setwd("~/Documents/Internal Research/Penelitian 2020 Ganjil")
library(psych)
library(GPArotation)

## load data mentah
data_ukur <- read.csv("Pengukuran Kualitas Layanan Elearning- 30 Nov-edited.csv", header = TRUE)

## ambil data numerik utk pengukuran statistik
data_ukur_num <- as.data.frame(data_ukur[,5:24])

## ambil data non-numberik utk pengukuran deskriptif
data_ukur_non_num <- as.data.frame(data_ukur[,2:4])

## kelompokan data berdasarkan faktor
x1 <- data_ukur_num[,1:4]
x2 <- data_ukur_num[,5:8]
x3 <- data_ukur_num[,9:12]
x4 <- data_ukur_num[,13:16]
x5 <- data_ukur_num[,17:20]

## hitung rata-rata utk setiap baris dan buat kolom baru
x1$rata <- rowMeans(x1)
x2$rata <- rowMeans(x2)
x3$rata <- rowMeans(x3)
x4$rata <- rowMeans(x4)
x5$rata <- rowMeans(x5)

## hitung KMO
kmoX1 <- KMO(x1[,1:4])
kmoX2 <- KMO(x2[,1:4])
kmoX3 <- KMO(x3[,1:4])
kmoX4 <- KMO(x4[,1:4])
kmoX5 <- KMO(x5[,1:4])

## hitung reliability
reliabelX1 <- alpha(x1[,1:4])
reliabelX2 <- alpha(x2[,1:4])
reliabelX3 <- alpha(x3[,1:4])
reliabelX4 <- alpha(x4[,1:4])
reliabelX5 <- alpha(x5[,1:4])

## hitung factor analysis
# Step 1 : cari jml factor yang dapat diterima untuk melakukan perhitungan factor analysis dan
#          fungsi yang dipakai fa.parallel() dengan metode 'minres'.
#          Hasil dari fa.parallel() berupa grafik dan informasi yang berupa saran jumlah factor 
#          yang dibutuhkan oleh fungsi perhitungan factor analysis yaitu fungsi fa().
#          
#          Selain itu juga dihasilkan grafik dengan judul Parallel Analysis Scree Plots, dimana
#          arti dari grafik garisnya adalah Garis biru menunjukkan nilai eigen dari data aktual 
#          dan dua garis merah (ditempatkan di atas satu sama lain) menunjukkan data simulasi 
#          dan sampel ulang.
# Step 2 : lakukan hitungan factor analysis dengan berdasarkan saran dari fa.paralel()
fa.parallel(x1[,1:4], fm = 'minres', fa = 'fa')
x1_fac <- factanal(x1[,1:4], 1, rotation = 'varimax')
fa.parallel(x2[,1:4], fm = 'minres', fa = 'fa')
x2_fac <- factanal(x2[,1:4], 1, rotation = 'varimax')
fa.parallel(x3[,1:4], fm = 'minres', fa = 'fa')
x3_fac <- factanal(x3[,1:4], 1, rotation = 'varimax')
fa.parallel(x4[,1:4], fm = 'minres', fa = 'fa')
x4_fac <- factanal(x4[,1:4], 1, rotation = 'varimax')
fa.parallel(x5[,1:4], fm = 'minres', fa = 'fa')
x5_fac <- factanal(x5[,1:4], 1, rotation = 'varimax')
# Step 3 : lakukan penggabungan rata-rata untuk masing-masing faktor dan hitung factor analysis spt step 1 dan 2
xgabungan <- data.frame(x1 = x1$rata, x2 = x2$rata, x3 = x3$rata, x4 = x4$rata, x5 = x5$rata)
fa.parallel(xgabungan, fm = 'minres', fa = 'fa')
xgabungan_fac <- factanal(xgabungan, 1, rotation = 'varimax')

## Data hitungan deskriptif
# Respondent berdasarkan prodi
resp_prodi <- as.data.frame(table(data_ukur_non_num$Program.Studi))
names(resp_prodi) <- c('Prodi','Freq')
# Respondent berdasarkan semester
resp_semester <- as.data.frame(table(data_ukur_non_num$Semester))
names(resp_semester) <- c('Semester','Freq')
# Respondent berdasarkan jenis kelamin
resp_jk <- as.data.frame(table(data_ukur_non_num$Jenis.Kelamin))
names(resp_jk) <- c('JenKel','Freq')

## Pengelompokan data berdasarkan Program Studi
Prodi_SI <- data_ukur_non_num[data_ukur_non_num$Program.Studi == 'Sistem Informasi',]
Prodi_MI <- data_ukur_non_num[data_ukur_non_num$Program.Studi == 'Manajemen Informatika',]
Prodi_TI <- data_ukur_non_num[data_ukur_non_num$Program.Studi == 'Teknik Informatika',]

##Prodi SI
# Respondent berdasarkan prodi dan jenis kelamin
SIprodi_jenkel <- as.data.frame(table(Prodi_SI$Jenis.Kelamin))
names(SIprodi_jenkel) <- c('Jenkel','Freq')
# Respondent berdasarkan prodi dan semester
SIprodi_semester <- as.data.frame(table(Prodi_SI$Semester))
names(SIprodi_semester) <- c('Semester','Freq')

##Prodi MI
# Respondent berdasarkan prodi dan jenis kelamin
MIprodi_jenkel <- as.data.frame(table(Prodi_MI$Jenis.Kelamin))
names(MIprodi_jenkel) <- c('Jenkel','Freq')
# Respondent berdasarkan prodi dan semester
MIprodi_semester <- as.data.frame(table(Prodi_MI$Semester))
names(MIprodi_semester) <- c('Semester','Freq')

##Prodi TI
# Respondent berdasarkan prodi dan jenis kelamin
TIprodi_jenkel <- as.data.frame(table(Prodi_TI$Jenis.Kelamin))
names(TIprodi_jenkel) <- c('Jenkel','Freq')
# Respondent berdasarkan prodi dan semester
TIprodi_semester <- as.data.frame(table(Prodi_TI$Semester))
names(TIprodi_semester) <- c('Semester', 'Freq')

## gambar grafik utk data hitungan deskriptif
library(plotly)
library(dplyr)
# responden berdasarkan program studi
fig1 <- plot_ly(resp_prodi, labels = ~Prodi, values = ~Freq, type = 'pie')
fig1 <- fig1 %>% layout(title = 'Responden Berdasarkan Program Studi',
                        xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
                        yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
fig1
# responden berdasarkan jenis kelamin
fig2 <- plot_ly(resp_jk, labels = ~JenKel, values = ~Freq, type = 'pie')
fig2 <- fig2 %>% layout(title = 'Responden Berdasarkan Jenis Kelamin',
                        xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
                        yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
fig2
# responden berdasarkan semester
fig3 <- plot_ly(resp_semester, x = ~Semester, y = ~Freq, type = 'bar', name = 'Proporsi Respoden berdasarkan Semester')
fig3 <- fig3 %>% layout(title = 'Responden berdasarkan Semester',
                        yaxis = list(title = 'Jumlah'))
fig3
# responden berdasarkan prodi dan jenis kelamin
fig4 <- plot_ly()
fig4 <- fig4 %>% add_pie(data = SIprodi_jenkel, label = ~Jenkel, values = ~Freq, 
                         name = 'SI', text = 'SI', domain = list(x = c(0,0.4),y = c(0.4,1)))
fig4 <- fig4 %>% add_pie(data = MIprodi_jenkel, label = ~Jenkel, values = ~Freq, 
                         name = 'MI', text = 'MI', domain = list(x = c(0.6,1),y = c(0.4,1)))
fig4 <- fig4 %>% add_pie(data = TIprodi_jenkel, label = ~Jenkel, values = ~Freq, 
                         name = 'TI', text = 'TI', domain = list(x = c(0.25,0.75), y = c(0,0.6)))
fig4 <- fig4 %>% layout(title = 'Responden berdasarkan Program Studi dan Jenis Kelamin', showlegend = T,
                        xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
                        yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))
fig4

## responden berdasarkan prodi dan semester
#SI
fig5 <- plot_ly(SIprodi_semester, x = ~Semester, y = ~Freq, type = 'bar')
fig5 <- fig5 %>% layout(title = 'Responden Program Studi SI berdasarkan Semester',
                        xaxis = list(title = 'Semester'),
                        yaxis = list(title = 'Jumlah'))
fig5
#MI
fig6 <- plot_ly(MIprodi_semester, x = ~Semester, y = ~Freq, type = 'bar')
fig6 <- fig6 %>% layout(title = 'Responden Program Studi MI berdasarkan Semester',
                        xaxis = list(title = 'Semester'),
                        yaxis = list(title = 'Jumlah'))
fig6
#TI
fig7 <- plot_ly(TIprodi_semester, x = ~Semester, y = ~Freq, type = 'bar')
fig7 <- fig7 %>% layout(title = 'Responden Program Studi TI berdasarkan Semester',
                        xaxis = list(title = 'Semester'),
                        yaxis = list(title = 'Jumlah'))
fig7