Laporan Analisis Data
Morris Alexander Pangaribuan
2026-05-18
- 1. Pendahuluan
- 2. Import Library dan Dataset
- 3. Data Understanding
- 4. Data Cleaning & Preparation
- 5. Data Transformation & Feature Engineering
- 6. Exploratory Data Analysis (EDA)
- 6.1 Distribusi Gender
- 6.2 Distribusi Department
- 6.3 Distribusi Umur
- 6.4 Distribusi Salary
- 6.5 Salary Berdasarkan Job Level
- 6.6 Performance Berdasarkan Department
- 6.7 Korelasi Variabel Numerik
- 7. Analisis & Insight
- 8. Hasil Analisis
- 9. Kesimpulan
- 10. Rekomendasi
- 11. Export Dataset Bersih
1. Pendahuluan
Laporan ini bertujuan untuk melakukan analisis data karyawan
berdasarkan dataset employee_raw_dataset.csv. Analisis
dilakukan menggunakan bahasa pemrograman R dan disusun dalam format R
Markdown.
Tahapan analisis meliputi:
- Data Cleaning & Preparation
- Data Transformation & Feature Engineering
- Exploratory Data Analysis (EDA)
- Analisis dan Insight
- Kesimpulan Hasil Analisis
Dataset berisi informasi terkait:
- Gender
- Department
- Education Level
- Job Level
- Age
- Years Experience
- Monthly Salary
- Performance Score
- Attendance Rate
- Promotion
- dan variabel lainnya.
2. Import Library dan Dataset
# Library utama
library(tidyverse)
library(janitor)
library(skimr)
library(scales)
library(corrplot)
library(knitr)
library(kableExtra)# Import dataset
employee <- read.csv("employee_raw_dataset - employee_raw_dataset.csv")
# Melihat struktur data
str(employee)## 'data.frame': 510 obs. of 13 variables:
## $ Employee_ID : int 1 2 3 4 5 6 7 8 9 10 ...
## $ Gender : chr "female" "Female" "MALE" "Female" ...
## $ Department : chr "Operations" "Hr" "Risk" "IT" ...
## $ Education_Level : chr "Master" "Bachelor" "Bachelor" "Bachelor" ...
## $ Job_Level : chr "Junior" "Officer" "Junior" "Manager" ...
## $ Age : int 15 70 32 27 16 41 29 27 35 38 ...
## $ Years_Experience : int 7 12 18 2 14 9 8 7 1 12 ...
## $ Training_Hours : num 25.6 10.2 33.2 13.8 104.1 ...
## $ Monthly_Salary : int 10702825 15735781 -5000000 21446446 10685352 4095998 6268244 12047713 16111528 9389963 ...
## $ Performance_Score: num 99.5 76.6 86 120 73.2 83.1 79.2 87.1 71.4 58.8 ...
## $ Attendance_Rate : num 98 85.5 91.6 85.7 130 ...
## $ Project_Completed: int 6 2 5 3 2 5 5 3 4 2 ...
## $ Promotion_Last_3Y: int 1 0 0 0 1 1 0 1 0 0 ...## Employee_ID Gender Department Education_Level Job_Level Age
## 1 1 female Operations Master Junior 15
## 2 2 Female Hr Bachelor Officer 70
## 3 3 MALE Risk Bachelor Junior 32
## 4 4 Female IT Bachelor Manager 27
## 5 5 Female Finance Bachelor Junior 16
## 6 6 Female Operations PhD Senior Manager 41
## 7 7 MALE Hr Bachelor Officer 29
## 8 8 MALE IT Bachelor Manager 27
## 9 9 MALE Hr Master Supervisor 35
## 10 10 Female Compliance Bachelor Supervisor 38
## Years_Experience Training_Hours Monthly_Salary Performance_Score
## 1 7 25.6 10702825 99.5
## 2 12 10.2 15735781 76.6
## 3 18 33.2 -5000000 86.0
## 4 2 13.8 21446446 120.0
## 5 14 104.1 10685352 73.2
## 6 9 16.0 4095998 83.1
## 7 8 16.7 6268244 79.2
## 8 7 22.3 12047713 87.1
## 9 1 4.8 16111528 71.4
## 10 12 32.8 9389963 58.8
## Attendance_Rate Project_Completed Promotion_Last_3Y
## 1 98.0 6 1
## 2 85.5 2 0
## 3 91.6 5 0
## 4 85.7 3 0
## 5 130.0 2 1
## 6 90.9 5 1
## 7 92.8 5 0
## 8 95.4 3 1
## 9 106.8 4 0
## 10 94.3 2 03. Data Understanding
3.1 Dimensi Dataset
## [1] 510 133.2 Nama Variabel
## [1] "Employee_ID" "Gender" "Department"
## [4] "Education_Level" "Job_Level" "Age"
## [7] "Years_Experience" "Training_Hours" "Monthly_Salary"
## [10] "Performance_Score" "Attendance_Rate" "Project_Completed"
## [13] "Promotion_Last_3Y"3.3 Statistik Deskriptif
| Name | employee |
| Number of rows | 510 |
| Number of columns | 13 |
| _______________________ | |
| Column type frequency: | |
| character | 4 |
| numeric | 9 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| Gender | 0 | 1 | 1 | 7 | 0 | 8 | 0 |
| Department | 0 | 1 | 0 | 10 | 15 | 14 | 0 |
| Education_Level | 0 | 1 | 1 | 15 | 0 | 8 | 0 |
| Job_Level | 0 | 1 | 6 | 14 | 0 | 5 | 0 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| Employee_ID | 0 | 1.00 | 245.65 | 147.12 | 1.00e+00 | 118.25 | 245.5 | 372.75 | 500.0 | ▇▇▇▇▇ |
| Age | 0 | 1.00 | 33.73 | 8.87 | 1.10e+01 | 28.00 | 33.0 | 40.00 | 70.0 | ▂▇▇▁▁ |
| Years_Experience | 0 | 1.00 | 8.05 | 4.94 | -6.00e+00 | 5.00 | 8.0 | 11.00 | 22.0 | ▁▅▇▅▁ |
| Training_Hours | 35 | 0.93 | 20.91 | 15.83 | 2.00e-01 | 9.70 | 17.1 | 27.90 | 104.1 | ▇▃▁▁▁ |
| Monthly_Salary | 25 | 0.95 | 12492230.33 | 5094031.94 | -5.00e+06 | 9554922.00 | 12061459.0 | 15254192.00 | 44648274.0 | ▁▇▃▁▁ |
| Performance_Score | 0 | 1.00 | 77.74 | 11.97 | 3.96e+01 | 70.90 | 77.3 | 84.50 | 120.0 | ▁▃▇▂▁ |
| Attendance_Rate | 0 | 1.00 | 92.94 | 5.61 | 7.78e+01 | 89.53 | 92.9 | 96.00 | 130.0 | ▂▇▁▁▁ |
| Project_Completed | 0 | 1.00 | 5.07 | 2.30 | 0.00e+00 | 4.00 | 5.0 | 6.00 | 13.0 | ▂▇▅▂▁ |
| Promotion_Last_3Y | 0 | 1.00 | 0.24 | 0.43 | 0.00e+00 | 0.00 | 0.0 | 0.00 | 1.0 | ▇▁▁▁▂ |
4. Data Cleaning & Preparation
Pada tahap ini dilakukan pembersihan data untuk memastikan kualitas data sebelum dianalisis.
4.1 Mengecek Missing Values
## Employee_ID Gender Department Education_Level
## 0 0 0 0
## Job_Level Age Years_Experience Training_Hours
## 0 0 0 35
## Monthly_Salary Performance_Score Attendance_Rate Project_Completed
## 25 0 0 0
## Promotion_Last_3Y
## 04.2 Membersihkan Nama Kolom
## [1] "employee_id" "gender" "department"
## [4] "education_level" "job_level" "age"
## [7] "years_experience" "training_hours" "monthly_salary"
## [10] "performance_score" "attendance_rate" "project_completed"
## [13] "promotion_last_3y"4.3 Standardisasi Data Gender
Dataset memiliki penulisan gender yang tidak konsisten seperti:
- female
- Female
- MALE
- male
Maka dilakukan standardisasi.
## [1] "Female" "Male" "unknown" "-"4.4 Standardisasi Department
## [1] "Operations" "Hr" "Risk" "It" "Finance"
## [6] "Compliance" "Marketing" "Xdept" "" "Unknown"
## [11] "999" "???"4.5 Menangani Nilai Tidak Wajar
Beberapa variabel memiliki nilai yang tidak realistis, seperti:
- Age kurang dari 18 atau terlalu tinggi
- Monthly Salary negatif
- Attendance Rate di atas 100
- Performance Score di atas 100
## employee_id gender department education_level
## Min. : 1.0 Length:510 Length:510 Length:510
## 1st Qu.:118.2 Class :character Class :character Class :character
## Median :245.5 Mode :character Mode :character Mode :character
## Mean :245.6
## 3rd Qu.:372.8
## Max. :500.0
##
## job_level age years_experience training_hours
## Length:510 Min. :11.00 Min. :-6.000 Min. : 0.20
## Class :character 1st Qu.:28.00 1st Qu.: 5.000 1st Qu.: 9.70
## Mode :character Median :33.00 Median : 8.000 Median : 17.10
## Mean :33.73 Mean : 8.049 Mean : 20.91
## 3rd Qu.:40.00 3rd Qu.:11.000 3rd Qu.: 27.90
## Max. :70.00 Max. :22.000 Max. :104.10
## NA's :35
## monthly_salary performance_score attendance_rate project_completed
## Min. :-5000000 Min. : 39.60 Min. : 77.80 Min. : 0.000
## 1st Qu.: 9554922 1st Qu.: 70.90 1st Qu.: 89.53 1st Qu.: 4.000
## Median :12061459 Median : 77.30 Median : 92.90 Median : 5.000
## Mean :12492230 Mean : 77.74 Mean : 92.94 Mean : 5.073
## 3rd Qu.:15254192 3rd Qu.: 84.50 3rd Qu.: 96.00 3rd Qu.: 6.000
## Max. :44648274 Max. :120.00 Max. :130.00 Max. :13.000
## NA's :25
## promotion_last_3y
## Min. :0.0000
## 1st Qu.:0.0000
## Median :0.0000
## Mean :0.2392
## 3rd Qu.:0.0000
## Max. :1.0000
## 4.6 Membersihkan Data Anomali
4.7 Dataset Setelah Cleaning
## [1] 423 13## employee_id gender department education_level
## Min. : 6.0 Length:423 Length:423 Length:423
## 1st Qu.:134.5 Class :character Class :character Class :character
## Median :256.0 Mode :character Mode :character Mode :character
## Mean :253.6
## 3rd Qu.:371.5
## Max. :500.0
##
## job_level age years_experience training_hours
## Length:423 Min. :18.00 Min. :-6.000 Min. : 0.200
## Class :character 1st Qu.:29.00 1st Qu.: 5.000 1st Qu.: 9.425
## Mode :character Median :33.00 Median : 8.000 Median :16.850
## Mean :34.07 Mean : 8.026 Mean :20.235
## 3rd Qu.:40.00 3rd Qu.:11.000 3rd Qu.:27.325
## Max. :58.00 Max. :22.000 Max. :98.200
## NA's :33
## monthly_salary performance_score attendance_rate project_completed
## Min. : 1403602 Min. :39.60 Min. : 77.80 Min. : 0.000
## 1st Qu.: 9562876 1st Qu.:70.55 1st Qu.: 89.50 1st Qu.: 4.000
## Median :12057961 Median :77.20 Median : 92.70 Median : 5.000
## Mean :12593885 Mean :76.82 Mean : 92.29 Mean : 5.099
## 3rd Qu.:15215972 3rd Qu.:83.80 3rd Qu.: 95.70 3rd Qu.: 6.000
## Max. :44648274 Max. :99.20 Max. :100.00 Max. :12.000
##
## promotion_last_3y
## Min. :0.0000
## 1st Qu.:0.0000
## Median :0.0000
## Mean :0.2222
## 3rd Qu.:0.0000
## Max. :1.0000
## 5. Data Transformation & Feature Engineering
Tahap ini bertujuan untuk membuat fitur baru yang dapat membantu analisis.
5.1 Membuat Kategori Umur
## [1] "Senior" "Adult" "Adult" "Adult" "Adult" "Adult"5.2 Membuat Kategori Salary
5.3 Membuat Status Performance
5.4 Konversi Variabel Menjadi Factor
## [1] "employee_id" "gender" "department"
## [4] "education_level" "job_level" "age"
## [7] "years_experience" "training_hours" "monthly_salary"
## [10] "performance_score" "attendance_rate" "project_completed"
## [13] "promotion_last_3y" "age_group" "salary_category"
## [16] "performance_status"6. Exploratory Data Analysis (EDA)
6.1 Distribusi Gender
6.2 Distribusi Department
6.3 Distribusi Umur
6.4 Distribusi Salary
6.5 Salary Berdasarkan Job Level
6.6 Performance Berdasarkan Department
6.7 Korelasi Variabel Numerik
7. Analisis & Insight
7.1 Analisis Gender
## gender n
## 1 - 2
## 2 Female 255
## 3 Male 164
## 4 unknown 2Insight:
- Distribusi gender dapat menunjukkan keseimbangan tenaga kerja.
- Jika salah satu gender mendominasi, perusahaan dapat mempertimbangkan strategi diversity hiring.
7.2 Analisis Salary
## # A tibble: 5 × 2
## job_level avg_salary
## <fct> <dbl>
## 1 Manager 14284178.
## 2 Supervisor 12901059.
## 3 Junior 12711203.
## 4 Officer 11709883.
## 5 Senior Manager 10865866.Insight:
- Job level yang lebih tinggi cenderung memiliki salary lebih besar.
- Salary dapat digunakan sebagai indikator senioritas dan tanggung jawab.
7.3 Analisis Performance
## # A tibble: 12 × 2
## department avg_performance
## <fct> <dbl>
## 1 "999" 83.1
## 2 "It" 79.2
## 3 "Risk" 78.3
## 4 "???" 78.2
## 5 "Finance" 78.1
## 6 "Compliance" 75.9
## 7 "Marketing" 75.9
## 8 "Unknown" 75.8
## 9 "Hr" 75.4
## 10 "Operations" 74.8
## 11 "" 74.0
## 12 "Xdept" 70.5Insight:
- Department dengan performa tinggi dapat menjadi benchmark bagi department lain.
- Department dengan performa rendah mungkin membutuhkan evaluasi sistem kerja atau pelatihan tambahan.
7.4 Analisis Promotion
Error sebelumnya terjadi karena kolom promotion_last_3_y
tidak ditemukan pada dataset. Oleh karena itu, analisis dibuat lebih
fleksibel dengan mengecek nama kolom terlebih dahulu.
## [1] "employee_id" "gender" "department"
## [4] "education_level" "job_level" "age"
## [7] "years_experience" "training_hours" "monthly_salary"
## [10] "performance_score" "attendance_rate" "project_completed"
## [13] "promotion_last_3y" "age_group" "salary_category"
## [16] "performance_status"## Kolom promotion tidak ditemukan pada dataset.Insight:
- Analisis promosi digunakan untuk melihat hubungan antara promosi, performa, dan salary.
- Jika kolom promosi tersedia, maka dapat diketahui apakah karyawan yang dipromosikan memiliki performa lebih tinggi.
- Kode dibuat fleksibel agar tidak error meskipun nama kolom berbeda.
7.5 Analisis Training Hours
Insight:
- Grafik digunakan untuk melihat apakah pelatihan berdampak pada performa.
- Jika tren meningkat, maka program pelatihan dapat dianggap efektif.
8. Hasil Analisis
Berdasarkan proses analisis data yang telah dilakukan, diperoleh beberapa hasil utama:
- Dataset memiliki beberapa data tidak konsisten dan nilai anomali.
- Proses data cleaning berhasil meningkatkan kualitas data.
- Job level memiliki hubungan kuat dengan salary.
- Department tertentu memiliki performa lebih tinggi dibanding department lain.
- Training hours menunjukkan hubungan positif terhadap performance score.
- Karyawan dengan performa tinggi cenderung lebih sering mendapatkan promosi.
9. Kesimpulan
Analisis data karyawan menggunakan R Programming berhasil dilakukan melalui beberapa tahapan penting mulai dari data cleaning, transformation, exploratory analysis, hingga insight generation.
Hasil analisis menunjukkan bahwa:
- Kualitas data sangat penting sebelum analisis dilakukan.
- Faktor seperti training, experience, dan job level memengaruhi performa serta salary.
- Visualisasi data membantu memahami pola dan hubungan antar variabel.
Laporan ini dapat digunakan sebagai dasar pengambilan keputusan HR dan pengembangan strategi manajemen karyawan.
10. Rekomendasi
Beberapa rekomendasi berdasarkan hasil analisis:
- Meningkatkan program pelatihan untuk meningkatkan performa.
- Melakukan monitoring terhadap department dengan performa rendah.
- Menstandarisasi input data agar tidak muncul nilai anomali.
- Mengembangkan sistem promosi berbasis performa.
11. Export Dataset Bersih
Dataset hasil cleaning akan tersimpan dengan nama:
employee_cleaned.csv