[Corpora-List] Statistical tests for corpus studies

[Rayson, Paul]

Adam, Josephine,

But there is a problem with the Mann-Whitney test of too many zeros in the slices, as your IJCL paper points out Adam. For example, in the LOB and Brown comparison only words with a frequency of 30 or more (in the joint corpus) had few enough zeros for the test to be applicable. This means that 92% of the word types in the joint corpus were omitted from the comparison.

Regards,
Paul.

Dr. Paul Rayson
Director of UCREL
University Centre for Computer Corpus Research on Language
Computing Department, Lancaster University,
Lancaster, LA1 4YR, UK.
Web: http://www.comp.lancs.ac.uk/computing/users/paul/
Tel: +44 1524 593786  Fax: +44 1524 593608


-----Original Message-----
From: Adam Kilgarriff [mailto:adam.kilgarriff at itri.brighton.ac.uk]
Sent: 07 May 2003 10:45
To: Josephine Lo
Cc: CORPORA at HD.UIB.NO
Subject: Re: [Corpora-List] Statistical tests for corpus studies


Josephine,

chi-square will probably not give you what you want, nor will 
log-likelihood - my paper on "Comparing Corpora" (Int Jnl Corpus 
Linguistics 2001) explains why.  Non-parametric tests are more suitable, 
I found the Mann-Whitney test did the job well.  It involves chopping 
each corpus up into same-size slices.

Regards,

    Adam

Josephine Lo wrote:

> Dear all,
>
> As a lay-man to statistics, I wish to get some advice on the tests 
> suitable for comparing the frequency of a specific type of word in 
> corpora of different genre. Having in mind are Chi-square and ANOVA 
> but I'm not sure they are the appropriate ones.
>
> Thanks in advance
>
>
> Josephine Lo
> Research Assistant
> Dept. of English and Communication
> City University of Hong Kong
>

-- 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Adam Kilgarriff                        
ITRI, University of Brighton                   tel: (44) 1273 642919        
Lewes Road, Brighton BN2 4GJ, UK               fax: (44) 1273 642908
adam at itri.bton.ac.uk     http://www.itri.bton.ac.uk/~Adam.Kilgarriff
  and                                 
Lexicography MasterClass Ltd.
71 Freshfield Road, Brighton BN2 0BL, UK       tel: (44) 1273 705773 
adam at lexmasterclass.com                http://www.lexmasterclass.com
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%