Pitfalls in Corpus Research.
This paper discusses some pitfalls in corpus research and suggests solutions on the basis of examples and computer simulations. We first address reliability problems in language transcriptions, agreement between transcribers, and how disagreements can be dealt with. We then show that the frequencies...
| Published in: | Computers & the Humanities Vol. 38; no. 4; pp. 343 - 363 |
|---|---|
| Main Authors: | , , |
| Format: | Article |
| Published: |
Springer Nature
Nov2004
|
| Subjects: | |
| Online Access: | View this record in EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=15825065&site=ehost-live header: @attributes: shortDbName: hlh uiTerm: 15825065 longDbName: Humanities International Complete uiTag: AN controlInfo: bkinfo: jinfo: jid: 00104817 CHM jtl: Computers & the Humanities issn: 00104817 maglogo: N pubinfo: dt: Nov2004 vid: 38 iid: 4 pid: 237 pub: Springer Nature artinfo: ui: 15825065 10.1007/s10579-004-1919-1 ppf: 343 ppct: 20 formats: fmt: @attributes: type: P size: 165KB tig: atl: Pitfalls in Corpus Research. aug: au: Rietveld, Toni Van Hout, Roeland Ernestus, Mirjam affil: Department of Linguistics, Radboud University Nijmegen, Erasmusplein 1, 6525 HT, Nijmegen, The Netherlands. Max Planck Institute for Psycholinguistics. su: Computer simulation Transcription (Linguistics) Statistical reliability Word frequency Statistical hypothesis testing Sequential analysis sug: subj: Computer simulation Transcription (Linguistics) Statistical reliability Word frequency Statistical hypothesis testing Sequential analysis keyword: χ² analysis ?2 analysis agreement between transcribers corpus research effect size log odds power of a test sequential dependence unit dependence ab: This paper discusses some pitfalls in corpus research and suggests solutions on the basis of examples and computer simulations. We first address reliability problems in language transcriptions, agreement between transcribers, and how disagreements can be dealt with. We then show that the frequencies of occurrence obtained from a corpus cannot always be analyzed with the traditional X² test, as corpus data are often not sequentially independent and unit independent. Next, we stress the relevance of the power of statistical tests, and the sizes of statistically significant effects. Finally, we point out that at-test based on log odds often provides a better alternative to a X² analysis based on frequency counts. pubtype: Academic Journal doctype: Article src: R language: English refInfo: copyright: @attributes: flag: Y custom: Computers & the Humanities is a copyright of Springer, 2004. All Rights Reserved. item: Computers & the Humanities holder: Springer Nature dt: @attributes: year: 2004 holdings: @attributes: islocal: N |
|---|