[R] Alternative for correspondance analysis

Sannr SanderSmit77 at hotmail.com
Thu Sep 3 10:18:15 CEST 2009


I am looking for an alternative way to inspect my data, other than doing an
correspondance analysis.

What I have is a list with 5 different measurements of a person and ratings
that that person gave to a number of objects, so:

person1.name, person1.score1, person1.score2, person1.score3,
person1.score4, person1.score5, person1.ratingobject1,
person1.ratingobject2, person1.ratingobject3, person1.ratingobject4, ...

person2.name, person2.score1, person2.score2, person2.score3,
person2.score4, person2.score5, person2.ratingobject1,
person2.ratingobject2, person2.ratingobject3, person2.ratingobject4, ...

person3.name, person3.score1, person3.score2, person3.score3,
person3.score4, person3.score5, person3.ratingobject1,
person3.ratingobject2, person3.ratingobject3, person3.ratingobject4, ...

etc...

Now I want to see how those 5 measurements match to certain high ratings for
certain objects, so for example a character type high on score1,score2 and
score3 and low on score5 & score6 generally rate object1 very high and
object 2 very low. I want to do this using R.

I thought about using correspondance analysis to tackle the problem, but
this means I will need to first group those people. I prefer not to do so,
because I feel I am loosing information when I do that. Is there any
alternative I could use to measure correspondances in the data?

Any suggestion is very welcome!
-- 
View this message in context: http://www.nabble.com/Alternative-for-correspondance-analysis-tp25271624p25271624.html
Sent from the R help mailing list archive at Nabble.com.




More information about the R-help mailing list