A further look at two grammatical measures from children's language samples and their contribution to the diagnostic process.
Minh Bùi, Laurence B Leonard
PMID 38349663WHAT IT FOUND
Two language-sample measures sorted four- and five-year-olds with DLD from typical peers: a tense/agreement score missed one child, Sentence Point missed none.
In children with DLD, scores did not track the SPELT-P2 grammar test. Use them with other assessments, not alone.
Key findings
01At a cutoff of 87.24, the FVMC misclassified one child with DLD, resulting in 95.5% sensitivity and 100% specificity.
02At a cutoff of 41, the Sentence Point misclassified no child, resulting in 100% sensitivity and 100% specificity.
03In the 22 Study 1 children and the larger group of 60 children with DLD, FVMC and Sentence Point scores were unrelated to SPELT-P2 scores.
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The Study 1 comparison had only 22 children with DLD and 22 children with typical development. The children with DLD were selected from earlier word-learning studies and had already been identified partly by SPELT-P2 scores, so they may not represent all children referred to clinic. The typical children came from an earlier study, and their language samples were collected years apart from the DLD samples, although the procedures were consistent. The study included mostly White, monolingual English-speaking children, so it may not apply to children with other language backgrounds. The FVMC can be inflated by memorized or routinized forms such as that's a or It's a. The Sentence Point excludes utterances without a verb, such as The baby hungry, which can inflate accuracy. The diagnostic accuracy was tested against the SPELT-P2, a grammar-based measure, not against a fully independent clinical diagnosis.
Declared interests
The authors declared no financial or non-financial conflicts of interest.
The easy way to misread this
Do not treat the 95.5% and 100% accuracy figures as proof that the FVMC or Sentence Point can diagnose DLD by itself. The children were already selected using the SPELT-P2, and the paper says these measures are useful only as part of a diagnostic battery.