current interest in large corpora and CDA

[John E Richardson]

hi Linnea,

I agree, it would be great to develop a macro-CDA research method,
though I am not entirely sure that this would be possible. Or, if it
were, it would probably be called something other than CDA like corpus
linguistics or content analysis, or something else. 
As others have said before, CDA is more of an approach to social
analysis, it is a problem centred research paradigm, rather than a
method. CDA examines the ways that language use is implicated in the
reproduction of inequality. Clearly there are a number of ways that we
can examine how this occurs, and hence different research methods (and
their accompanying methodologies) that need to be applied. I think that
the best research, whether CDA or not, employs a combination of research
methods to explore a research question or a research problem, and while
certain methods are useful in exploring meaning (or bias) across a
sample of texts (e.g. content analysis), others are far more suitable
for exploring meaning within texts (CA, S-F grammar & transitivity) or
examining text-context relations (e.g. rhetorical analysis). You already
know about my work on pre-2001 representations of Islam, but I have
recently completed a content analysis of headlines reporting the
invasion of Iraq. Here I coded the verb process, actor, object, tense &
a few other variables of 2100 headlines. The results are pretty
interesting & I'm writing them up at the moment. 

all the best

John


> Hi Cicero,
> 
> Welcome to the list!  I was actually waiting for someone more 
> knowledgeable than I about post-structuralism to reply to your question, 
> so I was glad that John did.  :)  
> 
> What other areas of CDA are you interested in?  I have been looking at 
> the representation of Muslims in U.S. print media, focusing on the years 
> prior to 2001, and exploring methods of macro-analysis across large 
> corpora.   I'm quite interested in some of the criticism of "bias" in 
> CDA: while I agree that all research is biased (in that researchers 
> always approach their area of study with an agenda), I think it would be 
> useful to develop approaches to CDA that balance micro- and 
> macro-analysis, so that when analyses are made about selected portions 
> of a text, statistics can also be provided about the prevalence of 
> patterns on a larger scale. 
> 
> I would be delighted to hear any suggestions people have about using 
> large corpora for CDA.
> 
> Linnea
> 
> 
>  


John E Richardson
Dept of Journalism Studies
Sheffield University