Return to Python

MLU Calculator

Linguistics MLU Calculator

In Linguistics, MLU (Mean Length of Utterance) is a measure used to represent linguistic proficiency. It is often used to track language development in children to ensure that they are on the right track to communicating verbally. In order to calculate MLU, you must take a spoken sample and count how many morphemes and utterances are contained within it. From there, you take the amount of morphemes and divide it by the amount of utterances to get your MLU- or the average amount of morphemes per utterance.

A morpheme is the smallest meaningful unit of a word, including roots, prefixes, and suffixes. For example, the word reusable can be broken down into the following morphemes: re-use-able. A morpheme can also include the english plural marker -s, past tense marker -ed, and irregular forms. An irregular example would be the word ‘was’- which still counts as two morphemes even though it is technically broken down into one visible component.

In this case, an utterance is simply a sentence or statement. So for the code, it looks for a sentence end with . , ? , or !

So let’s walk through a sample MLU Calculation with this sample:

A cat eyed a slice of pizza in a garbage can. The cat accidentally knocked down the garbage can and it caused a loud crash. All of a sudden, some rats scurried down the dim alley. The cat silently followed the rats to catch a tasty snack.

First, the code returns the morphemic breakdown of the input.
Next, you get the results.

We end up with an MLU of 14. The code simply counts up the morphemes and utterances and divides them for us in order to get the MLU. 56 morphemes divided by 4 utterances gives 14.