![[Search]](/static/images/search.gif){kind=small}
![[A-Z Index]](/static/images/az.gif){kind=small}
![[Contact]](/static/images/contact.gif){kind=small}
![[University of Cambridge]](/static/images/identifier2.gif){kind=small}
![[Talks.cam]](/static/images/talkslogosmall.gif)
Toni Rietveld, Radboud University Nijmegen
Wednesday 20 October 2010, 16:00-17:30
Statistical pitfalls in corpus research
- đ¤ Speaker: Toni Rietveld, Radboud University Nijmegen
- đ Date & Time: Wednesday 20 October 2010, 16:00 - 17:30
- đ Venue: GR-06/07, English Faculty Building
Questions? Contact
Chris Cummins
Abstract
The availability of large memory spaces in computers has promoted the creation of corpora of normal and pathological speech and language samples. The analysis of these corpora demands different strategies to those used in balanced factorial designs.
In this talk I will discuss four topics: 1) the incorrect use of “significance” as an index of effect size, 2) the importance of indices of agreement between transcribers, 3) the danger of unit dependencies in the analysis of data obtained in the context of dyadic communication and 4) a way of coping with sequential dependencies in the analysis of corpus data.
Series This talk is part of the RCEAL Tuesday Colloquia series.
Included in Lists
Note: Ex-directory lists are not shown.