Developmental Measures#

Several developmental measures are defined as Reader methods:

mlu([participant, exclude_switch])

Return the mean lengths of utterance (MLU).

mlum([participant, exclude_switch])

Return the mean lengths of utterance in morphemes.

mluw([participant, exclude_switch])

Return the mean lengths of utterance in words.

ttr([keep_case, participant])

Return the type-token ratios (TTR).

ipsyn([participant])

Return the indexes of productive syntax (IPSyn).

For the mean lengths of utterance (MLU) in Eve’s data from Brown:

>>> import pylangacq
>>> url = "https://childes.talkbank.org/data/Eng-NA/Brown.zip"
>>> eve = pylangacq.read_chat(url, "Eve")
>>> eve.mlu()
[2.309041835357625,
 2.488372093023256,
 2.8063241106719365,
 2.618803418803419,
 2.8852691218130313,
 3.203358208955224,
 3.179732313575526,
 3.4171011470281543,
 3.8439306358381504,
 3.822669104204753,
 3.8814317673378076,
 4.177847113884555,
 4.2631578947368425,
 3.9936974789915967,
 4.457182320441989,
 4.416536661466458,
 4.501661129568106,
 4.288242730720607,
 4.3813169984686064,
 3.3172541743970316]