Standard measures

In this excercise we shall go through some basic commands and standard measures that are useful for analyzing children's conversation. You will use an interface (see the bottom of this page) which provides various options for searching and viewing the given files. The files originally belong to CHILDES (CHIld Language Data Exchange System). Every file starts with some information about the circumstances of the taped conversation and about the participants. These lines start with '@' (the '@-tiers').

a) Give the name of the target child.

b) How old was the child when the recordings where made (yy/mm/dd)?
First recording:


Second recording:

Third recording:

Within research there are some standard measures which are being used a lot:
MLU: Mean Length of Utterance (words)
MLT: Mean Length of Turn
TTR: Type-Token Ratio
MLU is a quantitative measure for morpho-syntactic complexity. It indicates how many morphemes/words a child uses per utterance. In other words: a higher value of the MLU means that a child is capable of producing more complex sentences. In the acquisition of Dutch the MLU for morphemes and words are corelated. Therefore, we can restrict ourselves to words in this exercise. The value of MLU does not say tell us anything about the way in which the complexity of utterances is expressed; you still do not know whether a child produces more adverbial phrases or perhaps even embedded clauses in its speech.
MLT is the sort of measure as MLU, the only difference is that it is calculated over turns instead of utterances.
TTR is the number of different words (types) divided by the total number of words of a certain amount of utterances. It is an indicator of lexical diversity.

c) Give the MLU, MLT and TTR in mar20511, only for the child's utterances.


MLU:

MLT:

TTR:

As you have already seen, the page with the CHILDES-files contains several buttons, drop-down menu's and empty fields. The first menu, File, contains a selection of useful files for the exercise. The second one, Mode/Analysis, contains several options for making calculations and analyses. The one you probably have not used yet is the Frequency analysis; this option shows a list of all the words used by the participants and the frequency in which they are used.
If you want to use just a part of the file you can enter the number of the first and of the last utterance of your analysis in the fields Start at and End at respectively. Note that when you use the Participant filter, the Word filter or the begin/end option a second column of line numbers appears on the left of the utterances. The first column displays the line numbers of the output you can see, the second column with numbers between parentheses, shows the line numbers of the original file.

d) What is the MLU of the child and mother in the file mar10720 from line 60 to 310?
Child:


Mother:

e) How many times does the child say "niet" (or a variant) in mar20511?


And the father?

Note: it may be interesting to see in which context someone uses a certain word or phrase. If you want to see some of the context around the target word and its sentence, use the fields Context pre-/post- select.

Back
File:
Mode/Analysis:
Start at:
End at:
Word filter:
This can be a comma seperated list
Note you can use *
Show act data?
Show com data?