People First Inc.

: Exploratory Data Analysis (EDA) and

Visualization
Introduction:

People First Inc., an HR Analytics consulting firm, is tasked by a client

experiencing high employee attrition. This report presents an exploratory data
analysis (EDA) of a random sample employee dataset provided by the client.

Data Cleaning and Preparation:

1. Missing Values:
o Hourly Rate: Impute missing values using the median hourly rate of
the entire dataset. This is appropriate as it represents the central
tendency and is less susceptible to outliers compared to the mean.
o Performance Rating: Impute missing values using the mode (most
frequent) performance rating. This maintains the distribution of
ratings within the dataset.
2. Attrition Rate: Calculate the percentage of employees who left the company
by dividing the count of employees with "Yes" in the Attrition column by
the total number of employees and multiplying by 100%.

Analysis:

1. Manager Tenure and Attrition:

Analyze the relationship between the number of years an employee spends with the
same manager and their likelihood of leaving. This requires further investigation
beyond basic EDA and might involve statistical testing.

2. Marital Status and Attrition:

Create a pivot table with "Marital Status" as rows and "Attrition" as columns,
displaying data as percentages. This helps identify potential differences in attrition
rates based on marital status.

3. Gender, Job Role, and Attrition:

Cross-check the client's claim by creating a pivot table or chart with "Gender,"
"Job Role," and "Attrition" to understand if specific job roles have higher attrition
rates for certain genders.

4. Descriptive Statistics:
a) Monthly Income:

 Central tendency: Mean = $X, Median = $Y (replace X and Y with actual

values)
 Dispersion: Standard Deviation = $Z (replace Z with actual value)
 Symmetry: Analyze the skew and kurtosis values to assess the distribution's
symmetry and normality. Interpretation: The mean and median provide
insights into the average and typical income levels, while the standard
deviation measures income variability. Skew and kurtosis indicate if the
distribution deviates from a normal bell-shaped curve.

b) Age:

 Central tendency: Mean = X years, Median = Y years (replace X and Y with

actual values)
 Dispersion: Standard Deviation = Z years (replace Z with actual value)
 Visualization: Create a histogram to visualize the distribution of ages.
Interpretation: Similar to income analysis, understand the average and
typical employee ages, along with the spread of ages within the dataset. The
histogram visually depicts the age distribution.

c) Most Common Salary Hike:

 Use the MODE function to identify the most frequently occurring salary hike
amount.

d) Central Tendency for Monthly Income:

 Median is a more appropriate measure of central tendency for "Monthly

Income" due to its robustness to outliers present in income data. The mean
can be skewed by extreme values (very high or low incomes), whereas the
median represents the "middle" value, offering a better representation of the
typical income level.

e) Workforce Diversity:

 Calculate the range and standard deviation of "Total Working Years" for
both genders using a filter on the "Gender" variable.
 Interpretation: Analyze the range and standard deviation to understand the
spread and variability of total work experience within each gender group. A
larger range and standard deviation might indicate greater diversity in
experience levels.

f) Marital Status and Job Satisfaction:

  Calculate the average "Job Satisfaction" score for each marital status using
the AVERAGE function with a filter on the "Marital Status" variable.
 Interpretation: Compare the average job satisfaction scores across marital
statuses. While this doesn't definitively prove or disprove the claim, it can
indicate potential differences in job satisfaction based on marital status.

Conclusion:

This EDA provides initial insights into the provided employee data. Further
analysis, including statistical testing and visualizations, is crucial for drawing
stronger conclusions and addressing the client's concerns.

Note: Replace placeholders like X, Y, and Z with the actual values obtained from
your analysis.

People First Inc.: Unveiling the Employee Experience

(Humanized Version)
The Challenge:

Imagine this: you're part of the Human Capital Advisory (HCA) team at People
First Inc., a leading HR analytics firm. A client, concerned about high employee
turnover, has entrusted you with a mission – to delve into a sample of their
employee data and uncover valuable insights. This report presents the findings,
aiming to shed light on the factors influencing employee decisions and help the
client build a thriving workplace.

Missing Pieces:

Before diving in, we identified some gaps in the data. We noticed missing
information for "Hourly Rate" and "Performance Rating." To address this, we
filled in the blanks with the most frequently occurring values (for "Performance
Rating") and the average rate across all employees (for "Hourly Rate"). This
ensures our analysis reflects the broader picture without being swayed by
individual outliers.

Turnover Troubles:

One of the most pressing questions is: how many employees are leaving the
company? Analyzing the data, we discovered that a significant portion (XX%) of
employees have chosen to move on (replace XX with the actual percentage). This
highlights the need to understand the underlying reasons behind these departures.
Managerial Impact:

The client believes that employees who spend 15 years or more with the same
manager are less likely to leave. While our initial analysis doesn't provide a
definitive answer, it paves the way for further investigation. We encourage
exploring this connection through statistical testing to gain a clearer understanding.

Marital Status and Job Satisfaction:

We explored the relationship between marital status and employee decisions to

leave. By creating a special table, we discovered the percentages of employees
from different marital backgrounds who chose to depart (replace with specific
percentages). This information can shed light on potential factors influencing
employee retention, prompting further exploration.

Gender, Job Role, and Attrition:

The client also raised concerns about gender disparities in attrition rates within
specific job roles. We examined the data by creating a visualization, revealing
whether specific job roles show higher turnover rates for either gender (replace
with observations based on your analysis). This helps address the client's concerns
and guides further investigation into potential causes.

Employee Insights:

Beyond basic trends, we delved deeper into the data to understand the overall
employee experience. We analyzed:

 Monthly Income: The average and typical monthly income levels (replace
with actual values) provide a benchmark for understanding employee
compensation. Additionally, we explored the spread and variability in
income (replace with observations based on standard deviation), offering
insights into potential income disparities.
 Age: Similar to income, we analyzed the average and typical ages (replace
with actual values) of employees. We also created a visual representation of
the age distribution (replace with a description of the histogram), providing
insights into the age demographics of the workforce.
 Salary Hikes: We identified the most common salary increase amount
(replace with the actual value), offering a glimpse into the typical salary
progression within the company.
 Workforce Diversity: We examined the spread and variability of total work
experience (replace with observations based on the range and standard
deviation) for both genders. This helps assess the diversity in experience
levels within the workforce.
  Job Satisfaction and Marital Status: We compared the average job
satisfaction scores across different marital statuses (replace with
observations based on the analysis). While not conclusive, this provides a
starting point for exploring potential connections between marital status and
job satisfaction.

Moving Forward:

This data exploration serves as a springboard for further analysis. By combining

these initial insights with additional statistical tests and deeper investigation, we
can gain a comprehensive understanding of the factors influencing employee
decisions and guide the client in creating a more engaging and fulfilling work
environment for all.

This humanized version replaces technical jargon with simpler language and adds
context to the analysis, making it easier for a broader audience to understand the
findings and their implications