Good morning!
In correspondence analysis there are two kinds of analyses (Greenacre
1984): simple and multiple corr.analysis (CA & MCA). The first analysis
(CA) uses a cross tabulation of two variables. The second (MCA) analyses
large numbers of variables.
When I use ADE-4 and its COA (Correspondence analysis - simple
correspondence analysis) module, I can enter "indicator matrix" data
(0/1-data). This is strange. Why is it so? I was thinking that only
multiple correspondence analysis can use "indicator data".
Example CA data
21 32 34 32 24
21 23 11 23 07
00 12 31 24 23
12 11 54 23 21
Example MCA data
0 1 0 0
1 0 0 0
1 1 1 0
0 1 0 0
I think my questions are:
a) why can COA accept indicator matrix data? (This confuses me)
b) is ADE-4 module COA performing a simple correspondence analysis when I
enter indicator data?
>From a very snowy Umea, Sweden,
Kent Lofgren
University of Umea
Pedagogiska Institutionen
S-901 87 Umea
Sweden
Tel: 46 + (0)90 - 786 64 32 (office)
Fax: 46 + (0)90 - 786 66 93 (office)
Academic interests: Bourdieu, Foucault, methodology, multivariate
statistics, the social construction of the human body, the social
significance of sport, social reproduction, leisure science, pedagogy,
philosophy of science, power and knowledge, Wittgenstein
----------
This archive was generated by hypermail 2b30 : Sat Feb 10 2001 - 10:21:43 MET