Perception
Blanka
2025-02-03
Perception Map
The principal component analysis was performed on 18 standardized variables (n = 152). We wanted to see how cash, card, and mobile payment methods are perceived in terms of safety, speed of transaction, ease of use, convenience, privacy, and the ability to track expenses. The respondents evaluated these on the Likert scale, where 1 = strongly disagree and 7 = strongly agree.
library(readxl)
file.exists("./Data_Perception.xlsx")## [1] TRUEmydata <- read_xlsx("./Data_Perception.xlsx")colnames(mydata) <- c("Cash_Safety", "Card_Safety", "Mobile Payment_Safety", "Cash_Speed", "Card_Speed", "Mobile Payment_Speed", "Cash_Ease of Use", "Card_Ease of Use", "Mobile Payment_Ease of Use", "Cash_Convenience", "Card_Convenience", "Mobile Payment_Convenience", "Cash_Privacy", "Card_Privacy", "Mobile Payment_Privacy", "Cash_Tracking Expenses", "Card_Tracking Expenses", "Mobile Payment_Tracking Expenses")library(psych)
psych::describe(mydata[ ,-1])## vars n mean sd median trimmed mad min max
## Card_Safety 1 152 5.34 1.21 6 5.43 1.48 2 7
## Mobile Payment_Safety 2 152 5.26 1.49 6 5.44 1.48 1 7
## Cash_Speed 3 152 4.48 1.60 5 4.48 1.48 1 7
## Card_Speed 4 152 5.78 1.07 6 5.89 1.48 1 7
## Mobile Payment_Speed 5 152 5.79 1.41 6 6.05 1.48 1 7
## Cash_Ease of Use 6 152 5.63 1.49 6 5.84 1.48 2 7
## Card_Ease of Use 7 152 6.08 1.03 6 6.21 1.48 3 7
## Mobile Payment_Ease of Use 8 152 5.70 1.44 6 5.93 1.48 1 7
## Cash_Convenience 9 152 4.71 1.78 5 4.77 2.97 1 7
## Card_Convenience 10 152 5.92 1.07 6 6.05 1.48 2 7
## Mobile Payment_Convenience 11 152 5.91 1.31 6 6.15 1.48 1 7
## Cash_Privacy 12 152 6.32 1.04 7 6.52 0.00 2 7
## Card_Privacy 13 152 4.25 1.63 4 4.31 1.48 1 7
## Mobile Payment_Privacy 14 152 4.14 1.78 4 4.16 1.48 1 7
## Cash_Tracking Expenses 15 152 4.16 1.98 4 4.20 2.97 1 7
## Card_Tracking Expenses 16 152 5.52 1.30 6 5.68 1.48 2 7
## Mobile Payment_Tracking Expenses 17 152 5.61 1.45 6 5.82 1.48 1 7
## range skew kurtosis se
## Card_Safety 5 -0.70 -0.24 0.10
## Mobile Payment_Safety 6 -1.13 0.63 0.12
## Cash_Speed 6 -0.10 -1.14 0.13
## Card_Speed 6 -1.38 3.36 0.09
## Mobile Payment_Speed 6 -1.53 2.21 0.11
## Cash_Ease of Use 5 -0.91 -0.24 0.12
## Card_Ease of Use 4 -1.02 0.57 0.08
## Mobile Payment_Ease of Use 6 -1.18 0.85 0.12
## Cash_Convenience 6 -0.20 -1.33 0.14
## Card_Convenience 5 -1.13 1.59 0.09
## Mobile Payment_Convenience 6 -1.49 2.19 0.11
## Cash_Privacy 5 -1.62 2.34 0.08
## Card_Privacy 6 -0.21 -0.77 0.13
## Mobile Payment_Privacy 6 -0.07 -1.07 0.14
## Cash_Tracking Expenses 6 -0.01 -1.29 0.16
## Card_Tracking Expenses 5 -0.79 0.06 0.11
## Mobile Payment_Tracking Expenses 6 -0.99 0.21 0.12First, we performed some descriptive statistics. Cash is on average perceived as the safest (mean = 5.82), mobile payment methods as the speediest (mean = 5.79), card as the easiest to use (mean = 6.08) and also the most convenient (mean = 5.92), ever so slightly beating mobile payment platforms (mean = 5.91). Cash is seen offering the most privacy (n = 6.32), and mobile payment platforms allow for the easiest tracking of expenses (mean = 5.61).
The table below shows the correlation matrix, which displays correlations between the variables. For Card_Safety and Mobile Payment_Safety, for example, it is 0.593. This means that the linear combination between Card_Safety and Mobile Payment_Safety is positive and semi strong.
mydata_PCA <- mydata[ , c("Cash_Safety", "Card_Safety", "Mobile Payment_Safety", "Cash_Speed", "Card_Speed", "Mobile Payment_Speed", "Cash_Ease of Use", "Card_Ease of Use", "Mobile Payment_Ease of Use", "Cash_Convenience", "Card_Convenience", "Mobile Payment_Convenience", "Cash_Privacy", "Card_Privacy", "Mobile Payment_Privacy", "Cash_Tracking Expenses", "Card_Tracking Expenses", "Mobile Payment_Tracking Expenses")]
library(pastecs)round(stat.desc(mydata_PCA, basic = FALSE), 2)## Cash_Safety Card_Safety Mobile Payment_Safety Cash_Speed
## median 6.00 6.00 6.00 5.00
## mean 5.82 5.34 5.26 4.48
## SE.mean 0.11 0.10 0.12 0.13
## CI.mean.0.95 0.22 0.19 0.24 0.26
## var 1.90 1.46 2.22 2.57
## std.dev 1.38 1.21 1.49 1.60
## coef.var 0.24 0.23 0.28 0.36
## Card_Speed Mobile Payment_Speed Cash_Ease of Use Card_Ease of Use
## median 6.00 6.00 6.00 6.00
## mean 5.78 5.79 5.63 6.08
## SE.mean 0.09 0.11 0.12 0.08
## CI.mean.0.95 0.17 0.23 0.24 0.17
## var 1.14 2.00 2.22 1.07
## std.dev 1.07 1.41 1.49 1.03
## coef.var 0.18 0.24 0.26 0.17
## Mobile Payment_Ease of Use Cash_Convenience Card_Convenience
## median 6.00 5.00 6.00
## mean 5.70 4.71 5.92
## SE.mean 0.12 0.14 0.09
## CI.mean.0.95 0.23 0.28 0.17
## var 2.08 3.16 1.15
## std.dev 1.44 1.78 1.07
## coef.var 0.25 0.38 0.18
## Mobile Payment_Convenience Cash_Privacy Card_Privacy
## median 6.00 7.00 4.00
## mean 5.91 6.32 4.25
## SE.mean 0.11 0.08 0.13
## CI.mean.0.95 0.21 0.17 0.26
## var 1.71 1.08 2.67
## std.dev 1.31 1.04 1.63
## coef.var 0.22 0.16 0.38
## Mobile Payment_Privacy Cash_Tracking Expenses
## median 4.00 4.00
## mean 4.14 4.16
## SE.mean 0.14 0.16
## CI.mean.0.95 0.28 0.32
## var 3.15 3.94
## std.dev 1.78 1.98
## coef.var 0.43 0.48
## Card_Tracking Expenses Mobile Payment_Tracking Expenses
## median 6.00 6.00
## mean 5.52 5.61
## SE.mean 0.11 0.12
## CI.mean.0.95 0.21 0.23
## var 1.68 2.09
## std.dev 1.30 1.45
## coef.var 0.23 0.26R <- cor(mydata_PCA)
round(R, 3)## Cash_Safety Card_Safety Mobile Payment_Safety
## Cash_Safety 1.000 -0.054 -0.083
## Card_Safety -0.054 1.000 0.593
## Mobile Payment_Safety -0.083 0.593 1.000
## Cash_Speed -0.068 0.019 0.040
## Card_Speed 0.063 0.196 0.173
## Mobile Payment_Speed -0.071 0.158 0.098
## Cash_Ease of Use -0.059 0.132 0.025
## Card_Ease of Use -0.078 0.170 0.142
## Mobile Payment_Ease of Use -0.008 -0.033 -0.072
## Cash_Convenience -0.019 0.021 -0.022
## Card_Convenience -0.046 0.159 0.196
## Mobile Payment_Convenience 0.027 0.011 -0.052
## Cash_Privacy -0.032 0.087 0.100
## Card_Privacy -0.088 -0.060 -0.065
## Mobile Payment_Privacy 0.029 -0.080 -0.074
## Cash_Tracking Expenses -0.035 0.030 -0.025
## Card_Tracking Expenses 0.002 0.082 0.150
## Mobile Payment_Tracking Expenses 0.060 -0.001 0.102
## Cash_Speed Card_Speed Mobile Payment_Speed
## Cash_Safety -0.068 0.063 -0.071
## Card_Safety 0.019 0.196 0.158
## Mobile Payment_Safety 0.040 0.173 0.098
## Cash_Speed 1.000 -0.004 -0.227
## Card_Speed -0.004 1.000 0.317
## Mobile Payment_Speed -0.227 0.317 1.000
## Cash_Ease of Use 0.482 0.074 -0.135
## Card_Ease of Use 0.005 0.388 0.275
## Mobile Payment_Ease of Use -0.215 0.198 0.635
## Cash_Convenience 0.602 0.016 -0.193
## Card_Convenience 0.080 0.344 0.409
## Mobile Payment_Convenience -0.193 0.213 0.584
## Cash_Privacy 0.153 0.171 -0.012
## Card_Privacy 0.002 0.088 0.313
## Mobile Payment_Privacy -0.079 0.128 0.368
## Cash_Tracking Expenses 0.476 -0.031 -0.057
## Card_Tracking Expenses -0.099 0.341 0.172
## Mobile Payment_Tracking Expenses -0.176 0.280 0.313
## Cash_Ease of Use Card_Ease of Use
## Cash_Safety -0.059 -0.078
## Card_Safety 0.132 0.170
## Mobile Payment_Safety 0.025 0.142
## Cash_Speed 0.482 0.005
## Card_Speed 0.074 0.388
## Mobile Payment_Speed -0.135 0.275
## Cash_Ease of Use 1.000 0.294
## Card_Ease of Use 0.294 1.000
## Mobile Payment_Ease of Use -0.175 0.407
## Cash_Convenience 0.492 0.052
## Card_Convenience 0.235 0.563
## Mobile Payment_Convenience -0.147 0.343
## Cash_Privacy 0.261 0.223
## Card_Privacy 0.022 0.137
## Mobile Payment_Privacy 0.017 0.153
## Cash_Tracking Expenses 0.280 0.000
## Card_Tracking Expenses -0.048 0.236
## Mobile Payment_Tracking Expenses -0.097 0.211
## Mobile Payment_Ease of Use Cash_Convenience
## Cash_Safety -0.008 -0.019
## Card_Safety -0.033 0.021
## Mobile Payment_Safety -0.072 -0.022
## Cash_Speed -0.215 0.602
## Card_Speed 0.198 0.016
## Mobile Payment_Speed 0.635 -0.193
## Cash_Ease of Use -0.175 0.492
## Card_Ease of Use 0.407 0.052
## Mobile Payment_Ease of Use 1.000 -0.288
## Cash_Convenience -0.288 1.000
## Card_Convenience 0.396 0.082
## Mobile Payment_Convenience 0.781 -0.288
## Cash_Privacy 0.035 0.165
## Card_Privacy 0.316 -0.023
## Mobile Payment_Privacy 0.355 -0.015
## Cash_Tracking Expenses -0.143 0.302
## Card_Tracking Expenses 0.184 0.028
## Mobile Payment_Tracking Expenses 0.372 -0.116
## Card_Convenience Mobile Payment_Convenience
## Cash_Safety -0.046 0.027
## Card_Safety 0.159 0.011
## Mobile Payment_Safety 0.196 -0.052
## Cash_Speed 0.080 -0.193
## Card_Speed 0.344 0.213
## Mobile Payment_Speed 0.409 0.584
## Cash_Ease of Use 0.235 -0.147
## Card_Ease of Use 0.563 0.343
## Mobile Payment_Ease of Use 0.396 0.781
## Cash_Convenience 0.082 -0.288
## Card_Convenience 1.000 0.434
## Mobile Payment_Convenience 0.434 1.000
## Cash_Privacy 0.083 0.080
## Card_Privacy 0.246 0.200
## Mobile Payment_Privacy 0.278 0.253
## Cash_Tracking Expenses 0.009 -0.191
## Card_Tracking Expenses 0.221 0.227
## Mobile Payment_Tracking Expenses 0.220 0.317
## Cash_Privacy Card_Privacy
## Cash_Safety -0.032 -0.088
## Card_Safety 0.087 -0.060
## Mobile Payment_Safety 0.100 -0.065
## Cash_Speed 0.153 0.002
## Card_Speed 0.171 0.088
## Mobile Payment_Speed -0.012 0.313
## Cash_Ease of Use 0.261 0.022
## Card_Ease of Use 0.223 0.137
## Mobile Payment_Ease of Use 0.035 0.316
## Cash_Convenience 0.165 -0.023
## Card_Convenience 0.083 0.246
## Mobile Payment_Convenience 0.080 0.200
## Cash_Privacy 1.000 -0.216
## Card_Privacy -0.216 1.000
## Mobile Payment_Privacy -0.257 0.822
## Cash_Tracking Expenses 0.059 0.174
## Card_Tracking Expenses 0.155 0.048
## Mobile Payment_Tracking Expenses 0.106 0.125
## Mobile Payment_Privacy Cash_Tracking Expenses
## Cash_Safety 0.029 -0.035
## Card_Safety -0.080 0.030
## Mobile Payment_Safety -0.074 -0.025
## Cash_Speed -0.079 0.476
## Card_Speed 0.128 -0.031
## Mobile Payment_Speed 0.368 -0.057
## Cash_Ease of Use 0.017 0.280
## Card_Ease of Use 0.153 0.000
## Mobile Payment_Ease of Use 0.355 -0.143
## Cash_Convenience -0.015 0.302
## Card_Convenience 0.278 0.009
## Mobile Payment_Convenience 0.253 -0.191
## Cash_Privacy -0.257 0.059
## Card_Privacy 0.822 0.174
## Mobile Payment_Privacy 1.000 0.037
## Cash_Tracking Expenses 0.037 1.000
## Card_Tracking Expenses 0.075 -0.344
## Mobile Payment_Tracking Expenses 0.191 -0.368
## Card_Tracking Expenses
## Cash_Safety 0.002
## Card_Safety 0.082
## Mobile Payment_Safety 0.150
## Cash_Speed -0.099
## Card_Speed 0.341
## Mobile Payment_Speed 0.172
## Cash_Ease of Use -0.048
## Card_Ease of Use 0.236
## Mobile Payment_Ease of Use 0.184
## Cash_Convenience 0.028
## Card_Convenience 0.221
## Mobile Payment_Convenience 0.227
## Cash_Privacy 0.155
## Card_Privacy 0.048
## Mobile Payment_Privacy 0.075
## Cash_Tracking Expenses -0.344
## Card_Tracking Expenses 1.000
## Mobile Payment_Tracking Expenses 0.705
## Mobile Payment_Tracking Expenses
## Cash_Safety 0.060
## Card_Safety -0.001
## Mobile Payment_Safety 0.102
## Cash_Speed -0.176
## Card_Speed 0.280
## Mobile Payment_Speed 0.313
## Cash_Ease of Use -0.097
## Card_Ease of Use 0.211
## Mobile Payment_Ease of Use 0.372
## Cash_Convenience -0.116
## Card_Convenience 0.220
## Mobile Payment_Convenience 0.317
## Cash_Privacy 0.106
## Card_Privacy 0.125
## Mobile Payment_Privacy 0.191
## Cash_Tracking Expenses -0.368
## Card_Tracking Expenses 0.705
## Mobile Payment_Tracking Expenses 1.000library(tibble)
library(tidyr)##
## Attaching package: 'tidyr'## The following object is masked from 'package:pastecs':
##
## extractmydata_PCA <- mydata %>%
pivot_longer(everything(), names_to = "name", values_to = "score") %>%
separate(name, into = c("retailer", "dimension"), sep = "_")%>%
pivot_wider(names_from = dimension, values_from = score, values_fn = mean) %>%
column_to_rownames(var = "retailer")
print(mydata_PCA)## Safety Speed Ease of Use Convenience Privacy
## Cash 5.815789 4.480263 5.631579 4.710526 6.322368
## Card 5.335526 5.782895 6.078947 5.921053 4.250000
## Mobile Payment 5.256579 5.789474 5.697368 5.907895 4.138158
## Tracking Expenses
## Cash 4.157895
## Card 5.519737
## Mobile Payment 5.611842To check whether the principal component analysis can be performed, Bartlett’s test of sphericity is done, which tests the assumption that the population correlation matrix is equal to the identity matrix. The hypotheses are the following:
H0: the correlation matrix is the same to the identity matrix, which means that all correlations are equal.
H1: the correlation matrix is not the same to the identity matrix, which means that all correlations are not equal. Based on the low p-value, H0 can be rejected. This means that the principal component analysis can be performed.
library(psych)
cortest.bartlett(R, n = nrow(mydata))## $chisq
## [1] 1102.831
##
## $p.value
## [1] 1.715596e-143
##
## $df
## [1] 153det(R)## [1] 0.0004761888Determinant of the correlation matrix tells us how highly variables correlate among each other. The larger it is, the smaller the correlation between the variables. For the principal component analysis, there is no threshold as to how high it should we.
Let us now check the KMO and MSA statistics. The latter assesses the adequacy of a single variable, and KMO is a general measure of sampling adequacy. Both KMO (in the output written as Overall MSA) and MSA for each item should be higher than 0.50.
Cash_Safety is the only one with the MSA lower than 0.50, which means that it is the least appropriate for the principal component analysis. We have not dropped it, but it should be taken into account that when creating the principal components, the most information will be lost for Cash_Safety.
library(psych)
KMO(R)## Kaiser-Meyer-Olkin factor adequacy
## Call: KMO(r = R)
## Overall MSA = 0.7
## MSA for each item =
## Cash_Safety Card_Safety
## 0.27 0.56
## Mobile Payment_Safety Cash_Speed
## 0.54 0.67
## Card_Speed Mobile Payment_Speed
## 0.81 0.82
## Cash_Ease of Use Card_Ease of Use
## 0.72 0.73
## Mobile Payment_Ease of Use Cash_Convenience
## 0.76 0.71
## Card_Convenience Mobile Payment_Convenience
## 0.83 0.77
## Cash_Privacy Card_Privacy
## 0.73 0.60
## Mobile Payment_Privacy Cash_Tracking Expenses
## 0.62 0.72
## Card_Tracking Expenses Mobile Payment_Tracking Expenses
## 0.64 0.70Removing Cash_Safety because it’s less than 0.5?
library(FactoMineR)## Warning: package 'FactoMineR' was built under R version 4.4.2components <- PCA(mydata_PCA,
scale.unit = TRUE,
graph = FALSE)
library(factoextra)## Warning: package 'factoextra' was built under R version 4.4.2## Loading required package: ggplot2##
## Attaching package: 'ggplot2'## The following objects are masked from 'package:psych':
##
## %+%, alpha## Welcome! Want to learn more? See two factoextra-related books at https://goo.gl/ve3WBaget_eigenvalue(components)## eigenvalue variance.percent cumulative.variance.percent
## Dim.1 5.3678587 89.46431 89.46431
## Dim.2 0.6321413 10.53569 100.00000The output above shows the eigenvalues for both of the principal components. The eigenvalue – the variance – of the first principal component is 5.37, and about 89.46% of total information was transferred to PC1. The eigenvalue of PC2 is 0.63, and 10.53% of total information was transferred to PC2. 100% of all information was transferred to PC1 and PC2.
According to the Kaiser’s rule, the eigenvalues of components should be larger than 1. Even though the eigenvalue of PC2 is lower than 1, we have decided to retain them.
library(factoextra)
fviz_eig(components,
choice = "eigenvalue",
main = "Scree plot",
ylab = "Eigenvalue",
addlabels = TRUE)
The parallel analysis scree plot can also be used to determine the number of principal components. It follows the rule that PC of the actual data should be larger than PC of resampled data. Also according to this tool, one PC would suffice, but we have decided to keep two.
library(psych)
fa.parallel(mydata_PCA,
sim = FALSE,
fa = "pc")## In smc, smcs < 0 were set to .0
## In smc, smcs < 0 were set to .0
## In smc, smcs < 0 were set to .0## Warning in cor.smooth(model): Matrix was not positive definite, smoothing was
## done## Warning in cor.smooth(r): Matrix was not positive definite, smoothing was done## Warning in fac(r = r, nfactors = nfactors, n.obs = n.obs, rotate = rotate, : An
## ultra-Heywood case was detected. Examine the results carefully## In factor.scores, the correlation matrix is singular, the pseudo inverse is used## Warning in cor(sampledata, use = use): the standard deviation is zero## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Warning in cor(sampledata, use = use): the standard deviation is zero
## Parallel analysis suggests that the number of factors = NA and the number of components = 1library(FactoMineR)
components <- PCA(mydata_PCA,
scale.unit = TRUE,
graph = FALSE,
ncp = 2)
components## **Results for the Principal Component Analysis (PCA)**
## The analysis was performed on 3 individuals, described by 6 variables
## *The results are available in the following objects:
##
## name description
## 1 "$eig" "eigenvalues"
## 2 "$var" "results for the variables"
## 3 "$var$coord" "coord. for the variables"
## 4 "$var$cor" "correlations variables - dimensions"
## 5 "$var$cos2" "cos2 for the variables"
## 6 "$var$contrib" "contributions of the variables"
## 7 "$ind" "results for the individuals"
## 8 "$ind$coord" "coord. for the individuals"
## 9 "$ind$cos2" "cos2 for the individuals"
## 10 "$ind$contrib" "contributions of the individuals"
## 11 "$call" "summary statistics"
## 12 "$call$centre" "mean of the variables"
## 13 "$call$ecart.type" "standard error of the variables"
## 14 "$call$row.w" "weights for the individuals"
## 15 "$call$col.w" "weights for the variables"Let us now look at the correlations between individual variables and PC1 (in the output Dim.1) and PC2 (in the output Dim.2).
print(components$var$cor)## Dim.1 Dim.2
## Safety -0.9828532 0.18438962
## Speed 0.9982590 -0.05898223
## Ease of Use 0.6555122 0.75518462
## Convenience 0.9989791 -0.04517381
## Privacy -0.9949910 0.09996487
## Tracking Expenses 0.9938182 -0.11101969The output above shows the loadings (rescaled coefficients), and shows how much information of each of the variables was transferred to both PC’s. For safety, for example, (-0.9828532)2 + (0.18438962)2 = 1.000, which means that 100% of information of the variable was transferred to PC1 and PC2. The table also shows the correlation coefficients – the correlation between ease of use and PC1, for example, is positive and semi strong. The signs being mixed for both principal components indicate that both dimensions will show contrast.
PC1 (dimension 1, which will follow the x-axis) will mean that the means of payment – cash, card, or mobile payment – is low on safety and privacy, and high in speed, ease of use, convenience, and tracking expenses.
PC2 (dimension 2, which will follow the y-axis) will mean that the means of payment – cash, card, or mobile payment – is low on speed, convenience, and tracking expenses, and high on safety, ease of use, and privacy.
print(components$var$contrib)## Dim.1 Dim.2
## Safety 17.996011 5.3784705
## Speed 18.564592 0.5503364
## Ease of Use 8.004983 90.2177791
## Convenience 18.591386 0.3228191
## Privacy 18.443239 1.5808137
## Tracking Expenses 18.399788 1.9497812The output above shows how much information each of the variables contributed to PC’s. The variable Safety, for example, contributed about 18% of information to PC1 and 5.38% of information to PC2. Ease of use, on the other hand, contributed about 8% of information to PC1 and 90.22% of information to PC2.
library(factoextra)
fviz_pca_biplot(components)
It can be observed that Card is positively associated with Ease of Use, Convenience, and Tracking Expenses. Cash, on the other hand, is negatively associated with Ease of Use, Convenience, and Tracking Expenses, but positively associated with Privacy and Safety. Mobile Payment is positioned lower, indicating that it has some similarities with Card. It is closer to Convenience and Tracking Expenses vectors, meaning that users associate it with these benefits, but it is further away from Ease of Use compared to Card, implying that Cards may be seen as easier to use than Mobile Payments.