[Corpora-List] Announcement: blog for experimental statistics in corpus linguistics

Wallis, Sean s.wallis at ucl.ac.uk
Tue May 29 11:51:13 CEST 2012

Dear colleagues

I am pleased to announce the publication of a new practical research blog aimed at addressing questions of experimental design and statistics for researchers working with corpus data.

I will be briefly introducing this blog, called corp.ling.stats (http://corplingstats.wordpress.com<http://corplingstats.wordpress.com/>) at ICAME22 in Leuven this week.

Corpus linguistics poses a number of problems for conventional statistical analysis that are not always recognised. Some problems can be addressed by changing experimental design, others, by identifying better statistical methods. A corpus linguist cannot ignore statistical methods in order to draw inferences from their corpora, and a statistician who wishes to be relevant cannot ignore linguistic problems.

The statistical methods covered have all been employed in research papers, and one or two papers in the collection cover original innovations.

Throughout the emphasis is on core methods (the scientific reliability of testable hypotheses) that hopefully bear on linguistic theory, rather than on exploratory and descriptive techniques.

The blog contains a number of posts on different subjects, several papers and Excel spreadsheets demonstrating the methods referred to.

Please do have a look at the blog, post questions or email me direct, and circulate widely.

I am open to suggestions regarding content: so far the statistical focus has been driven by research papers originating from the Survey of English Usage and collaborating colleagues.


Sean Wallis Survey of English Usage University College London -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 5149 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20120529/f47c2129/attachment.txt>

More information about the Corpora mailing list