The influence of Entrepreneurial Capital on the performance of subcontracting SMMEs in Gauteng, South Africa,Explanatory Factor analysis (Mmtsatsi Rampa SA)
Lovemore Chipindu [lovemore.datascience@gmail.com], [0778796212]
30 January 2019
library(readr)
rampa <- read_csv("~/SA projects 2019/Rampa/rampa.csv")## Parsed with column specification:
## cols(
## .default = col_double()
## )## See spec(...) for full column specifications.View(rampa)
attach(rampa)
names(rampa)## [1] "ENTCAP1" "ENTCAP2" "ENTCAP3" "ENTCAP4" "ENTCAP5" "ENTCAP6" "ENTCAP7"
## [8] "AC1" "AC2" "AC3" "CC4" "CC5" "CC6" "NC7"
## [15] "NC8" "NC9" "EC1" "EC2" "EC3" "EC4" "EC5"
## [22] "EC6" "EC7" "EC8" "EC9" "EC10" "EC11" "EC12"
## [29] "EC13" "EC14" "EC15" "EC16" "EC17" "EC18" "BP1"
## [36] "BP2" "BP3" "BP4" "BP5" "BP6" "BP7" "BP8"library(factoextra)## Loading required package: ggplot2## Welcome! Related Books: `Practical Guide To Cluster Analysis in R` at https://goo.gl/13EFCZlibrary(FactoMineR)
library(ggplot2)
library(nFactors)## Loading required package: MASS## Loading required package: psych##
## Attaching package: 'psych'## The following objects are masked from 'package:ggplot2':
##
## %+%, alpha## Loading required package: boot##
## Attaching package: 'boot'## The following object is masked from 'package:psych':
##
## logit## Loading required package: lattice##
## Attaching package: 'lattice'## The following object is masked from 'package:boot':
##
## melanoma##
## Attaching package: 'nFactors'## The following object is masked from 'package:lattice':
##
## parallellibrary(psych)overal
##screen plot , determining the number of factors or components in form of dimensions to be considered
res.pca <- prcomp(rampa, scale = TRUE)
fviz_eig(res.pca)
fviz_pca_var(res.pca,
col.var = "contrib", # Color by contributions to the PC
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
fviz_pca_ind(res.pca,
col.ind = "cos2", # Color by the quality of representation
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
eig.val <- get_eigenvalue(res.pca)
eig.val## eigenvalue variance.percent cumulative.variance.percent
## Dim.1 10.5075399 25.0179522 25.01795
## Dim.2 4.7586571 11.3301359 36.34809
## Dim.3 2.3471921 5.5885526 41.93664
## Dim.4 1.9247603 4.5827626 46.51940
## Dim.5 1.8088960 4.3068952 50.82630
## Dim.6 1.5100305 3.5953108 54.42161
## Dim.7 1.1544225 2.7486249 57.17023
## Dim.8 1.1035469 2.6274927 59.79773
## Dim.9 1.0073483 2.3984483 62.19618
## Dim.10 0.9527419 2.2684331 64.46461
## Dim.11 0.8789091 2.0926408 66.55725
## Dim.12 0.8001449 1.9051069 68.46236
## Dim.13 0.7883770 1.8770882 70.33944
## Dim.14 0.7707083 1.8350196 72.17446
## Dim.15 0.7549146 1.7974156 73.97188
## Dim.16 0.7170489 1.7072592 75.67914
## Dim.17 0.6814080 1.6224000 77.30154
## Dim.18 0.6416539 1.5277473 78.82929
## Dim.19 0.6132215 1.4600511 80.28934
## Dim.20 0.6097805 1.4518584 81.74120
## Dim.21 0.5776161 1.3752765 83.11647
## Dim.22 0.5394145 1.2843203 84.40079
## Dim.23 0.4967296 1.1826895 85.58348
## Dim.24 0.4802555 1.1434654 86.72695
## Dim.25 0.4692729 1.1173165 87.84426
## Dim.26 0.4364474 1.0391605 88.88342
## Dim.27 0.4232944 1.0078439 89.89127
## Dim.28 0.4008760 0.9544667 90.84573
## Dim.29 0.3906535 0.9301274 91.77586
## Dim.30 0.3854160 0.9176572 92.69352
## Dim.31 0.3640172 0.8667076 93.56023
## Dim.32 0.3434633 0.8177697 94.37800
## Dim.33 0.3286344 0.7824629 95.16046
## Dim.34 0.3094295 0.7367369 95.89720
## Dim.35 0.2938088 0.6995448 96.59674
## Dim.36 0.2747415 0.6541464 97.25089
## Dim.37 0.2561308 0.6098353 97.86072
## Dim.38 0.2114844 0.5035343 98.36426
## Dim.39 0.2079349 0.4950832 98.85934
## Dim.40 0.1937186 0.4612349 99.32058
## Dim.41 0.1577261 0.3755384 99.69611
## Dim.42 0.1276322 0.3038863 100.00000Non graphical solution to screen test
ev<-eigen(cor(rampa))
ap<-parallel(subject=nrow(rampa),var=ncol(rampa),rep=100,cent=.05)
nS<-nScree(x=ev$values,parallel=ap$eigen$qevpea)
plotnScree(nS)
fac<-factanal(rampa,12,rotation="varimax")
print(fac,digits = 2,cutoff=.3,sort=TRUE)##
## Call:
## factanal(x = rampa, factors = 12, rotation = "varimax")
##
## Uniquenesses:
## ENTCAP1 ENTCAP2 ENTCAP3 ENTCAP4 ENTCAP5 ENTCAP6 ENTCAP7 AC1 AC2
## 0.60 0.34 0.35 0.27 0.64 0.50 0.43 0.17 0.43
## AC3 CC4 CC5 CC6 NC7 NC8 NC9 EC1 EC2
## 0.68 0.23 0.50 0.63 0.80 0.40 0.83 0.70 0.57
## EC3 EC4 EC5 EC6 EC7 EC8 EC9 EC10 EC11
## 0.60 0.42 0.49 0.37 0.60 0.53 0.52 0.16 0.39
## EC12 EC13 EC14 EC15 EC16 EC17 EC18 BP1 BP2
## 0.45 0.50 0.33 0.49 0.38 0.00 0.58 0.45 0.22
## BP3 BP4 BP5 BP6 BP7 BP8
## 0.16 0.13 0.33 0.14 0.38 0.57
##
## Loadings:
## Factor1 Factor2 Factor3 Factor4 Factor5 Factor6 Factor7 Factor8
## EC9 0.51
## EC10 0.65
## EC11 0.72
## EC12 0.64
## EC13 0.58
## EC14 0.75
## EC15 0.61
## EC16 0.63 0.32
## EC17 0.73
## EC18 0.52
## BP1 0.71
## BP2 0.86
## BP3 0.90
## BP4 0.91
## BP5 0.78
## BP6 0.73
## BP7 0.57
## BP8 0.56
## EC2 0.59
## EC3 0.57
## EC4 0.68
## EC5 0.35 0.59
## EC6 0.49 0.57
## EC7 0.53
## ENTCAP3 0.74
## ENTCAP4 0.78
## ENTCAP6 0.61
## ENTCAP7 0.43 0.51
## AC1 0.85
## AC2 0.65
## CC5 0.65
## CC6 0.57
## CC4 0.87
## NC8 0.67
## ENTCAP2
## ENTCAP1
## ENTCAP5 0.41
## AC3 0.47
## NC7
## NC9 0.33
## EC1 0.47
## EC8 0.47 0.45
## Factor9 Factor10 Factor11 Factor12
## EC9
## EC10 0.57
## EC11
## EC12
## EC13
## EC14
## EC15
## EC16
## EC17 0.61
## EC18
## BP1
## BP2
## BP3
## BP4
## BP5
## BP6 0.52
## BP7 0.48
## BP8
## EC2
## EC3
## EC4
## EC5
## EC6
## EC7
## ENTCAP3
## ENTCAP4
## ENTCAP6
## ENTCAP7
## AC1
## AC2
## CC5
## CC6
## CC4
## NC8
## ENTCAP2 0.74
## ENTCAP1 0.42
## ENTCAP5
## AC3
## NC7
## NC9
## EC1
## EC8
##
## Factor1 Factor2 Factor3 Factor4 Factor5 Factor6 Factor7
## SS loadings 5.53 4.85 3.42 2.52 1.49 1.17 1.12
## Proportion Var 0.13 0.12 0.08 0.06 0.04 0.03 0.03
## Cumulative Var 0.13 0.25 0.33 0.39 0.42 0.45 0.48
## Factor8 Factor9 Factor10 Factor11 Factor12
## SS loadings 0.94 0.93 0.69 0.54 0.53
## Proportion Var 0.02 0.02 0.02 0.01 0.01
## Cumulative Var 0.50 0.52 0.54 0.55 0.57
##
## Test of the hypothesis that 12 factors are sufficient.
## The chi square statistic is 558.27 on 423 degrees of freedom.
## The p-value is 1.05e-05Entrepreneurial Capital Variables: Concept Entrepreneurial capital: ENTCAP
ent<-rampa[,1:7]
names(ent)## [1] "ENTCAP1" "ENTCAP2" "ENTCAP3" "ENTCAP4" "ENTCAP5" "ENTCAP6" "ENTCAP7"##screen plot , determining the number of factors or components in form of dimensions to be considered
res.pca <- prcomp(ent, scale = TRUE)
fviz_eig(res.pca)
fviz_pca_var(res.pca,
col.var = "contrib", # Color by contributions to the PC
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
fviz_pca_ind(res.pca,
col.ind = "cos2", # Color by the quality of representation
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
eig.val <- get_eigenvalue(res.pca)
eig.val## eigenvalue variance.percent cumulative.variance.percent
## Dim.1 3.3281227 47.544610 47.54461
## Dim.2 1.0223078 14.604397 62.14901
## Dim.3 0.6942637 9.918052 72.06706
## Dim.4 0.6282599 8.975141 81.04220
## Dim.5 0.5338448 7.626354 88.66855
## Dim.6 0.4796305 6.851865 95.52042
## Dim.7 0.3135707 4.479581 100.00000Non graphical solution to screen test
ev<-eigen(cor(ent))
ap<-parallel(subject=nrow(ent),var=ncol(ent),rep=100,cent=.05)
nS<-nScree(x=ev$values,parallel=ap$eigen$qevpea)
plotnScree(nS)
fac<-factanal(ent,3,rotation="varimax")
print(fac,digits = 2,cutoff=.3,sort=TRUE)##
## Call:
## factanal(x = ent, factors = 3, rotation = "varimax")
##
## Uniquenesses:
## ENTCAP1 ENTCAP2 ENTCAP3 ENTCAP4 ENTCAP5 ENTCAP6 ENTCAP7
## 0.78 0.00 0.49 0.04 0.59 0.52 0.49
##
## Loadings:
## Factor1 Factor2 Factor3
## ENTCAP3 0.57 0.40
## ENTCAP4 0.92
## ENTCAP5 0.58
## ENTCAP6 0.40 0.53
## ENTCAP7 0.39 0.58
## ENTCAP2 0.98
## ENTCAP1 0.32 0.33
##
## Factor1 Factor2 Factor3
## SS loadings 1.58 1.32 1.18
## Proportion Var 0.23 0.19 0.17
## Cumulative Var 0.23 0.41 0.58
##
## Test of the hypothesis that 3 factors are sufficient.
## The chi square statistic is 3.67 on 3 degrees of freedom.
## The p-value is 0.3Entrepreneurial Commitment (Allen & Meyer, 1990)
commit<-rampa[,8:16]
names(commit)## [1] "AC1" "AC2" "AC3" "CC4" "CC5" "CC6" "NC7" "NC8" "NC9"##screen plot , determining the number of factors or components in form of dimensions to be considered
res.pca <- prcomp(commit, scale = TRUE)
fviz_eig(res.pca)
fviz_pca_var(res.pca,
col.var = "contrib", # Color by contributions to the PC
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
fviz_pca_ind(res.pca,
col.ind = "cos2", # Color by the quality of representation
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
eig.val <- get_eigenvalue(res.pca)
eig.val## eigenvalue variance.percent cumulative.variance.percent
## Dim.1 2.3102954 25.669949 25.66995
## Dim.2 1.8095529 20.106144 45.77609
## Dim.3 1.0795397 11.994886 57.77098
## Dim.4 0.9452178 10.502420 68.27340
## Dim.5 0.7687530 8.541700 76.81510
## Dim.6 0.7365458 8.183842 84.99894
## Dim.7 0.5392467 5.991630 90.99057
## Dim.8 0.4900384 5.444871 96.43544
## Dim.9 0.3208103 3.564559 100.00000Non graphical solution to screen test
ev<-eigen(cor(commit))
ap<-parallel(subject=nrow(commit),var=ncol(commit),rep=100,cent=.05)
nS<-nScree(x=ev$values,parallel=ap$eigen$qevpea)
plotnScree(nS)
fac<-factanal(commit,3,rotation="varimax")
print(fac,digits = 2,cutoff=.3,sort=TRUE)##
## Call:
## factanal(x = commit, factors = 3, rotation = "varimax")
##
## Uniquenesses:
## AC1 AC2 AC3 CC4 CC5 CC6 NC7 NC8 NC9
## 0.33 0.33 0.76 0.00 0.55 0.67 0.82 0.72 0.90
##
## Loadings:
## Factor1 Factor2 Factor3
## AC1 0.80
## AC2 0.80
## CC5 0.64
## CC6 0.55
## CC4 1.00
## AC3 0.41
## NC7 0.42
## NC8 0.47
## NC9
##
## Factor1 Factor2 Factor3
## SS loadings 1.39 1.28 1.25
## Proportion Var 0.15 0.14 0.14
## Cumulative Var 0.15 0.30 0.43
##
## Test of the hypothesis that 3 factors are sufficient.
## The chi square statistic is 30.76 on 12 degrees of freedom.
## The p-value is 0.00214Entrepreneurial Competence Variables: Opportunity competence, Relationship competence, Conceptual competence, Organising competence , Strategic competence, Commitment competence (Ahmad, 2007; Lans et al., 2011; Man et al., 2002)
comp<-rampa[,17:34]
names(comp)## [1] "EC1" "EC2" "EC3" "EC4" "EC5" "EC6" "EC7" "EC8" "EC9" "EC10"
## [11] "EC11" "EC12" "EC13" "EC14" "EC15" "EC16" "EC17" "EC18"##screen plot , determining the number of factors or components in form of dimensions to be considered
res.pca <- prcomp(comp, scale = TRUE)
fviz_eig(res.pca)
fviz_pca_var(res.pca,
col.var = "contrib", # Color by contributions to the PC
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
fviz_pca_ind(res.pca,
col.ind = "cos2", # Color by the quality of representation
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
eig.val <- get_eigenvalue(res.pca)
eig.val## eigenvalue variance.percent cumulative.variance.percent
## Dim.1 7.9063760 43.924311 43.92431
## Dim.2 1.6623056 9.235031 53.15934
## Dim.3 0.8773588 4.874215 58.03356
## Dim.4 0.8398995 4.666109 62.69967
## Dim.5 0.7806484 4.336935 67.03660
## Dim.6 0.6880940 3.822744 70.85935
## Dim.7 0.6417564 3.565314 74.42466
## Dim.8 0.5882737 3.268187 77.69285
## Dim.9 0.5551719 3.084288 80.77713
## Dim.10 0.5231415 2.906341 83.68348
## Dim.11 0.4660764 2.589314 86.27279
## Dim.12 0.4493204 2.496225 88.76901
## Dim.13 0.4212618 2.340343 91.10936
## Dim.14 0.4036609 2.242561 93.35192
## Dim.15 0.3386012 1.881118 95.23304
## Dim.16 0.3250996 1.806109 97.03915
## Dim.17 0.2962271 1.645706 98.68485
## Dim.18 0.2367267 1.315148 100.00000Non graphical solution to screen test
ev<-eigen(cor(comp))
ap<-parallel(subject=nrow(comp),var=ncol(comp),rep=100,cent=.05)
nS<-nScree(x=ev$values,parallel=ap$eigen$qevpea)
plotnScree(nS)
fac<-factanal(comp,7,rotation="varimax")
print(fac,digits = 2,cutoff=.3,sort=TRUE)##
## Call:
## factanal(x = comp, factors = 7, rotation = "varimax")
##
## Uniquenesses:
## EC1 EC2 EC3 EC4 EC5 EC6 EC7 EC8 EC9 EC10 EC11 EC12 EC13 EC14 EC15
## 0.00 0.56 0.58 0.45 0.45 0.39 0.63 0.00 0.56 0.36 0.37 0.46 0.41 0.33 0.00
## EC16 EC17 EC18
## 0.41 0.00 0.58
##
## Loadings:
## Factor1 Factor2 Factor3 Factor4 Factor5 Factor6 Factor7
## EC2 0.58
## EC3 0.56
## EC4 0.67
## EC5 0.67
## EC6 0.56 0.32 0.33
## EC7 0.51
## EC9 0.52
## EC10 0.68
## EC11 0.57 0.34
## EC14 0.54 0.48
## EC12 0.32 0.51
## EC13 0.68
## EC17 0.37 0.84
## EC15 0.30 0.87
## EC8 0.33 0.30 0.85
## EC1 0.31 0.94
## EC16 0.34 0.33 0.33 0.38 0.30
## EC18 0.33 0.32 0.33
##
## Factor1 Factor2 Factor3 Factor4 Factor5 Factor6 Factor7
## SS loadings 2.95 2.26 1.71 1.35 1.13 1.05 1.01
## Proportion Var 0.16 0.13 0.09 0.07 0.06 0.06 0.06
## Cumulative Var 0.16 0.29 0.38 0.46 0.52 0.58 0.64
##
## Test of the hypothesis that 7 factors are sufficient.
## The chi square statistic is 59.77 on 48 degrees of freedom.
## The p-value is 0.119Overall Business Performance (Sales growth vs Competitor) , Level of satisfaction with performance: BP
perf<-rampa[,35:42]
names(perf)## [1] "BP1" "BP2" "BP3" "BP4" "BP5" "BP6" "BP7" "BP8"##screen plot , determining the number of factors or components in form of dimensions to be considered
res.pca <- prcomp(perf, scale = TRUE)
fviz_eig(res.pca)
fviz_pca_var(res.pca,
col.var = "contrib", # Color by contributions to the PC
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
fviz_pca_ind(res.pca,
col.ind = "cos2", # Color by the quality of representation
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE # Avoid text overlapping
)
eig.val <- get_eigenvalue(res.pca)
eig.val## eigenvalue variance.percent cumulative.variance.percent
## Dim.1 5.0888911 63.611139 63.61114
## Dim.2 0.8472688 10.590861 74.20200
## Dim.3 0.6251670 7.814587 82.01659
## Dim.4 0.4611110 5.763887 87.78047
## Dim.5 0.3826654 4.783317 92.56379
## Dim.6 0.2559139 3.198924 95.76271
## Dim.7 0.1930125 2.412656 98.17537
## Dim.8 0.1459704 1.824630 100.00000Non graphical solution to screen test
ev<-eigen(cor(perf))
ap<-parallel(subject=nrow(perf),var=ncol(perf),rep=100,cent=.05)
nS<-nScree(x=ev$values,parallel=ap$eigen$qevpea)
plotnScree(nS)
fac<-factanal(perf,4,rotation="varimax")
print(fac,digits = 2,cutoff=.3,sort=TRUE)##
## Call:
## factanal(x = perf, factors = 4, rotation = "varimax")
##
## Uniquenesses:
## BP1 BP2 BP3 BP4 BP5 BP6 BP7 BP8
## 0.00 0.00 0.20 0.11 0.33 0.11 0.45 0.65
##
## Loadings:
## Factor1 Factor2 Factor3 Factor4
## BP6 0.87
## BP7 0.69
## BP3 0.38 0.67 0.35
## BP4 0.33 0.78 0.35
## BP5 0.53 0.53
## BP1 0.89
## BP2 0.50 0.77
## BP8 0.43
##
## Factor1 Factor2 Factor3 Factor4
## SS loadings 2.10 1.86 1.20 0.98
## Proportion Var 0.26 0.23 0.15 0.12
## Cumulative Var 0.26 0.49 0.64 0.77
##
## Test of the hypothesis that 4 factors are sufficient.
## The chi square statistic is 2.09 on 2 degrees of freedom.
## The p-value is 0.352