, lack of the entire fricative class at 36 months would be of concern, whereas errors on /f/, which according to Prather et al., 1975, is mastered by this age, would not be. The third important concept emerges from the case studies; both studies demonstrate that longitudinal assessment is necessary to document changing profiles over the course of development. A child such as David, who has a normal-but-delayed profile in language and phonology at one age, may subsequently exhibit atypical patterns as phonology and language dissociate.(ABSTRACT TRUNCATED AT 400 WORDS)