How To Calculate Mlu

Calculate the Mean Length of Utterance (MLU) for language development assessment. Enter the child’s speech sample below to get an accurate MLU score.

Comprehensive Guide: How to Calculate MLU (Mean Length of Utterance)

Mean Length of Utterance (MLU) is a fundamental measure in linguistics and speech-language pathology used to assess language development in children. Developed by Roger Brown in 1973, MLU provides a quantitative measure of syntactic complexity by calculating the average number of morphemes per utterance in a speech sample.

Why MLU Matters in Language Development

MLU serves several critical functions in language assessment:

• Developmental Benchmark: MLU correlates strongly with chronological age in typically developing children, making it a reliable indicator of language progression.
• Diagnostic Tool: Clinicians use MLU to identify potential language delays or disorders when scores fall significantly below age expectations.
• Treatment Planning: MLU helps speech-language pathologists design appropriate intervention strategies by quantifying current syntactic abilities.
• Research Applications: Linguists and psychologists use MLU in studies of language acquisition across different languages and cultures.

The Science Behind MLU Calculation

MLU is calculated using the formula:

MLU = Total number of morphemes ÷ Total number of utterances

Key Components

• Morphemes: The smallest meaningful units in a language (e.g., “cats” contains two morphemes: “cat” + “s”)
• Utterances: A single speech turn bounded by pauses or changes in speaker
• Speech Sample: Typically 50-100 utterances collected in naturalistic settings

Brown’s Morphemes

The standard English MLU calculation follows Brown’s (1973) system which includes:

• Free morphemes (standalone words)
• Bound morphemes (prefixes/suffixes like -ing, -s, -ed)
• Special counting rules for contractions and irregular forms

Step-by-Step Guide to Calculating MLU

1. Collect a Speech Sample

   Record or transcribe 50-100 utterances from the child in a natural setting. Ideal samples include:

   • Play sessions with caregivers
   • Storytelling activities
   • Daily routines (meals, bath time)
   • Interactions with peers

   Pro Tip: Use a digital recorder to capture accurate samples, then transcribe verbatim including false starts and repetitions.

2. Segment the Sample into Utterances

   An utterance is defined as:

   • A single speech turn by the child
   • Bounded by pauses or changes in speaker
   • May be a single word, phrase, or complete sentence

   Example segmentation:

   Child: “Want cookie” [utterance 1] • “Mommy give” [utterance 2] • “Big one!” [utterance 3]

3. Count Morphemes in Each Utterance

   Apply Brown’s morpheme counting rules:

   Morpheme Type	Example	Count
   Regular noun plural	“dogs”	2 (dog + s)
   Third person singular	“runs”	2 (run + s)
   Past tense -ed	“walked”	2 (walk + ed)
   Present progressive -ing	“running”	2 (run + ing)
   Possessive ‘s	“daddy’s”	2 (daddy + ‘s)
   Articles	“a”, “the”	1 each
   Auxiliary verbs	“is”, “are”, “was”	1 each
   Contracted forms	“don’t”	2 (do + not)

4. Calculate the Average

   Sum all morphemes across utterances and divide by the total number of utterances:

   Example calculation for three utterances:

   • “I want cookie” = 3 morphemes
   • “Mommy give” = 2 morphemes
   • “Big one” = 2 morphemes

   Total morphemes = 7 ÷ 3 utterances = MLU of 2.33

5. Interpret the Results

   Compare the MLU score to developmental norms:

   Age (months)	Typical MLU Range	Language Stage
   18-24	1.0 – 1.5	Single words
   24-30	1.5 – 2.0	Two-word combinations
   30-36	2.0 – 2.5	Simple sentences
   36-42	2.5 – 3.0	Complex sentences
   42-48	3.0 – 3.5	Embedded clauses
   48+	3.5+	Adult-like syntax

   Note: These are approximate ranges. Individual variation is normal, and cultural/linguistic differences may affect scores.

Common Challenges in MLU Calculation

Unintelligible Utterances

When parts of the speech sample are unclear:

• Exclude completely unintelligible utterances
• For partially intelligible utterances, count only the clear morphemes
• Note the percentage of unintelligible speech in your report

False Starts and Repetitions

Handling disfluencies:

• False starts: “I want… can I have cookie?” → count “can I have cookie”
• Repetitions: “I I I want it” → count as “I want it” (2 morphemes)
• Self-corrections: “He go went there” → count “went” not “go”

Dialectal Variations

For non-standard dialects:

• Use dialect-specific morpheme rules when available
• Document the dialect/variety in your report
• Compare to norms from similar linguistic communities

MLU Across Different Languages

While MLU was originally developed for English, researchers have adapted it for other languages:

Language	Key Differences	Example MLU at 36 months
English	Rich inflectional morphology (plurals, tense markers)	2.5-3.0
Spanish	Verb conjugations add multiple morphemes per word	3.0-3.5
Mandarin	Tonal language with minimal inflectional morphology	1.8-2.2
French	Complex verb morphology but many silent morphemes	2.2-2.7
German	Compound words and case markings increase morpheme count	2.8-3.3

For accurate cross-linguistic comparisons, researchers often use MLU-w (mean length of utterance in words) instead of morpheme-based MLU to control for morphological differences between languages.

Advanced Applications of MLU

Beyond basic assessment, MLU has several advanced applications:

• Tracking Intervention Progress:

  SLPs use MLU to measure changes in syntactic complexity before and after language intervention. A study by Paul & Smith (1993) found that children with specific language impairment showed an average MLU increase of 0.8 morphemes after 6 months of targeted syntax therapy.

• Bilingual Language Development:

  Research shows that bilingual children’s MLU scores may initially lag behind monolingual peers but typically converge by age 5 (Paradis et al., 2006). Clinicians should:

  • Assess MLU separately for each language
  • Consider combined MLU across languages (controversial but sometimes used)
  • Compare to bilingual norms when available
• Autism Spectrum Disorder Assessment:

  MLU patterns in ASD often show:

  • Higher MLU for formulaic/echolalic speech
  • Lower MLU for spontaneous, novel utterances
  • Atypical morpheme acquisition sequences

  A 2018 study in Journal of Autism and Developmental Disorders found that MLU in spontaneous speech below 2.0 at age 4 was 89% sensitive and 85% specific for identifying ASD in verbal children.

Limitations of MLU

While MLU is a valuable tool, clinicians should be aware of its limitations:

1. Morpheme Counting Variability:

   Different clinicians may count morphemes differently, especially for irregular forms. Inter-rater reliability averages 85-90% in research studies (Eisenberg et al., 2001).

2. Sample Size Dependence:

   MLU scores stabilize with larger samples. Research shows that:

   • 50 utterances: ±0.3 MLU variability
   • 100 utterances: ±0.15 MLU variability
   • 200 utterances: ±0.08 MLU variability
3. Ceiling Effects:

   MLU becomes less sensitive for children with MLU > 4.0, as it doesn’t capture complex syntax like:

   • Embedded clauses (“The dog that chased the cat…”)
   • Passive constructions (“The ball was thrown…”)
   • Conjunctions (“I went to the store and then…”)

   For advanced language users, clinicians often supplement MLU with measures like:

   • Clausal density
   • Subordination index
   • Sentence complexity scores

Best Practices for MLU Assessment

Sample Collection

• Use naturalistic contexts (play, meals, routines)
• Aim for 100+ utterances when possible
• Record during peak alertness times
• Include familiar communication partners

Transcription

• Transcribe immediately after recording
• Use phonetic transcription for unclear words
• Mark unintelligible segments clearly
• Include nonverbal vocalizations that serve communicative functions

Analysis

• Double-check morpheme counts
• Calculate MLU for different contexts separately
• Compare to multiple norm references
• Consider qualitative analysis alongside quantitative MLU

MLU in Research: Key Studies

The following foundational studies have shaped our understanding of MLU:

1. Brown (1973)

   The original MLU study established:

   • MLU correlates with age (r = 0.91)
   • Morpheme acquisition follows predictable sequence
   • MLU stages correspond to grammatical development

   Original paper: Brown, R. (1973). A first language: The early stages. Harvard University Press.

2. Miller & Chapman (1981)

   Expanded MLU research by:

   • Developing age norms for MLU in English
   • Establishing reliability measures
   • Comparing MLU to other language measures

   Key finding: MLU accounts for 60-70% of variance in syntactic complexity scores.

3. Rice et al. (2010)

   Longitudinal study showing:

   • MLU growth trajectories predict later literacy skills
   • Children with slow MLU growth (≤0.1/month) at risk for reading difficulties
   • MLU at age 3 correlates with narrative skills at age 6 (r = 0.72)

   Study link: Rice, M. L., et al. (2010). Language growth and grammar in children with specific language impairment.

Frequently Asked Questions About MLU

Q: How often should MLU be measured?

A: For typically developing children, every 6 months is sufficient. For children in intervention, every 3 months allows for closer progress monitoring. More frequent measurement (monthly) may be warranted during periods of rapid language growth or intensive therapy.

Q: Can MLU be used for adults with aphasia?

A: While originally designed for children, MLU has been adapted for adult aphasia assessment. Research shows that:

• MLU correlates with Western Aphasia Battery scores (r = 0.78)
• MLU < 3.0 often indicates significant syntactic impairment
• Combined with other measures, MLU helps track recovery progress

However, adult MLU interpretation requires different normative data than child MLU.

Q: What’s the difference between MLU and MLU-w?

A: MLU counts morphemes while MLU-w (mean length of utterance in words) counts whole words. Key differences:

Feature	MLU	MLU-w
Unit of measurement	Morphemes	Words
Language specificity	High (varies by morphology)	Lower (more cross-linguistic comparability)
Sensitivity to inflections	High	Low
Typical values at 36 months	2.5-3.0	3.0-4.0
Best for	Detailed syntactic analysis	Quick screening, cross-linguistic studies

Tools and Resources for MLU Calculation

Professionals can use these resources to enhance MLU assessment:

• Systematic Analysis of Language Transcripts (SALT):

  The gold standard software for language sample analysis, including automated MLU calculation. Visit SALT website

• Child Language Data Exchange System (CHILDES):

  Large database of child language transcripts with MLU calculation tools. Explore CHILDES

• ASHA MLU Norms:

  The American Speech-Language-Hearing Association provides clinical norms and guidelines. ASHA Practice Portal

• MLU Calculator Apps:

  Several mobile apps (iOS/Android) offer quick MLU calculation for clinical use, though manual verification is recommended.

Future Directions in MLU Research

Emerging areas of study include:

• Automated MLU Calculation:

  Natural language processing algorithms are being developed to:

  • Automatically transcribe and segment speech samples
  • Calculate MLU with 90%+ accuracy compared to human raters
  • Provide real-time MLU feedback during therapy sessions
• MLU in Digital Communication:

  Researchers are studying how MLU manifests in:

  • Text messages and social media posts
  • Augmentative and alternative communication (AAC) systems
  • Virtual reality communication environments
• Neurolinguistic Correlates:

  Advanced neuroimaging studies are exploring:

  • Brain regions activated during MLU growth spurts
  • Neural predictors of MLU development
  • Differences in brain connectivity between typical and atypical MLU trajectories

Conclusion: The Enduring Value of MLU

Nearly five decades after Roger Brown’s seminal work, MLU remains one of the most robust and clinically useful measures of language development. Its simplicity belies its power to:

• Capture the essence of syntactic development in a single number
• Provide a common metric across languages and cultures
• Serve as both a screening tool and progress monitor
• Bridge research and clinical practice

While newer, more complex measures of language have been developed, MLU’s accessibility and strong empirical foundation ensure its continued relevance. For speech-language pathologists, educators, and researchers, mastering MLU calculation and interpretation remains an essential skill in understanding and supporting language development.

As our understanding of language grows, MLU continues to evolve—from its origins in tape-recorded samples to potential future applications in AI-powered language analysis. Yet its core purpose remains the same: to quantify the beautiful complexity of human language acquisition, one morpheme at a time.