Practicum Week 4

Dhea Putri Khasanah

Data Science Student at ITSB

R Programming Data Science Python

📍 Institut Teknologi Sains Bandung

Lecturer In Charge

Bakti Siregar, M.Sc., CDS

Mata Kuliah: Programming for Data Science 1

---

1 Introduction

This practicum demonstrates the use of conditional statements and control flow in programming.
The analysis uses a small employee dataset to calculate bonuses, filter salaries, and demonstrate loop control such as for, while, break, and continue.

2 Process

2.1 Dataset

employees <- list(
  list(name = "Bagas", age = 25, salary =  5000, position = "Staff",      performance = "Good"),
  list(name = "Joan",  age = 30, salary =  7000, position = "Supervisor", performance = "Very Good"),
  list(name = "Alya",  age = 27, salary =  6500, position = "Staff",      performance = "Average"),
  list(name = "Dwi",   age = 35, salary = 10000, position = "Manager",    performance = "Good"),
  list(name = "Nabil", age = 40, salary = 12000, position = "Director",   performance = "Very Good")
)

Table 2.1: EMPLOYEE DATA REPORTS

Name	Age	Salary	Position	Performance
Bagas	25	5,000	Staff	Good
Joan	30	7,000	Supervisor	Very Good
Alya	27	6,500	Staff	Average
Dwi	35	10,000	Manager	Good
Nabil	40	12,000	Director	Very Good

Dataset Interpretation.

1. Workforce Overview & Demographics

• Analysis: The dataset represents a cross-section of a corporate hierarchy, ranging from Staff to Director levels. The age distribution spans from 25 to 40 years, with an average age of approximately 31.4 years.

• Insight: This indicates a relatively young and productive workforce, where career progression (from Staff to Management) is visible as age and experience increase.

2. Compensation Hierarchy

• Analysis: There is a clear positive correlation between Position and Salary. Compensation starts at 5,000 for entry-level Staff and peaks at 12,000 for the Director position.

• Insight: The salary structure is well-defined and follows a standard corporate grading system. The significant gap between the Manager (\(10,000\)) and Director (\(12,000\)) levels reflects the high level of responsibility and strategic decision-making required at the top of the organization.

3. Performance Distribution

• Analysis: Performance metrics are varied, with 40% of employees achieving a “Very Good” rating, 40% at “Good”, and 20% at “Average”.

• Insight: The high percentage of “Very Good” and “Good” ratings (80% combined) suggests a high-performing culture. However, the presence of an “Average” rating for a Staff position (Alya) indicates a potential area for targeted training or performance improvement programs.

4. Strategic Relationship (Salary vs. Performance)

• Analysis: It is noteworthy that higher-salaried roles (Joan and Nabil) are also the top performers (“Very Good”).

• Business Insight: This alignment suggests that the company successfully incentivizes top talent. High earners are delivering high value, which justifies the company’s investment in senior personnel and leadership.

2.2 Helper Function

print_header <- function(title) {
  cat("\n", strrep("=", 60), "\n", sep = "")
  cat(format(title, width = 60, justify = "centre"), "\n")
  cat(strrep("=", 60), "\n")
}

---

2.3 1. Bonus Calculation

Bonus Calculation Rules:

• Very Good -> 20% of salary
• Good -> 10% of salary
• Average -> 5% of salary

print_header("EMPLOYEE BONUS TABLE")

## 
## ============================================================
##                     EMPLOYEE BONUS TABLE                     
## ============================================================

cat(sprintf("%-10s %-10s %-15s %-10s\n", "Name", "Salary", "Performance", "Bonus"))

## Name       Salary     Performance     Bonus

cat(strrep("-", 60), "\n")

## ------------------------------------------------------------

for (emp in employees) {
  if (emp$performance == "Very Good") {
    bonus <- emp$salary * 0.20
  } else if (emp$performance == "Good") {
    bonus <- emp$salary * 0.10
  } else {
    bonus <- emp$salary * 0.05
  }
  cat(sprintf("%-10s %-10d %-15s %-10.0f\n",
              emp$name, emp$salary, emp$performance, bonus))
}

## Bagas      5000       Good            500       
## Joan       7000       Very Good       1400      
## Alya       6500       Average         325       
## Dwi        10000      Good            1000      
## Nabil      12000      Very Good       2400

Table 2.2: EMPLOYEE BONUS TABLE

Name	Position	Performance	Salary	Bonus	Total_Pay
Bagas	Staff	Good	5,000	500	5,500
Joan	Supervisor	Very Good	7,000	1,400	8,400
Alya	Staff	Average	6,500	325	6,825
Dwi	Manager	Good	10,000	1,000	11,000
Nabil	Director	Very Good	12,000	2,400	14,400

---

2.4 2. Salary Filter

print_header("EMPLOYEES WITH SALARY > 6000")

## 
## ============================================================
##                 EMPLOYEES WITH SALARY > 6000                 
## ============================================================

cat(sprintf("%-10s %-10s\n", "Name", "Salary"))

## Name       Salary

cat(strrep("-", 25), "\n")

## -------------------------

for (emp in employees) {
  if (emp$salary > 6000) {
    cat(sprintf("%-10s %-10d\n", emp$name, emp$salary))
  }
}

## Joan       7000      
## Alya       6500      
## Dwi        10000     
## Nabil      12000

2.4.1 Employees With Salary > 6000

Table 2.3: 2.5.3.1 - EMPLOYEES WITH SALARY > 6000

	Name	Salary
2	Joan	7000
3	Alya	6500
4	Dwi	10000
5	Nabil	12000

Analysis of Employees with High Salary (> 6,000)

• Logic Applied: Iterative filtering (For Loop equivalent).

• Interpretation: This table identifies key personnel with high compensation levels. From the dataset, four employees (Joan, Alya, Dwi, and Nabil) meet the criteria.

• Business Insight: This list helps management monitor high-tier labor costs. It shows that 80% of the sample workforce earns above the 6,000 threshold, indicating a senior-heavy or high-skill distribution within this group.

---

2.5 3. While Loop — Search Until Manager

print_header("SEARCH UNTIL MANAGER")

## 
## ============================================================
##                     SEARCH UNTIL MANAGER                     
## ============================================================

cat(sprintf("%-10s %-15s\n", "Name", "Position"))

## Name       Position

cat(strrep("-", 30), "\n")

## ------------------------------

i <- 1  # Di R, index dimulai dari 1
while (i <= length(employees)) {
  emp <- employees[[i]]
  cat(sprintf("%-10s %-15s\n", emp$name, emp$position))
  
  if (emp$position == "Manager") {
    cat("\nStop here (Manager Found)\n")
    break
  }
  i <- i + 1
}

## Bagas      Staff          
## Joan       Supervisor     
## Alya       Staff          
## Dwi        Manager        
## 
## Stop here (Manager Found)

2.5.1 Scanning Until Manager Found

Table 2.4: 2.5.3.2 - SCANNING UNTIL MANAGER FOUND

Name	Position	Note
Bagas	Staff
Joan	Supervisor
Alya	Staff
Dwi	Manager	Stop here

Scanning Process (While Loop Logic)

• Logic Applied: Conditional termination using a while loop.

• Interpretation: The system scans the organizational hierarchy sequentially starting from the first entry. The process remains active through Bagas (Staff) and Joan (Supervisor) but terminates immediately upon reaching Dwi (Manager).

• Business Insight: This reflects an automated search efficiency. Instead of processing the entire database, the system stops as soon as the target role is identified, saving computational resources and focusing on the relevant chain of command.

---

2.6 4. Break — Stop at High Salary

print_header("BREAK CONDITION (SALARY > 10000)")

## 
## ============================================================
##               BREAK CONDITION (SALARY > 10000)               
## ============================================================

cat(sprintf("%-10s %-10s\n", "Name", "Salary"))

## Name       Salary

cat(strrep("-", 25), "\n")

## -------------------------

for (emp in employees) {
  if (emp$salary > 10000) {
    cat("\nStopped because salary above 10000\n")
    break
  }
  cat(sprintf("%-10s %-10d\n", emp$name, emp$salary))
}

## Bagas      5000      
## Joan       7000      
## Alya       6500      
## Dwi        10000     
## 
## Stopped because salary above 10000

2.6.1 Break Loop (Salary > 10,000)

Table 2.5: 2.5.3.3 - BREAK LOOP (SALARY > 10,000)

Name	Salary
Bagas	5000
Joan	7000
Alya	6500
Dwi	10000
Status: Stopped because Nabil has a salary above 10,000

Budget Threshold Control (Break Logic)

• Logic Applied: Early exit using the break statement.

• Interpretation: The table displays employees processed before the system encountered a salary exceeding 10,000. The loop processed Bagas, Joan, Alya, and Dwi successfully. However, the sequence was halted before including Nabil, whose salary of 12,000 triggered the safety break.

• Business Insight: This acts as a “financial circuit breaker.” It ensures that reporting or processing stops once an outlier or a specific budget limit is breached, preventing further data inclusion that might violate fiscal constraints.

---

2.7 5. Next — Skip Average Performance

print_header("CONTINUE (SKIP AVERAGE PERFORMANCE)")

## 
## ============================================================
##             CONTINUE (SKIP AVERAGE PERFORMANCE)              
## ============================================================

cat(sprintf("%-10s %-15s\n", "Name", "Performance"))

## Name       Performance

cat(strrep("-", 30), "\n")

## ------------------------------

for (emp in employees) {
  if (emp$performance == "Average") {
    cat(sprintf("%-10s %-15s\n", emp$name, "Skipped"))
    next
  }
  cat(sprintf("%-10s %-15s\n", emp$name, emp$performance))
}

## Bagas      Good           
## Joan       Very Good      
## Alya       Skipped        
## Dwi        Good           
## Nabil      Very Good

2.7.1 Continue (Skip Average Performance)

Table 2.6: 2.5.3.4 - CONTINUE (SKIP AVERAGE PERFORMANCE)

	Name	Performance
1	Bagas	Good
2	Joan	Very Good
4	Dwi	Good
5	Nabil	Very Good
Note: (Alya is skipped because the performance is ‘Average’)

Performance-Based Filtering (Skip/Next Logic)

• Logic Applied: Iteration skipping using the next statement.

• Interpretation: This table presents a curated list of employees, intentionally omitting those with underperforming metrics. Alya is absent from this list because her “Average” performance triggered the skip logic.

• Business Insight: This is a “Focus-Driven” reporting method. By automatically skipping “Average” performers, management can quickly generate a “High-Value Talent List” to prioritize bonus distributions or promotion reviews without manual sorting.

3 Visualization

3.0.1 Bonus Allocation Interpretation

• Budget Concentration: The Director and Manager positions typically consume the largest portion of the bonus budget due to the combination of higher base salaries and performance multipliers.
• Efficiency: Even though there are more Staff members, their total bonus share might be lower if their performance falls into the “Average” category, which only triggers a 5% bonus.
• Visual Clarity: This chart allows stakeholders to see at a glance where the company’s incentive funds are actually going.

4 Conclusion

This comprehensive analysis demonstrates the effective integration of Conditional Statements and Iterative Loops to manage and analyze an employee dataset. By utilizing R’s robust data processing capabilities, we achieved the following strategic objectives:

• Dynamic Compensation Logic: Through the implementation of if-else structures, we successfully automated the calculation of bonuses based on performance tiers: Very Good (20%), Good (10%), Average (5%). This ensures a fair and transparent merit-based reward system.

• Operational Control: The use of For Loops and While Loops provided a systematic way to filter data and control program execution. Specifically:

• Filtered high-salary earners (> 6,000) for budget monitoring.

• Executed precise scanning of organizational structures until specific roles (Manager) were identified.

• Implemented safety breaks to prevent over-budgeting when salaries exceeded 10,000.

• Streamlined reporting by skipping non-priority performance categories (Average) using the next/continue logic.

• Data-Driven Visual Insights: The inclusion of Stacked Bar Charts, Point Plots, and Pie Charts transformed raw numbers into actionable business intelligence. These visuals confirm that bonus allocations are aligned with both individual performance and departmental budgeting.

In summary, this automated approach not only minimizes manual calculation errors but also provides management with a scalable tool for real-time employee performance and salary analysis.