Assignment 6

library(readr)
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, union

library(plotly)

## Loading required package: ggplot2

## 
## Attaching package: 'plotly'

## The following object is masked from 'package:ggplot2':
## 
##     last_plot

## The following object is masked from 'package:stats':
## 
##     filter

## The following object is masked from 'package:graphics':
## 
##     layout

load dataset

hr <- read_csv("https://raw.githubusercontent.com/aiplanethub/Datasets/refs/heads/master/HR_comma_sep.csv")

## Rows: 14999 Columns: 10
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (2): Department, salary
## dbl (8): satisfaction_level, last_evaluation, number_project, average_montly...
## 
## 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.

1. Histogram: Distribution of Employee Satisfaction

plot_ly(
  data = hr,
  x = ~satisfaction_level,
  type = "histogram"
) |>
  layout(
    title = "Employee Satisfaction: Most Employees Report Moderate to High Satisfaction",
    xaxis = list(title = "Satisfaction Level (0–1)"),
    yaxis = list(title = "Count of Employees")
  )

The distribution is right-skewed, with many employees reporting satisfaction above 0.6. Very low satisfaction (<0.3) is relatively rare, affecting a small subset of workers. This suggests overall positive morale, which may correlate with lower turnover in most of the workforce.

2. Box Plot: Last Evaluation Scores

plot_ly(
  data = hr,
  y = ~last_evaluation,
  type = "box",
  boxpoints = "outliers"
) |>
  layout(
    title = "Last Evaluation Scores: Generally High With Some Low Outliers",
    yaxis = list(title = "Evaluation Score (0–1 Scale)")
  )

The median evaluation score is high (~0.85), indicating strong performance among employees. The tight IQR suggests relatively consistent performance across workers. A few low outliers indicate a minority of employees may be underperforming and require support.

3. Comparative Box Plot: Monthly Hours by Department

plot_ly(
  data = hr,
  x = ~Department,
  y = ~average_montly_hours,
  type = "box"
) |>
  layout(
    title = "Monthly Working Hours by Department: Noticeable Variability in Workload",
    xaxis = list(title = "Department"),
    yaxis = list(title = "Average Monthly Hours")
  )

Departments such as technical and sales show slightly higher median working hours. Some departments display wide ranges, suggesting uneven workload distribution within teams. Overworked departments may face increased stress, lower satisfaction, or higher turnover risks.

4. Pie Chart: Attrition by Salary Level

left_data <- hr[hr$left == 1, ]
salary_counts <- table(left_data$salary)

plot_ly(
  labels = names(salary_counts),
  values = as.numeric(salary_counts),
  type = "pie"
) |>
  layout(
    title = "Attrition by Salary Level: Lower Salaries Strongly Associated With Turnover",
    showlegend = TRUE
  )

Employees with low salaries account for roughly 60% of total turnover. Very few high-salary employees leave, suggesting compensation is an important retention mechanism. Medium-salary departures show that factors beyond pay also influence turnover (culture, workload, growth opportunities).

5. Bar Plot: Average Satisfaction by Department

dept_satisfaction <- hr |>
  group_by(Department) |>
  summarise(avg_satisfaction = mean(satisfaction_level, na.rm = TRUE)) |>
  arrange(Department)

plot_ly(
  data = dept_satisfaction,
  x = ~Department,
  y = ~avg_satisfaction,
  type = "bar"
) |>
  layout(
    title = "Average Satisfaction by Department: 
    Small but Meaningful Differences in Morale",
    xaxis = list(title = "Department"),
    yaxis = list(title = "Average Satisfaction Level"),
    bargap = 0.2
  )

Average satisfaction is consistently moderate-to-high (~0.55–0.65) across all departments. Departments such as sales and support show slightly lower averages, which may contribute to turnover or burnout. Overall consistency suggests a generally positive organizational culture, despite higher workloads in some areas.

Key Takeaways from the Visualizations

Employees generally report high satisfaction and strong evaluations, indicating a capable workforce. A few departments carry heavier workload burdens, which may influence morale and retention. Low-salary employees are most likely to leave, confirming the role of compensation in turnover. Departmental satisfaction levels are similar, though small variations may signal areas for intervention.

Organizational Implications

Improving compensation for low-salary workers may significantly reduce attrition. Monitoring workload distribution could help prevent burnout in high-hour departments. Supporting low performers through training or coaching may improve efficiency and morale.