Analisis Multivariat Pengaruh faktor Aktivitas non Akademik terhadap Nilai Akhir Semester dan Indeks Prestasi Kumulatif (IPK) Menggunakan MANOVA dan MANCOVA
#Load Dataset
library(readr)
df<-read_csv("ResearchInformation3.csv")## Rows: 493 Columns: 16
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (10): Department, Gender, Income, Hometown, Preparation, Gaming, Attenda...
## dbl (6): HSC, SSC, Computer, English, Last, Overall
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.df## # A tibble: 493 × 16
## Department Gender HSC SSC Income Hometown Computer Preparation Gaming
## <chr> <chr> <dbl> <dbl> <chr> <chr> <dbl> <chr> <chr>
## 1 Business Admi… Male 4.17 4.84 Low (… Village 3 More than … 0-1 H…
## 2 Business Admi… Female 4.92 5 Upper… City 3 0-1 Hour 0-1 H…
## 3 Business Admi… Male 5 4.83 Lower… Village 3 0-1 Hour More …
## 4 Business Admi… Male 4 4.5 High … City 5 More than … More …
## 5 Business Admi… Female 2.19 3.17 Lower… Village 3 0-1 Hour 2-3 H…
## 6 Computer Scie… Male 4.75 4.05 Lower… Village 3 0-1 Hour More …
## 7 Computer Scie… Male 4.42 5 High … Village 4 0-1 Hour More …
## 8 Computer Scie… Male 4.5 4.81 Upper… City 3 2-3 Hours More …
## 9 Computer Scie… Male 3.32 4.5 Low (… City 4 0-1 Hour More …
## 10 Computer Scie… Female 3.33 4.95 Lower… City 3 0-1 Hour More …
## # ℹ 483 more rows
## # ℹ 7 more variables: Attendance <chr>, Job <chr>, English <dbl>, Extra <chr>,
## # Semester <chr>, Last <dbl>, Overall <dbl>#Variabel dependen, independen, covariat
df <- df[, c( "HSC","Gender","Income", "Hometown","Job","Gaming","Extra","Last","Overall")]
df## # A tibble: 493 × 9
## HSC Gender Income Hometown Job Gaming Extra Last Overall
## <dbl> <chr> <chr> <chr> <chr> <chr> <chr> <dbl> <dbl>
## 1 4.17 Male Low (Below 15,000) Village No 0-1 H… Yes 3.22 3.35
## 2 4.92 Female Upper middle (30,000-… City No 0-1 H… Yes 3.47 3.47
## 3 5 Male Lower middle (15,000-… Village No More … Yes 4 3.72
## 4 4 Male High (Above 50,000) City No More … Yes 3.8 3.75
## 5 2.19 Female Lower middle (15,000-… Village No 2-3 H… Yes 3.94 3.94
## 6 4.75 Male Lower middle (15,000-… Village No More … No 1 1
## 7 4.42 Male High (Above 50,000) Village No More … No 1.06 1.06
## 8 4.5 Male Upper middle (30,000-… City No More … Yes 2.95 1.25
## 9 3.32 Male Low (Below 15,000) City No More … Yes 1.42 1.44
## 10 3.33 Female Lower middle (15,000-… City No More … No 1.5 1.5
## # ℹ 483 more rowsEXPLORATORY DATA ANALYSIS
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, unionyz <- c("Last", "Overall", "HSC")
for (var in yz) {
meanval <- mean(df[[var]], na.rm = TRUE)
sdval <- sd(df[[var]], na.rm = TRUE)
ggplot(df, aes_string(x = var)) +
geom_histogram(aes(y = ..density..),
fill = "skyblue", bins = 20, color = "black") +
stat_function(fun = dnorm,
args = list(mean = meanval, sd = sdval),
color = "red", size = 1) +
ggtitle(paste("Distribusi", var)) +
theme_minimal() -> p
print(p)
}## Warning: `aes_string()` was deprecated in ggplot2 3.0.0.
## ℹ Please use tidy evaluation idioms with `aes()`.
## ℹ See also `vignette("ggplot2-in-packages")` for more information.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.## Warning: The dot-dot notation (`..density..`) was deprecated in ggplot2 3.4.0.
## ℹ Please use `after_stat(density)` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
library(ggplot2)
X <- c("Gender", "Hometown", "Job", "Gaming", "Extra", "Income")
for (var in X) {
ggplot(df, aes_string(x = var)) +
geom_bar(fill = "orange", color = "black") +
geom_text(stat = "count", aes(label = ..count..), vjust = -0.5) +
ggtitle(paste("Frekuensi:", var)) +
theme_minimal() -> p
print(p)
}
# PRE-PROCESSING
#Cek Miss Value
colSums(is.na(df))## HSC Gender Income Hometown Job Gaming Extra Last
## 0 0 0 0 0 0 0 0
## Overall
## 0#Del Data Duplikat
sum(duplicated(df))## [1] 2df <- df[!duplicated(df), ]
sum(duplicated(df))## [1] 0#Informasi Data
summary(df)## HSC Gender Income Hometown
## Min. :2.170 Length:491 Length:491 Length:491
## 1st Qu.:3.830 Class :character Class :character Class :character
## Median :4.170 Mode :character Mode :character Mode :character
## Mean :4.156
## 3rd Qu.:4.500
## Max. :5.000
## Job Gaming Extra Last
## Length:491 Length:491 Length:491 Min. :1.000
## Class :character Class :character Class :character 1st Qu.:2.810
## Mode :character Mode :character Mode :character Median :3.250
## Mean :3.161
## 3rd Qu.:3.670
## Max. :4.000
## Overall
## Min. :1.000
## 1st Qu.:2.880
## Median :3.270
## Mean :3.186
## 3rd Qu.:3.670
## Max. :4.000str(df)## tibble [491 × 9] (S3: tbl_df/tbl/data.frame)
## $ HSC : num [1:491] 4.17 4.92 5 4 2.19 4.75 4.42 4.5 3.32 3.33 ...
## $ Gender : chr [1:491] "Male" "Female" "Male" "Male" ...
## $ Income : chr [1:491] "Low (Below 15,000)" "Upper middle (30,000-50,000)" "Lower middle (15,000-30,000)" "High (Above 50,000)" ...
## $ Hometown: chr [1:491] "Village" "City" "Village" "City" ...
## $ Job : chr [1:491] "No" "No" "No" "No" ...
## $ Gaming : chr [1:491] "0-1 Hour" "0-1 Hour" "More than 3 Hours" "More than 3 Hours" ...
## $ Extra : chr [1:491] "Yes" "Yes" "Yes" "Yes" ...
## $ Last : num [1:491] 3.22 3.47 4 3.8 3.94 ...
## $ Overall : num [1:491] 3.35 3.47 3.72 3.75 3.94 ...#Variabel Kategori dijadikan Factor
df$Gaming <- factor(df$Gaming,
levels = c("0-1 Hour", "2-3 Hours", "More than 3 Hours"),
ordered = TRUE)
df$Income <- factor(df$Income,
levels = c("Low (Below 15,000)",
"Lower middle (15,000-30,000)",
"Upper middle (30,000-50,000)",
"High (Above 50,000)"),
ordered = TRUE)
df$Job <- factor(df$Job)
df$Extra <- factor(df$Extra)
df$Gender <- factor(df$Gender)
df$Hometown <- factor(df$Hometown)#Transform manly
library(e1071)
manly <- function(x, lambda) {
if (lambda == 0) return(x)
(exp(lambda * x) - 1) / lambda}
optlambdaoverall <- optimize(function(l) abs(skewness(manly(df$Overall, l), na.rm = TRUE)),interval = c(-1, 1))
optlambdalast <- optimize(function(l) abs(skewness(manly(df$Last, l), na.rm = TRUE)),interval = c(-1, 1))
lambda_overall <- optlambdaoverall$minimum
lambda_last <- optlambdalast$minimum
df$Overall_mn <- manly(df$Overall, lambda_overall)
df$Last_mn <- manly(df$Last, lambda_last)
lambda_overall## [1] 0.7977598lambda_last## [1] 0.7027244library(ggplot2)
mean_last <- mean(df$Last_mn, na.rm = TRUE)
sd_last <- sd(df$Last_mn, na.rm = TRUE)
ggplot(df, aes(x = Last_mn)) +
geom_histogram(aes(y = ..density..), fill = "skyblue", color = "black", bins = 20) +
stat_function(fun = dnorm, args = list(mean = mean_last, sd = sd_last),
color = "red", size = 1) +
labs(title = "Distribusi Last Manly Transform") +
theme_minimal()
mean_overall <- mean(df$Overall_mn, na.rm = TRUE)
sd_overall <- sd(df$Overall_mn, na.rm = TRUE)
ggplot(df, aes(x = Overall_mn)) +
geom_histogram(aes(y = ..density..), fill = "skyblue", color = "black", bins = 20) +
stat_function(fun = dnorm, args = list(mean = mean_overall, sd = sd_overall),
color = "red", size = 1) +
labs(title = "Distribusi Overall Manly Transform") +
theme_minimal()
UJI ASUMSI
1. Uji Normalitas y (Shapiro-Wilk)
cat("================================================\n")## ================================================cat("Shapiro Wilk Manly Transform \n")## Shapiro Wilk Manly Transformshapiro.test(df$Last_mn)##
## Shapiro-Wilk normality test
##
## data: df$Last_mn
## W = 0.97183, p-value = 4.114e-08shapiro.test(df$Overall_mn)##
## Shapiro-Wilk normality test
##
## data: df$Overall_mn
## W = 0.97652, p-value = 4.229e-07#2. Uji Dependensi antar Y (Barlett)
#install.packages("psych")
library(psych)##
## Attaching package: 'psych'## The following objects are masked from 'package:ggplot2':
##
## %+%, alphacortest.bartlett(df[, c("Overall_mn", "Last_mn")])## R was not square, finding R from data## $chisq
## [1] 969.4672
##
## $p.value
## [1] 7.782109e-213
##
## $df
## [1] 1#3. Uji Homogenitas Varians-Kovarians (Box’s M Test)
library(heplots)## Loading required package: broom## Warning in rgl.init(initValue, onlyNULL): RGL: unable to open X11 display## Warning: 'rgl.init' failed, will use the null device.
## See '?rgl.useNULL' for ways to avoid this warning.cat("============================Box's M Test Gender==============")## ============================Box's M Test Gender==============boxM(df[, c("Overall_mn", "Last_mn")], df$Gender)##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: df[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 2.0133, df = 3, p-value = 0.5696cat("============================Box's M Test Income==============")## ============================Box's M Test Income==============boxM(df[, c("Overall_mn", "Last_mn")], df$Income)##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: df[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 16.242, df = 9, p-value = 0.062cat("============================Box's M Test Gaming==============")## ============================Box's M Test Gaming==============boxM(df[, c("Overall_mn", "Last_mn")], df$Gaming)##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: df[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 37.106, df = 6, p-value = 1.679e-06cat("============================Box's M Test Job==============")## ============================Box's M Test Job==============boxM(df[, c("Overall_mn", "Last_mn")], df$Job)##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: df[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 55.371, df = 3, p-value = 5.722e-12cat("============================Box's M Test Hometown==============")## ============================Box's M Test Hometown==============boxM(df[, c("Overall_mn", "Last_mn")], df$Hometown)##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: df[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 11.648, df = 3, p-value = 0.008692cat("============================Box's M Test Extra==============")## ============================Box's M Test Extra==============boxM(df[, c("Overall_mn", "Last_mn")], df$Extra)##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: df[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 38.528, df = 3, p-value = 2.185e-08zx <- na.omit(df[, c("Overall_mn", "Last_mn", "Gender", "Hometown", "Job", "Gaming", "Extra", "Income")])
factors <- c("Gender", "Hometown", "Job", "Gaming", "Extra", "Income")
for (i in 1:(length(factors) - 1)) {
for (j in (i + 1):length(factors)) {
f1 <- factors[i]
f2 <- factors[j]
cat("\n============================Box's M Test", f1, "*", f2,"============================\n")
group <- interaction(zx[[f1]], zx[[f2]])
group_counts <- table(group)
if (all(group_counts >= 2)) {result <- boxM(zx[, c("Overall_mn", "Last_mn")], group)
print(result)}
else {cat("ada grup < 2 observasi\n")}}}##
## ============================Box's M Test Gender * Hometown ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 29.99, df = 9, p-value = 0.0004404
##
##
## ============================Box's M Test Gender * Job ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 56.387, df = 9, p-value = 6.62e-09
##
##
## ============================Box's M Test Gender * Gaming ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 104.54, df = 15, p-value = 1.79e-15
##
##
## ============================Box's M Test Gender * Extra ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 54.984, df = 9, p-value = 1.226e-08
##
##
## ============================Box's M Test Gender * Income ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 48.814, df = 21, p-value = 0.0005322
##
##
## ============================Box's M Test Hometown * Job ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 68.766, df = 9, p-value = 2.656e-11
##
##
## ============================Box's M Test Hometown * Gaming ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 62.859, df = 15, p-value = 8.08e-08
##
##
## ============================Box's M Test Hometown * Extra ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 79.363, df = 9, p-value = 2.162e-13
##
##
## ============================Box's M Test Hometown * Income ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 45.624, df = 21, p-value = 0.001433
##
##
## ============================Box's M Test Job * Gaming ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 122.23, df = 15, p-value < 2.2e-16
##
##
## ============================Box's M Test Job * Extra ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 90.823, df = 9, p-value = 1.113e-15
##
##
## ============================Box's M Test Job * Income ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 79.058, df = 21, p-value = 1.16e-08
##
##
## ============================Box's M Test Gaming * Extra ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 128.99, df = 15, p-value < 2.2e-16
##
##
## ============================Box's M Test Gaming * Income ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 87.366, df = 33, p-value = 8.339e-07
##
##
## ============================Box's M Test Extra * Income ============================
##
## Box's M-test for Homogeneity of Covariance Matrices
##
## data: zx[, c("Overall_mn", "Last_mn")]
## Chi-Sq (approx.) = 65.844, df = 21, p-value = 1.604e-06MODELING
#MANOVA one way
varx <- c("Gender", "Hometown", "Job", "Gaming", "Extra", "Income")
for (var in varx) {
cat("\n======================One-Way MANOVA", var,"==========================\n")
subset_df <- df[, c("Last_mn", "Overall_mn", var)]
names(subset_df)[3] <- "X"
model <- manova(cbind(Last_mn, Overall_mn) ~ X, data = subset_df)
print(summary(model, test = "Pillai"))
}##
## ======================One-Way MANOVA Gender ==========================
## Df Pillai approx F num Df den Df Pr(>F)
## X 1 0.022464 5.6073 2 488 0.003912 **
## Residuals 489
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ======================One-Way MANOVA Hometown ==========================
## Df Pillai approx F num Df den Df Pr(>F)
## X 1 0.0028314 0.69281 2 488 0.5007
## Residuals 489
##
## ======================One-Way MANOVA Job ==========================
## Df Pillai approx F num Df den Df Pr(>F)
## X 1 0.0024587 0.60139 2 488 0.5485
## Residuals 489
##
## ======================One-Way MANOVA Gaming ==========================
## Df Pillai approx F num Df den Df Pr(>F)
## X 2 0.20743 28.235 4 976 < 2.2e-16 ***
## Residuals 488
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ======================One-Way MANOVA Extra ==========================
## Df Pillai approx F num Df den Df Pr(>F)
## X 1 0.13566 38.295 2 488 3.56e-16 ***
## Residuals 489
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ======================One-Way MANOVA Income ==========================
## Df Pillai approx F num Df den Df Pr(>F)
## X 3 0.0030842 0.25072 6 974 0.9591
## Residuals 487#MANOVA two way
varx <- c("Gender", "Hometown", "Job", "Gaming", "Extra", "Income")
combinasi <- combn(varx, 2, simplify = FALSE)
for (combo in combinasi) {
var1 <- combo[1]
var2 <- combo[2]
cat("\n=========================Two-Way MANOVA", var1, "and", var2,"=======================\n")
subdf <- df[, c("Last_mn", "Overall_mn", var1, var2)]
names(subdf)[3:4] <- c("X1", "X2")
model <- manova(cbind(Last_mn, Overall_mn) ~ X1 * X2, data = subdf)
print(summary(model, test = "Pillai"))}##
## =========================Two-Way MANOVA Gender and Hometown =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.0224955 5.5922 2 486 0.003971 **
## X2 1 0.0015838 0.3855 2 486 0.680333
## X1:X2 1 0.0017764 0.4324 2 486 0.649173
## Residuals 487
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Gender and Job =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.0224826 5.5889 2 486 0.003983 **
## X2 1 0.0021180 0.5158 2 486 0.597366
## X1:X2 1 0.0032101 0.7826 2 486 0.457804
## Residuals 487
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Gender and Gaming =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.026890 6.6871 2 484 0.001365 **
## X2 2 0.209020 28.3015 4 970 < 2.2e-16 ***
## X1:X2 2 0.006198 0.7538 4 970 0.555516
## Residuals 485
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Gender and Extra =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.025929 6.469 2 486 0.001688 **
## X2 1 0.137491 38.736 2 486 2.458e-16 ***
## X1:X2 1 0.002330 0.568 2 486 0.567252
## Residuals 487
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Gender and Income =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.0227155 5.6017 2 482 0.003936 **
## X2 3 0.0042569 0.3434 6 966 0.913859
## X1:X2 3 0.0187549 1.5241 6 966 0.166931
## Residuals 483
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Hometown and Job =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.00283885 0.69181 2 486 0.5012
## X2 1 0.00307238 0.74889 2 486 0.4734
## X1:X2 1 0.00037088 0.09016 2 486 0.9138
## Residuals 487
##
## =========================Two-Way MANOVA Hometown and Gaming =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.003540 0.8598 2 484 0.4239
## X2 2 0.208193 28.1765 4 970 <2e-16 ***
## X1:X2 2 0.011986 1.4621 4 970 0.2117
## Residuals 485
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Hometown and Extra =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.003272 0.798 2 486 0.4509
## X2 1 0.135976 38.242 2 486 3.765e-16 ***
## X1:X2 1 0.003810 0.929 2 486 0.3955
## Residuals 487
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Hometown and Income =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.0028809 0.69631 2 482 0.4989
## X2 3 0.0039175 0.31597 6 966 0.9288
## X1:X2 3 0.0212718 1.73078 6 966 0.1106
## Residuals 483
##
## =========================Two-Way MANOVA Job and Gaming =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.003024 0.7341 2 484 0.4805
## X2 2 0.211536 28.6824 4 970 <2e-16 ***
## X1:X2 2 0.000935 0.1134 4 970 0.9778
## Residuals 485
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Job and Extra =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.002756 0.671 2 486 0.5114420
## X2 1 0.136330 38.357 2 486 3.408e-16 ***
## X1:X2 1 0.031467 7.895 2 486 0.0004225 ***
## Residuals 487
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Job and Income =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.0024822 0.59970 2 482 0.5494
## X2 3 0.0031514 0.25409 6 966 0.9577
## X1:X2 3 0.0108921 0.88162 6 966 0.5077
## Residuals 483
##
## =========================Two-Way MANOVA Gaming and Extra =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 2 0.239696 33.021 4 970 < 2.2e-16 ***
## X2 1 0.137578 38.605 2 484 2.781e-16 ***
## X1:X2 2 0.062531 7.827 4 970 3.295e-06 ***
## Residuals 485
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Gaming and Income =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 2 0.209695 28.0522 4 958 <2e-16 ***
## X2 3 0.002236 0.1787 6 958 0.9827
## X1:X2 6 0.024428 0.9872 12 958 0.4591
## Residuals 479
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANOVA Extra and Income =======================
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.139463 39.058 2 482 <2e-16 ***
## X2 3 0.002936 0.237 6 966 0.9645
## X1:X2 3 0.032805 2.685 6 966 0.0137 *
## Residuals 483
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#MANCOVA one way
varx <- c("Gender", "Hometown", "Job", "Gaming", "Extra", "Income")
for (var in varx) {
cat("\n====================One-Way MANCOVA", var, "dngan covariat HSC====================\n")
subsetdf <- df[, c("Last_mn", "Overall_mn", var, "HSC")]
names(subsetdf)[3] <- "X"
model <- manova(cbind(Last_mn, Overall_mn) ~ X + HSC, data = subsetdf)
print(summary(model, test = "Pillai"))}##
## ====================One-Way MANCOVA Gender dngan covariat HSC====================
## Df Pillai approx F num Df den Df Pr(>F)
## X 1 0.024400 6.09 2 487 0.002442 **
## HSC 1 0.086271 22.99 2 487 2.878e-10 ***
## Residuals 488
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ====================One-Way MANCOVA Hometown dngan covariat HSC====================
## Df Pillai approx F num Df den Df Pr(>F)
## X 1 0.003101 0.7574 2 487 0.4694
## HSC 1 0.087129 23.2408 2 487 2.289e-10 ***
## Residuals 488
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ====================One-Way MANCOVA Job dngan covariat HSC====================
## Df Pillai approx F num Df den Df Pr(>F)
## X 1 0.002627 0.6413 2 487 0.527
## HSC 1 0.088845 23.7434 2 487 1.448e-10 ***
## Residuals 488
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ====================One-Way MANCOVA Gaming dngan covariat HSC====================
## Df Pillai approx F num Df den Df Pr(>F)
## X 2 0.217051 29.643 4 974 < 2.2e-16 ***
## HSC 1 0.060768 15.722 2 486 2.42e-07 ***
## Residuals 487
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ====================One-Way MANCOVA Extra dngan covariat HSC====================
## Df Pillai approx F num Df den Df Pr(>F)
## X 1 0.145099 41.328 2 487 < 2.2e-16 ***
## HSC 1 0.076241 20.097 2 487 4.107e-09 ***
## Residuals 488
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ====================One-Way MANCOVA Income dngan covariat HSC====================
## Df Pillai approx F num Df den Df Pr(>F)
## X 3 0.003168 0.257 6 972 0.9565
## HSC 1 0.090203 24.043 2 485 1.107e-10 ***
## Residuals 486
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#MANCOVA two way
varx <- c("Gender", "Hometown", "Job", "Gaming", "Extra", "Income")
combinasi <- combn(varx, 2, simplify = FALSE)
for (combo in combinasi) {
var1 <- combo[1]
var2 <- combo[2]
cat("\n=========================Two-Way MANCOVA", var1, "&", var2, "dengan covariat HSC =====\n")
subdf <- df[, c("Last_mn", "Overall_mn", var1, var2, "HSC")]
names(subdf)[3:4] <- c("X1", "X2")
model <- manova(cbind(Last_mn, Overall_mn) ~ X1 * X2 + HSC, data = subdf)
print(summary(model, test = "Pillai"))
}##
## =========================Two-Way MANCOVA Gender & Hometown dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.024408 6.0671 2 485 0.002497 **
## X2 1 0.001730 0.4203 2 485 0.657105
## HSC 1 0.085211 22.5886 2 485 4.171e-10 ***
## X1:X2 1 0.001769 0.4296 2 485 0.650999
## Residuals 486
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Gender & Job dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.024424 6.0710 2 485 0.002488 **
## X2 1 0.002243 0.5450 2 485 0.580179
## HSC 1 0.086425 22.9408 2 485 3.023e-10 ***
## X1:X2 1 0.003330 0.8103 2 485 0.445325
## Residuals 486
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Gender & Gaming dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.028468 7.0766 2 483 0.0009351 ***
## X2 2 0.218420 29.6689 4 968 < 2.2e-16 ***
## HSC 1 0.058565 15.0234 2 483 4.681e-07 ***
## X1:X2 2 0.006498 0.7889 4 968 0.5324686
## Residuals 484
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Gender & Extra dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.027828 6.941 2 485 0.001066 **
## X2 1 0.146790 41.721 2 485 < 2.2e-16 ***
## HSC 1 0.074028 19.387 2 485 7.944e-09 ***
## X1:X2 1 0.002960 0.720 2 485 0.487338
## Residuals 486
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Gender & Income dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.024828 6.1233 2 481 0.002366 **
## X2 3 0.004467 0.3597 6 964 0.904395
## HSC 1 0.088748 23.4227 2 481 1.963e-10 ***
## X1:X2 3 0.022662 1.8414 6 964 0.088117 .
## Residuals 482
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Hometown & Job dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.003110 0.7566 2 485 0.4698
## X2 1 0.003296 0.8019 2 485 0.4491
## HSC 1 0.087096 23.1358 2 485 2.53e-10 ***
## X1:X2 1 0.000982 0.2382 2 485 0.7881
## Residuals 486
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Hometown & Gaming dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.003781 0.9167 2 483 0.40055
## X2 2 0.218451 29.6736 4 968 < 2.2e-16 ***
## HSC 1 0.060726 15.6135 2 483 2.687e-07 ***
## X1:X2 2 0.016340 1.9934 4 968 0.09345 .
## Residuals 484
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Hometown & Extra dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.003536 0.861 2 485 0.4235
## X2 1 0.145277 41.218 2 485 < 2.2e-16 ***
## HSC 1 0.074670 19.569 2 485 6.714e-09 ***
## X1:X2 1 0.004183 1.019 2 485 0.3619
## Residuals 486
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Hometown & Income dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.003166 0.7638 2 481 0.46645
## X2 3 0.004085 0.3288 6 964 0.92199
## HSC 1 0.089780 23.7218 2 481 1.495e-10 ***
## X1:X2 3 0.023422 1.9039 6 964 0.07733 .
## Residuals 482
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Job & Gaming dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.003162 0.7661 2 483 0.4654
## X2 2 0.221284 30.1064 4 968 < 2.2e-16 ***
## HSC 1 0.060629 15.5869 2 483 2.755e-07 ***
## X1:X2 2 0.000900 0.1090 4 968 0.9794
## Residuals 484
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Job & Extra dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.002912 0.708 2 485 0.4930039
## X2 1 0.145608 41.328 2 485 < 2.2e-16 ***
## HSC 1 0.076715 20.149 2 485 3.926e-09 ***
## X1:X2 1 0.029255 7.308 2 485 0.0007465 ***
## Residuals 486
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Job & Income dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.002664 0.6423 2 481 0.5265
## X2 3 0.003239 0.2606 6 964 0.9550
## HSC 1 0.090644 23.9728 2 481 1.19e-10 ***
## X1:X2 3 0.013704 1.1085 6 964 0.3552
## Residuals 482
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Gaming & Extra dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 2 0.248140 34.278 4 968 < 2.2e-16 ***
## X2 1 0.143609 40.497 2 483 < 2.2e-16 ***
## HSC 1 0.056566 14.480 2 483 7.813e-07 ***
## X1:X2 2 0.056875 7.083 4 968 1.274e-05 ***
## Residuals 484
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Gaming & Income dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 2 0.219917 29.5268 4 956 < 2.2e-16 ***
## X2 3 0.002293 0.1829 6 956 0.9816
## HSC 1 0.061213 15.5514 2 477 2.865e-07 ***
## X1:X2 6 0.026647 1.0758 12 956 0.3771
## Residuals 478
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## =========================Two-Way MANCOVA Extra & Income dengan covariat HSC =====
## Df Pillai approx F num Df den Df Pr(>F)
## X1 1 0.148629 41.986 2 481 < 2.2e-16 ***
## X2 3 0.002991 0.241 6 964 0.96302
## HSC 1 0.079927 20.892 2 481 1.992e-09 ***
## X1:X2 3 0.027186 2.214 6 964 0.03966 *
## Residuals 482
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1