Hi all
I am a lil bit confused about these two multivariate methods. Both of them are developed to handle categorical variables. My problem is as follows.
I conducted a questionnaires about the recreational enjoyment for a rural area. So I got columns for the sites, rows for the activities, and the count numbers of these 2 variables. More over, I am also interested in exploring potential relationships between respondents' social and demographic characteristics (age, occupation, ethic-groups) and their choice of places and activities.
In order to handle these questions, what might be the best multivariate method ? or other statistical methods ?
Thank you for your time reading this question.
I suggest to let the sites be the rows and the activities the columns. Apply CA to this matrix. The social and demographic characteristics may be introduced as supplementary variables (columns).
Jorge Ortiz
You use Correspondence Analysis for two categorical (nominal/ordinal) variables and MCA for more than two categorical variables. But you have count data, it might not be the right method to use with it. Your question could be better answered by regression analysis.
thank you for your answers. what i mean by count data is the contingency table. places vs activities. I think CA plus supplementary variables might be appropriate
- Badges
- Science topic
- Similar topics
- Biological Science
- Zoology
- Ornithology