perceptual map
Peng Kangjing
6/13/2021
##### Load libraries
library(ggplot2)
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionsports<-read.csv("~/Desktop/R/sports.csv",header=T,sep=',')
sports## brand price quality fashion gender
## 1 Nike 1 3 3 Female
## 2 Nike 2 3 1 Female
## 3 Nike 3 3 3 Female
## 4 Nike 2 2 3 Female
## 5 Nike 3 2 2 Female
## 6 Nike 1 1 2 Female
## 7 Nike 4 3 4 Female
## 8 Nike 2 2 3 Female
## 9 Nike 1 1 1 Female
## 10 Nike 2 2 1 Female
## 11 Nike 1 3 2 Female
## 12 Nike 3 3 5 Female
## 13 Nike 2 4 4 Female
## 14 Nike 2 2 3 Female
## 15 Nike 2 1 2 Male
## 16 Nike 4 3 3 Male
## 17 Nike 1 1 1 Male
## 18 Nike 1 1 1 Male
## 19 Nike 2 2 2 Male
## 20 Nike 2 5 2 Male
## 21 Nike 4 1 1 Male
## 22 Nike 1 2 1 Male
## 23 Nike 2 1 2 Male
## 24 Adidas 2 2 2 Female
## 25 Adidas 1 3 3 Female
## 26 Adidas 3 3 3 Female
## 27 Adidas 2 2 3 Female
## 28 Adidas 3 3 2 Female
## 29 Adidas 1 1 4 Female
## 30 Adidas 3 4 3 Female
## 31 Adidas 2 2 3 Female
## 32 Adidas 1 1 NA Female
## 33 Adidas 2 2 2 Female
## 34 Adidas 2 2 1 Female
## 35 Adidas 3 3 4 Female
## 36 Adidas 3 4 3 Female
## 37 Adidas 2 2 2 Female
## 38 Adidas 2 1 1 Male
## 39 Adidas 4 3 3 Male
## 40 Adidas 1 1 1 Male
## 41 Adidas 2 1 2 Male
## 42 Adidas 2 2 2 Male
## 43 Adidas 3 4 3 Male
## 44 Adidas 4 2 2 Male
## 45 Adidas 1 2 1 Male
## 46 Adidas 3 3 2 Male
## 47 Reebok 3 4 4 Female
## 48 Reebok 4 4 4 Female
## 49 Reebok 4 5 6 Female
## 50 Reebok 2 2 4 Female
## 51 Reebok 4 4 4 Female
## 52 Reebok 3 2 1 Female
## 53 Reebok 5 2 5 Female
## 54 Reebok 2 2 4 Female
## 55 Reebok 4 2 5 Female
## 56 Reebok 3 3 3 Female
## 57 Reebok 4 4 4 Female
## 58 Reebok 3 4 4 Female
## 59 Reebok 4 4 4 Female
## 60 Reebok 3 3 3 Female
## 61 Reebok 4 4 4 Male
## 62 Reebok 5 5 6 Male
## 63 Reebok 3 2 4 Male
## 64 Reebok 3 1 3 Male
## 65 Reebok 4 3 5 Male
## 66 Reebok 4 5 6 Male
## 67 Reebok 3 5 5 Male
## 68 Reebok 4 2 3 Male
## 69 Reebok 5 4 1 Male
## 70 Puma 3 4 5 Female
## 71 Puma 2 5 4 Female
## 72 Puma 3 4 3 Female
## 73 Puma 2 2 4 Female
## 74 Puma 5 5 3 Female
## 75 Puma 2 2 4 Female
## 76 Puma 4 5 2 Female
## 77 Puma 3 4 6 Female
## 78 Puma 3 3 4 Female
## 79 Puma 3 3 1 Female
## 80 Puma 3 4 2 Female
## 81 Puma 2 3 3 Female
## 82 Puma 3 3 4 Female
## 83 Puma 3 3 2 Female
## 84 Puma 4 4 4 Male
## 85 Puma 4 5 5 Male
## 86 Puma 4 5 3 Male
## 87 Puma 3 1 3 Male
## 88 Puma 3 4 3 Male
## 89 Puma 4 4 5 Male
## 90 Puma 4 3 3 Male
## 91 Puma 3 2 3 Male
## 92 Puma 4 3 2 Male
## 93 FILA 2 4 7 Female
## 94 FILA 4 4 2 Female
## 95 FILA 3 3 3 Female
## 96 FILA 3 3 4 Female
## 97 FILA 5 4 1 Female
## 98 FILA 2 2 7 Female
## 99 FILA 4 4 3 Female
## 100 FILA 2 2 4 Female
## 101 FILA 3 3 3 Female
## 102 FILA 3 4 3 Female
## 103 FILA 2 2 1 Female
## 104 FILA 2 4 2 Female
## 105 FILA 2 4 2 Female
## 106 FILA 3 3 3 Female
## 107 FILA 4 4 4 Male
## 108 FILA 4 5 3 Male
## 109 FILA 1 4 3 Male
## 110 FILA 4 1 1 Male
## 111 FILA 2 3 1 Male
## 112 FILA 3 5 3 Male
## 113 FILA 1 1 1 Male
## 114 FILA 4 3 3 Male
## 115 FILA 1 2 3 Malesports<- na.omit(sports)
#female
##The horizontal axis may be interpreted as price.Brands on the left on this axis is perceived to be low-priced including Reebok,Puma and Fila. Adidas and Nike are perceived to be high-priced.
## The vertical axis may be interpreted as fashion vs quality. Reebok is perceived to be low-quality and fashion and loads on the lowest part on vertical axis.
## In the aspect of price, fashion and qulity, Adidas and Nike are percived to be similar,Puma and Fila are similar. Reebok is perceived to be unique.
female<-sports %>%
filter(gender=="Female") %>%
group_by(brand) %>%
summarize(price=mean(price),quality=mean(quality),fashion=mean(fashion))
female=data.frame(female,row.names = 1)
female## price quality fashion
## Adidas 2.230769 2.538462 2.692308
## FILA 2.857143 3.285714 3.214286
## Nike 2.071429 2.428571 2.642857
## Puma 2.928571 3.571429 3.357143
## Reebok 3.428571 3.214286 3.928571# treat as distance
female = dist(female)
female_result = cmdscale(female, k=2, eig=T)
female_result## $points
## [,1] [,2]
## Adidas 0.8150227 -0.05960631
## FILA -0.2657948 0.17455948
## Nike 1.0000684 -0.08852520
## Puma -0.5401169 0.33436531
## Reebok -1.0091793 -0.36079328
##
## $eig
## [1] 3.045215e+00 2.838326e-01 5.706996e-03 1.183411e-16 3.816392e-17
##
## $x
## NULL
##
## $ac
## [1] 0
##
## $GOF
## [1] 0.9982886 0.9982886# save results in new dataset
female_data = data.frame(female_result$points)
colnames(female_data) = c("Coordinate1", "Coordinate2")
female_data## Coordinate1 Coordinate2
## Adidas 0.8150227 -0.05960631
## FILA -0.2657948 0.17455948
## Nike 1.0000684 -0.08852520
## Puma -0.5401169 0.33436531
## Reebok -1.0091793 -0.36079328plot solution
ggplot(data = female_data, aes(x= Coordinate1, y = Coordinate2)) +
geom_point(size = 4, color = "blue") +
annotate(geom = "text", x = female_data$Coordinate1, y = female_data$Coordinate2, label = row.names(female_data), vjust = 2)+
ggtitle("female MDS")
#male
##Compared to female, male think Fila and Puma are not very similar, and Fila has higher fashion and quality than Puma.
male<-sports %>%
filter(gender=="Male") %>%
group_by(brand) %>%
summarize(price=mean(price),quality=mean(quality),fashion=mean(fashion))
male=data.frame(male,row.names = 1)
male## price quality fashion
## Adidas 2.444444 2.111111 1.888889
## FILA 2.666667 3.111111 2.444444
## Nike 2.111111 1.888889 1.666667
## Puma 3.666667 3.444444 3.444444
## Reebok 3.888889 3.444444 4.111111# treat as distance
male = dist(male)
male_result = cmdscale(male, k=2, eig=T)
male_result## $points
## [,1] [,2]
## Adidas -1.1768317 -0.14343769
## FILA -0.1900256 0.45945173
## Nike -1.6146078 -0.15526830
## Puma 1.1979600 0.07770704
## Reebok 1.7835051 -0.23845279
##
## $eig
## [1] 8.643999e+00 3.186766e-01 3.979319e-02 -2.220446e-16 -6.946148e-16
##
## $x
## NULL
##
## $ac
## [1] 0
##
## $GOF
## [1] 0.9955797 0.9955797# save results in new dataset
male_data = data.frame(male_result$points)
colnames(male_data) = c("Coordinate1", "Coordinate2")
male_data## Coordinate1 Coordinate2
## Adidas -1.1768317 -0.14343769
## FILA -0.1900256 0.45945173
## Nike -1.6146078 -0.15526830
## Puma 1.1979600 0.07770704
## Reebok 1.7835051 -0.23845279plot solution
ggplot(data = male_data, aes(x= Coordinate1, y = Coordinate2)) +
geom_point(size = 4, color = "blue") +
annotate(geom = "text", x = male_data$Coordinate1, y = male_data$Coordinate2, label = row.names(male_data), vjust = 2)+
ggtitle("male MDS")