ConductScienceConductScience
MLU-W

Mean Length of Utterance in Words (MLU-W)

MLU-W counts whole words rather than morphemes and is the practical fallback when morpheme tagging is not reliable.

What MLU-W measures

Mean Length of Utterance in words divides the total number of words in a language sample by the total number of complete, intelligible utterances. It strips away the bound-morpheme accounting that makes MLU-M demanding and lets clinicians derive a length index directly from a fast transcript. For that reason it is the format most frequently used in automated language-sample software and in SUGAR — Pavelko & Owens built the SUGAR norms around MLU-W precisely to lower the transcription burden on school-based SLPs.

Formula

MLU-W = total words ÷ number of complete and intelligible utterances

Normative ranges and benchmarks

  • Age 3;0 — mean 2.9 words, SUGAR
  • Age 4;0 — mean 3.7 words, SUGAR
  • Age 5;0 — mean 4.5 words, SUGAR
  • Age 6;0 — mean 5.3 words, SUGAR
  • MLU-W typically runs about 0.7 – 0.9 units below MLU-M for the same sample

Normative bands are central estimates drawn from the cited literature. Individual variation is wide — always cross-reference against the source paper and your assessment's own manual before quoting a cut-score in a report.

Clinical use

MLU-W pays for itself when time is the binding constraint. A school-based SLP with four new evaluations on the schedule can collect 50 utterances at the lunch table, split them at pauses, and run MLU-W from the transcript in under ten minutes. The SUGAR norms publish exactly this protocol and provide 10th, 25th, 50th, 75th, and 90th percentile bands, which makes MLU-W defensible in a report without the clinician having to defend a bound-morpheme decision on every verb. The limitation is the ceiling — once MLU-W passes about 6.5 words it is measuring topic more than grammar.

“When you have four evaluations due Friday, MLU-W is the metric that gets the report written. It loses a decimal of precision versus MLU-M and saves you an hour per child — the trade-off is worth it for school-age screening.”
The pragmatist’s MLU

Free tools that compute MLU-W

MLU Calculator

Paste a language sample and get Mean Length of Utterance in morphemes and words, total utterances, total morphemes, and the matching Brown's stage. Implements Brown (1973) morpheme counting rules and runs entirely in your browser.

Open tool

SUGAR Norms Lookup

Interactive lookup for SUGAR (Pavelko & Owens 2017) language sample normative values. Enter the child's age in years and months and the tool returns the matching MLU, TNW, CPS, and MLUL means with ±1 SD typical ranges plus the full SUGAR table for context. Built for speech-language pathologists running 50-utterance samples.

Open tool

Language Sample Worksheet

Free printable and fillable language sample analysis worksheet for speech-language pathologists. Five columns (utterance #, transcription, morpheme count, grammatical Y/N, notes), configurable row count up to 100 utterances, browser print produces a clean PDF, and an inline running summary tracks total utterances, total morphemes, and rolling MLU as you fill it in.

Open tool

Related LSA metrics

MLU-M

Mean Length of Utterance in Morphemes (MLU-M)

MLU-M is the average number of morphemes per utterance and remains the single most-used index of early grammatical development in English.

TNW

Total Number of Words (TNW)

TNW is the raw word count in a language sample — the denominator behind almost every other LSA metric and a useful productivity index in its own right.

NDW

Number of Different Words (NDW)

NDW counts unique word roots across a fixed sample length and is the most stable lexical-diversity measure for school-age children.

References

  1. Pavelko, S. L., & Owens, R. E. (2017). Sampling Utterances and Grammatical Analysis Revised (SUGAR). LSHSS, 48(3), 197–215.
  2. Miller, J. F., & Chapman, R. S. (1981). The relation between age and mean length of utterance in morphemes. JSHR, 24(2), 154–161.
  3. Parker, M. D., & Brorson, K. (2005). A comparative study between mean length of utterance in morphemes (MLUm) and mean length of utterance in words (MLUw). First Language, 25(3), 365–376.