Competition - HR Analytics in Tableau
  • AI Chat
  • Code
  • Report
    • Beta
    Spinner

    📷 Tableau screenshot

    🌐 Link to Tableau Public: HR Analytics: Atlas Labs Attrition Insights

    🧾 Executive summary

    Atlas Labs is a fictitious software company for which an exploratory data analysis needs to be conducted. The aim of the analysis is to explore factors that impact attrition and help the organization retain more employees. The objective is to identify differences in salary, attrition, job satisfaction level, and work-life balance across demographics and other dimensions, and additionally identify factors with the most significant relationship with attrition.

    Key Findings

    • Salary and Attrition: The analysis revealed a correlation between lower average salary and attrition. Employees with lower salaries are more likely to leave the company.

    • Attrition Duration: On average, there is no attrition after 2.5 years of employment, 1 year with the same manager, 1.5 years since the last promotion, and 1 year in the most recent role. These milestones seem to have a stabilizing effect on attrition.

    • Travel and Marital Status: Employees who frequently travel and are single have a higher attrition rate compared to other groups. The combination of frequent travel and being single appears to contribute to higher attrition.

    • Department-Wise Attrition: The HR and Sales departments exhibit significantly higher attrition rates compared to the Technology department. This indicates a need for further investigation into the underlying causes specific to these departments.

    • Employee Satisfaction: The analysis identified that employees are most satisfied with the work environment. Job satisfaction and satisfaction with work-life balance are relatively similar.

    • Education and Attrition: among employees with Doctorate degrees in Illinois and No Formal Education in California, there are the highest attrition rate by Education.

    Conclusions

    The analysis provides valuable insights into attrition at Atlas Labs, but further investigation is warranted. Exploring additional factors such as performance ratings and career growth opportunities could provide a more comprehensive understanding of attrition drivers. Regarding Attrition and other variables such as age or distance to work, the analysis indicates that there are certain factors that have a stronger impact on attrition rates compared to these variables.

    Overall, the analysis highlights several key insights regarding attrition at Atlas Labs. Employees with lower salaries are more prone to attrition. Frequent travelers, especially those who are single, exhibit higher attrition rates. The HR and Sales departments require attention to address their higher attrition rates. Satisfaction with the work environment is a positive aspect, however, in other satisfaction ratings the count of employees dissatisfied or neutral is similar to count of employees who responded satisfied or very satisfied. There may be a group of neutral ratings employees who are passively satisfied or lack of interest or engagement. It is crucial to identify and address the underlying factors contributing to this neutrality to ensure a more positive and engaged workforce.

    By considering the identified factors and conducting further investigation, the organization can develop targeted strategies to mitigate attrition, improve employee satisfaction and retention, and foster a more engaged and productive workforce.

    R Notebook

    Thanks for reviewing the analysis. I genuinely appreciate your support in the DataCamp competition. If you enjoy my work and believe it deserves recognition, I kindly ask for your vote. Your support means a lot to me and will contribute to my motivation and growth as a data analyst.

    Main questions to address in this challenge:

    • What differences can be found in salary, attrition, job satisfaction and work-life balance across demographics, time and other dimensions?
    • What differences can be found between job satisfaction, work-life balance and other satisfaction influences?
    • Which factor of satisfaction reflects the most insatisfaction?
    • What factors have more relationship with attrition?

    Prepare data for process: read data files and explore tables

    There are three files in the data folder provided by DataCamp of a fictious company called Atlas Labs. The dataset is already described by columns, so we can get an idea of the content of the files before importing data into notebook. First steps should be load packages and read csv files into R environtment.

    # Loading R essential tidyverse packages and data.table for fast reading tables

library(tidyverse) 
library(data.table)
    Hidden output 
    # Import files into R environtment

education_level <- fread("./data/education_level.csv")
employee <- fread("./data/employee.csv")
performance_rating <- fread("./data/performance_rating.csv")
    # Explore tables

glimpse(education_level)
glimpse(employee)
glimpse(performance_rating)

    It appears that there are no missing values, and the data is consistent. The data provided as part of the competition may be pre-processed and checked for integrity before being made available for analysis, and we may not need to conduct an extensive cleaning task and there may not be any missing values or other issues to address. However, it is still important to process the data and verify its suitability for our analysis. Once we have confirmed that the data is clean, we can move to analysis. Tables we have to work with are 'education_level' with 2 columns and 5 rows that do not contain errors or misspellings, 'employee' with 23 variables of integers, characters and dates and 1470 rows and 'performance_rating' with 6709 observations of 11 variables with integers and characters.

    Process data: addressing missing values, duplicates, outliers and fix inconsistencies

    ‌
    ‌
    ‌