Type/Token Ratios (TTRs) frequently fail to discriminate between children at widely different stages of language development, and may fall as children get older. Such effects are caused by a negative, though non-linear, relationship between sample size and TTR. Standardization of the number of tokens before computing TTRs is recommended.…

Research on child language acquisition should distinguish between different possible causes of variation and not just attribute variation to individual variation. An alternative analysis using a different methodology can show that children's patterns of acquisition are actually relatively similar. (Author/CB)

Presents four principles for archive-based language research: maximum readability and minimum bias; consistent encoding for exhaustive computer search; systematic contrastiveness; and data comparability in elicitation, transcription and coding. Examples from existing computer archives illustrate these and other principles, and strategies are…

Provides critical analysis of Edwards (1992) presentation of a set of examples from the Child Language Data Exchange System as prototypes of bad transcription practice. The analysis argues that Edward's approach causes the development of an overly rigid set of principles that could stop progress in the analysis of spontaneous language samples.…