how to calculate mean length of utterance

Understanding Mean Length of Utterance (MLU)

Mean Length of Utterance (MLU) is a vital metric in linguistics, particularly in the study of child language development. It measures the average number of morphemes a speaker produces in an utterance. Often used by speech-language pathologists, developmental psychologists, and researchers, MLU provides a reliable indicator of a child's linguistic complexity and syntactic development.

A higher MLU generally suggests more advanced language abilities, as it reflects the capacity to combine more morphemes into longer, more complex sentences. While it's most commonly associated with children, MLU can also be used to analyze language samples from adults or individuals with language impairments.

What is an Utterance?

In the context of MLU calculation, an utterance is a natural unit of speech bounded by pauses, changes in intonation, or a new speaker's turn. It can be a single word, a phrase, or a complete sentence. The key is that it represents a single, cohesive thought or communicative act.

• "Go." (1 utterance)
• "I want cookie." (1 utterance)
• "That's a big dog!" (1 utterance)

What is a Morpheme?

A morpheme is the smallest meaningful unit of language. Unlike a word, a morpheme cannot be broken down into smaller parts that still carry meaning. Morphemes can be:

• Free Morphemes: Can stand alone as words (e.g., "cat," "run," "happy").
• Bound Morphemes: Must be attached to a free morpheme to have meaning. These include prefixes, suffixes, and grammatical endings (e.g., "-s" for plural, "-ed" for past tense, "un-" for negation).

For MLU calculation, every meaningful unit counts. For example, "running" has two morphemes: "run" (free) and "-ing" (bound). "Cats" has "cat" (free) and "-s" (bound).

Step-by-Step Guide to Calculating MLU

Calculating MLU involves a systematic process to ensure accuracy and consistency. Here’s how to do it:

1. Collect a Language Sample

Record a spontaneous speech sample from the individual. For children, this usually involves engaging them in natural play or conversation. A typical sample might be 50-100 utterances, usually collected over 15-30 minutes.

2. Transcribe the Sample

Carefully transcribe the recorded speech, noting every word, sound, and pause. Standard transcription conventions are often used to ensure consistency.

3. Segment Utterances

Divide the transcribed text into individual utterances. This is a critical step and often requires practice. General rules for segmentation include:

• Pauses (usually 2 seconds or more).
• Changes in speaker.
• Clear shifts in thought or topic.
• Intonation patterns indicating a complete thought (e.g., a question or statement).
• Repetitions or false starts are generally excluded or treated carefully.

4. Count Morphemes per Utterance

For each segmented utterance, count the number of morphemes. Remember to count both free and bound morphemes. Common rules for English include:

• Root words (e.g., "dog," "eat," "big") count as 1 morpheme.
• Plural "-s" (e.g., "dogs") counts as 1 additional morpheme.
• Possessive "-'s" (e.g., "Mommy's") counts as 1 additional morpheme.
• Past tense "-ed" (e.g., "walked") counts as 1 additional morpheme.
• Present progressive "-ing" (e.g., "running") counts as 1 additional morpheme.
• Third-person singular "-s" (e.g., "he runs") counts as 1 additional morpheme.
• Derivational prefixes (e.g., "un-," "re-") and suffixes (e.g., "-ly," "-ness") count as 1 additional morpheme.
• Contractions (e.g., "it's," "don't") typically count as two morphemes (it + is, do + not).

5. Sum Totals

After counting morphemes for each utterance, sum the total number of morphemes across all utterances in your sample. Also, sum the total number of utterances.

6. Apply the Formula

Finally, calculate the MLU using the simple formula:

MLU = Total Number of Morphemes / Total Number of Utterances

An Example Calculation

Let's consider a small language sample from a child:

• Utterance 1: "Doggy run." (Doggy - 1, run - 1 = 2 morphemes)
• Utterance 2: "I want cookies." (I - 1, want - 1, cookie - 1, -s - 1 = 4 morphemes)
• Utterance 3: "He's playing." (He - 1, is - 1, play - 1, -ing - 1 = 4 morphemes)

In this example:

• Total Morphemes = 2 + 4 + 4 = 10
• Total Utterances = 3
• MLU = 10 / 3 = 3.33

This MLU of 3.33 gives us an insight into the child's average utterance length and complexity.

Interpreting MLU Scores

MLU typically increases with age during the early years of language acquisition. It's often compared to normative data, such as Brown's Stages of Syntactic Development, to gauge a child's language progress. For instance:

• MLU 1.0-2.0: Early multi-word combinations.
• MLU 2.0-2.5: Emergence of grammatical morphemes (e.g., -ing, plural -s).
• MLU 2.5-3.0: Simple sentence forms, auxiliary verbs.
• MLU 3.0-3.5: More complex sentence structures, questions, negatives.
• MLU 3.5-4.0+: Development of complex sentences, embedded clauses.

It's important to note that these are general guidelines, and individual development can vary.

Important Considerations and Limitations

While MLU is a valuable tool, it's not without its nuances:

• Age Appropriateness: MLU is most informative for children between 18 months and 5 years. Beyond this age, utterances become so long and complex that MLU loses some of its sensitivity as a measure of grammatical complexity.
• Language Variations: The rules for morpheme counting can vary slightly depending on the specific dialect or language being analyzed.
• Transcription Accuracy: The accuracy of the MLU calculation heavily relies on the precision of the transcription and morpheme segmentation. Inconsistencies can lead to skewed results.
• Not a Sole Measure: MLU should be considered alongside other measures of language development, such as vocabulary size, semantic complexity, and pragmatic skills, for a complete picture.

Conclusion

Calculating Mean Length of Utterance is a foundational skill for anyone working with language development. By systematically analyzing language samples, we gain quantitative insights into a speaker's evolving linguistic abilities. While it requires careful attention to detail in transcription and morpheme counting, the MLU provides an invaluable window into the complexity of human language.