WDSIZE — Word Size Distribution
Status: Current
Character-length histogram for word stems from the %mor tier.
Usage
chatter clan wdsize file.cha
chatter clan wdsize corpus/ --speaker CHI
chatter clan wdsize file.cha --main-tier # Use main tier words instead
chatter clan wdsize file.cha --format json
What It Measures
WDSIZE counts the character length of each word stem extracted from the %mor tier and produces a histogram showing how many words of each length appear. By default it uses morphological stems (lemmas); with --main-tier it uses surface forms from the main tier.
Output
Per speaker:
- Character-length histogram (length → count)
- Total words measured
- Mean word size in characters
Differences from WDLEN
| Feature | WDSIZE | WDLEN |
|---|---|---|
| Default source | %mor stems | Main tier words |
| Sections | 1 (character lengths only) | 6 (chars, words/utt, turns, morphemes) |
| Morpheme counting | No | Yes (sections 5-6) |
Differences from CLAN
- Uses typed
MorTieritems withMorWord.lemmarather than raw string parsing - Supports JSON and CSV output
- Falls back to main tier words when
%moris absent