Analisis SUSENAS Maret 2019 (KOR)
Yazid Ahmad Hisyam
2024-12-21
1. Pendahuluan
Analisis ini bertujuan untuk melakukan preprocessing, analisis deskriptif, dan analisis multivariat pada data SUSENAS 2019. Dataset terdiri dari informasi pengeluaran rumah tangga, konsumsi, dan variabel terkait lainnya.
2. Preprocessing Data
2.1 Muat Dataset
library(foreign)
library(dplyr)
# Baca dataset (file blok43.dbf)
susenas_raw <- read.dbf("D:/Nomor 6/blok43.dbf")
# Tampilkan struktur awal dataset
str(susenas_raw)## 'data.frame': 340032 obs. of 18 variables:
## $ RENUM : int 285340 285346 285337 285334 285331 285319 285322 285325 285343 285328 ...
## $ R101 : int 11 11 11 11 11 11 11 11 11 11 ...
## $ R102 : int 1 1 1 1 1 1 1 1 1 1 ...
## $ R105 : int 2 2 2 2 2 2 2 2 2 2 ...
## $ R203 : int 1 1 1 1 1 1 1 1 1 1 ...
## $ R301 : int 4 4 3 2 2 1 5 4 4 3 ...
## $ FOOD : num 1795114 2108331 1810200 1561971 1178940 ...
## $ NONFOOD : num 1183000 868198 1074350 790975 778892 ...
## $ EXPEND : num 2978114 2976530 2884550 2352946 1957832 ...
## $ KAPITA : num 744529 744132 961517 1176473 978916 ...
## $ KALORI_KAP: num 2436 2451 2496 3385 3555 ...
## $ PROTE_KAP : num 63.9 78.4 74.5 109.2 105.3 ...
## $ LEMAK_KAP : num 49.3 48.2 45.3 82.6 59 ...
## $ KARBO_KAP : num 397 404 419 506 612 ...
## $ WERT : num 35.2 36.6 35.5 35 31 ...
## $ WEIND : num 140.8 146.6 106.6 70 61.9 ...
## $ WI1 : int 9976 9976 9976 9976 9976 9976 9976 9976 9976 9976 ...
## $ WI2 : int 177146 60810 99379 141157 123223 154278 90478 206467 24522 279725 ...
## - attr(*, "data_types")= chr [1:18] "N" "N" "N" "N" ...2.2 Pembersihan Data
# Pilih variabel yang relevan
selected_vars <- c("RENUM", "R101", "R102", "R105", "R203", "R301",
"FOOD", "NONFOOD", "EXPEND", "KAPITA",
"KALORI_KAP", "PROTE_KAP", "LEMAK_KAP", "KARBO_KAP",
"WERT", "WEIND", "WI1", "WI2")
susenas_cleaned <- susenas_raw %>% select(all_of(selected_vars))
# Hapus data dengan nilai NA
susenas_cleaned <- na.omit(susenas_cleaned)
# Tampilkan ringkasan data setelah pembersihan
summary(susenas_cleaned)## RENUM R101 R102 R105 R203
## Min. : 1 Min. :11.00 Min. : 1.00 Min. :1.000 Min. :1
## 1st Qu.: 85009 1st Qu.:18.00 1st Qu.: 4.00 1st Qu.:1.000 1st Qu.:1
## Median :170017 Median :35.00 Median :10.00 Median :2.000 Median :1
## Mean :170017 Mean :43.05 Mean :21.68 Mean :1.579 Mean :1
## 3rd Qu.:255024 3rd Qu.:64.00 3rd Qu.:23.00 3rd Qu.:2.000 3rd Qu.:1
## Max. :340032 Max. :94.00 Max. :79.00 Max. :2.000 Max. :1
## R301 FOOD NONFOOD EXPEND
## Min. : 1.000 Min. : 114857 Min. : 38208 Min. : 182190
## 1st Qu.: 3.000 1st Qu.: 1295486 1st Qu.: 857667 1st Qu.: 2277443
## Median : 4.000 Median : 1916079 Median : 1403417 Median : 3429452
## Mean : 3.757 Mean : 2226646 Mean : 2142186 Mean : 4368832
## 3rd Qu.: 5.000 3rd Qu.: 2785714 3rd Qu.: 2393183 3rd Qu.: 5212515
## Max. :29.000 Max. :31272857 Max. :193333898 Max. :201254112
## KAPITA KALORI_KAP PROTE_KAP LEMAK_KAP
## Min. : 114515 Min. :1000 Min. : 4.166 Min. : 2.023
## 1st Qu.: 656004 1st Qu.:1737 1st Qu.: 47.371 1st Qu.: 38.230
## Median : 997299 Median :2116 Median : 59.678 Median : 51.136
## Mean : 1308460 Mean :2217 Mean : 64.088 Mean : 55.374
## 3rd Qu.: 1543848 3rd Qu.:2580 3rd Qu.: 75.468 3rd Qu.: 67.453
## Max. :94740858 Max. :4500 Max. :364.666 Max. :293.561
## KARBO_KAP WERT WEIND WI1
## Min. : 25.66 Min. : 1.165 Min. : 1.165 Min. : 1
## 1st Qu.: 254.84 1st Qu.: 67.080 1st Qu.: 212.398 1st Qu.: 7180
## Median : 312.18 Median : 141.845 Median : 474.874 Median :15780
## Mean : 327.74 Mean : 222.376 Mean : 798.704 Mean :15840
## 3rd Qu.: 382.61 3rd Qu.: 296.702 3rd Qu.: 1011.605 3rd Qu.:24378
## Max. :1042.51 Max. :2082.520 Max. :22907.723 Max. :32974
## WI2
## Min. : 1
## 1st Qu.: 71016
## Median :156026
## Mean :156601
## 3rd Qu.:241034
## Max. :3260433. Analisis Deskriptif
3.1 Statistik Deskriptif
# Statistik deskriptif untuk variabel numerik
summary(select(susenas_cleaned, where(is.numeric)))## RENUM R101 R102 R105 R203
## Min. : 1 Min. :11.00 Min. : 1.00 Min. :1.000 Min. :1
## 1st Qu.: 85009 1st Qu.:18.00 1st Qu.: 4.00 1st Qu.:1.000 1st Qu.:1
## Median :170017 Median :35.00 Median :10.00 Median :2.000 Median :1
## Mean :170017 Mean :43.05 Mean :21.68 Mean :1.579 Mean :1
## 3rd Qu.:255024 3rd Qu.:64.00 3rd Qu.:23.00 3rd Qu.:2.000 3rd Qu.:1
## Max. :340032 Max. :94.00 Max. :79.00 Max. :2.000 Max. :1
## R301 FOOD NONFOOD EXPEND
## Min. : 1.000 Min. : 114857 Min. : 38208 Min. : 182190
## 1st Qu.: 3.000 1st Qu.: 1295486 1st Qu.: 857667 1st Qu.: 2277443
## Median : 4.000 Median : 1916079 Median : 1403417 Median : 3429452
## Mean : 3.757 Mean : 2226646 Mean : 2142186 Mean : 4368832
## 3rd Qu.: 5.000 3rd Qu.: 2785714 3rd Qu.: 2393183 3rd Qu.: 5212515
## Max. :29.000 Max. :31272857 Max. :193333898 Max. :201254112
## KAPITA KALORI_KAP PROTE_KAP LEMAK_KAP
## Min. : 114515 Min. :1000 Min. : 4.166 Min. : 2.023
## 1st Qu.: 656004 1st Qu.:1737 1st Qu.: 47.371 1st Qu.: 38.230
## Median : 997299 Median :2116 Median : 59.678 Median : 51.136
## Mean : 1308460 Mean :2217 Mean : 64.088 Mean : 55.374
## 3rd Qu.: 1543848 3rd Qu.:2580 3rd Qu.: 75.468 3rd Qu.: 67.453
## Max. :94740858 Max. :4500 Max. :364.666 Max. :293.561
## KARBO_KAP WERT WEIND WI1
## Min. : 25.66 Min. : 1.165 Min. : 1.165 Min. : 1
## 1st Qu.: 254.84 1st Qu.: 67.080 1st Qu.: 212.398 1st Qu.: 7180
## Median : 312.18 Median : 141.845 Median : 474.874 Median :15780
## Mean : 327.74 Mean : 222.376 Mean : 798.704 Mean :15840
## 3rd Qu.: 382.61 3rd Qu.: 296.702 3rd Qu.: 1011.605 3rd Qu.:24378
## Max. :1042.51 Max. :2082.520 Max. :22907.723 Max. :32974
## WI2
## Min. : 1
## 1st Qu.: 71016
## Median :156026
## Mean :156601
## 3rd Qu.:241034
## Max. :326043# Distribusi variabel kategorikal (contoh: R203)
table(susenas_cleaned$R203)##
## 1
## 3400323.2 Visualisasi Deskriptif
library(ggplot2)
# Visualisasi distribusi pengeluaran makanan
ggplot(susenas_cleaned, aes(x = FOOD)) +
geom_histogram(binwidth = 500000, fill = "blue", alpha = 0.7) +
labs(title = "Distribusi Pengeluaran Makanan", x = "Pengeluaran Makanan", y = "Frekuensi") +
theme_minimal()
4. Analisis Multivariat: K-Means Clustering
4.1 Normalisasi Data
# Pilih variabel numerik untuk clustering
var_cluster <- c("FOOD", "NONFOOD", "EXPEND", "KAPITA",
"KALORI_KAP", "PROTE_KAP", "LEMAK_KAP", "KARBO_KAP")
cluster_data <- susenas_cleaned[, var_cluster]
# Normalisasi data
cluster_data_normalized <- scale(cluster_data)4.2 Penentuan Jumlah Cluster Optimal
set.seed(123)
wss <- sapply(1:10, function(k) {
kmeans(cluster_data_normalized, centers = k, nstart = 25)$tot.withinss
})
# Plot elbow method
plot(1:10, wss, type = "b", pch = 19, frame = FALSE,
xlab = "Number of Clusters (k)",
ylab = "Total Within-Cluster Sum of Squares",
main = "Elbow Method for Optimal k")
4.3 Clustering dan Visualisasi
library(factoextra)
# Lakukan clustering (3 cluster)
kmeans_result <- kmeans(cluster_data_normalized, centers = 3, nstart = 25)
# Tambahkan hasil cluster ke data
susenas_cleaned$Cluster <- as.factor(kmeans_result$cluster)
# Visualisasi clustering
fviz_cluster(kmeans_result, data = cluster_data_normalized,
geom = "point", ellipse.type = "convex",
palette = "jco", ggtheme = theme_minimal()) +
labs(title = "K-Means Clustering")
4.4 Analisis Hasil
# Distribusi data per cluster
table(susenas_cleaned$Cluster)##
## 1 2 3
## 101550 16406 222076# Ringkasan statistik tiap cluster
aggregate(cluster_data, by = list(Cluster = susenas_cleaned$Cluster), FUN = mean)## Cluster FOOD NONFOOD EXPEND KAPITA KALORI_KAP PROTE_KAP LEMAK_KAP
## 1 1 2464382 1974974 4439356 1693360.9 2976.117 88.59248 78.14801
## 2 2 5517235 10824287 16341522 4352082.1 2523.204 82.54083 74.04720
## 3 3 1874841 1577253 3452094 907604.3 1847.794 51.51926 43.58053
## KARBO_KAP
## 1 430.4330
## 2 335.7839
## 3 280.18465. Kesimpulan
K-Means clustering berhasil mengelompokkan rumah tangga menjadi tiga kelompok berdasarkan pengeluaran dan konsumsi mereka. Analisis ini memberikan wawasan awal tentang pola pengeluaran dan konsumsi masyarakat dalam data SUSENAS 2019.