I want to produce a graph for the relative positioning of six periodicals: “El Pais”, “Expansion”, “Hola”, “National Geographic”, “Mia”, “Viajar”. A certain number of people are asked which publications they read.
library(ca)
public = file.choose()
public # The data content is a text string. In our case we select the file whose path is shown below. In each case, the output depends on the file structure of each disk.
## [1] "C:\\Users\\Alex\\Desktop\\periodicos.txt"
periodicos = read.table(file=public,header=T)
periodicos
## ElPais Expansion Hola NationalG Mia Viajar
## ElPais 500 300 20 40 5 40
## Expansion 300 400 10 50 4 50
## Hola 20 10 600 15 300 80
## NationalG 40 50 15 200 10 150
## Mia 5 4 300 10 400 40
## Viajar 40 50 80 150 40 250
I analyze the meaning of the values in the table above. In the first row, the value 500 indicates that 500 people read El Pais; 300 read El País and also Expansión; 20 The Country and Hello, etc.
plot(ca(periodicos))
The center means the least differentiated place. If a product is located there, that means that it is in the market average; that is, it does not have any particularly significant characteristics. The further from the center, the more differentiated. If we were to draw lines from the center to the groups of posts that we see on the map, those lines signify feature axes. In our case, such axes can be identified with “Information”, “home / society” and “travel”.