[Corpora-List] Factor Analysis for Biber's MDA in SPSS

[Angus Grieve-Smith]

     I was also interested in Biber's multidimensional analysis, and 
tried to replicate it.  But just when I began looking at the 
implementation of it, a problem confronted me.

     The idea behind multidimensional analysis was that we have an 
intuition that linguistic variables pattern differently in different 
registers and genres, but the pattern is rarely simple enough to say 
"Genre 1 has x, y and z values, but Genre 2 has x, a and b values."  
Biber basically threw all the variables in a pot and shook it to see 
which ones patterned together.  Those factors were labeled as the 
dimensions of variation.

     The problem is that variables don't just pattern together for 
reasons of register and genre.  Sometimes they pattern together for 
grammatical reasons; for example, his Dimension 1 is essentially 
frequent noun morphology vs. frequent verb morphology, and thus could 
simply be explained by the relative frequencies of nouns vs. verbs in 
the text.

     I had an idea that this could be overcome by bringing the notion of 
the "envelope of variation" over from variation theory, and I mentioned 
it to Biber.  He was interested, but my study did not support the idea.  
If anyone else can think of a way to overcome this obstacle, I'd be 
interested to know it.

     I discussed this in greater detail in my 2006 paper; if anyone 
wants to read it but doesn't have access, please let me know.

http://www.ingentaconnect.com/content/rodopi/lang/2006/00000060/00000001/art00003?crawler=true

On 2/2/2014 10:48 AM, Muhammad Shakir Aziz wrote:
> Dear Corpora Members
> I am interested in Biber's MD Analysis. I wan to conduct a complete 
> factor analysis using Biber's original 67 variables. For this purpose, 
> MDA Tagger 
> <https://www.academia.edu/4285869/Multidimensional_Analysis_Tagger_v_1.0> is 
> very useful which has tagged my corpus with those 67 features and has 
> calculated their score in each text. Since it does not perform Factor 
> Analysis, I have provided myself with SPSS 17. As I am a beginner in 
> MD Analysis, and have least idea about using SPSS, I am requesting 
> here for assistance. I have tried to perform Factor Analysis in SPSS 
> by following steps provided in online tutorials. But as a result SPSS 
> output window doesn't show anything under Factor Analysis heading. I 
> had selected Maximum Likelihood Extraction method and Promax Rotation. 
> Is there a guide/ tutorial available to select appropriate options in 
> SPSS for MD Analysis?
> I will be highly grateful for your help.
> -- 
> *Muhammad Shakir Aziz**???? ???? ????*
> *Master in English Linguistics
> Translator, Course Developer, Linguist for Urdu, Punjabi and English*
> Urdu:- http://awaz-e-dost.blogspot.com/
> English:- http://linguisticslearner.blogspot.com/
> Facebook:- http://www.facebook.com/truefriend2004
> Skype:- true_friend2004
>
>
> _______________________________________________
> UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
> Corpora mailing list
> Corpora at uib.no
> http://mailman.uib.no/listinfo/corpora


-- 
				-Angus B. Grieve-Smith
				grvsmth at panix.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20140202/9589edfa/attachment.htm>
-------------- next part --------------
_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora