Employee Survey Analysis

Analysis of employee satisfaction survey data with ordinal and categorical variables.

Dataset Overview

Synthetic employee survey with 1,000 responses covering:

  • Demographics: Age, department, tenure

  • Satisfaction Measures: Job satisfaction, work-life balance

  • Performance: Ratings and promotion history

  • Attitudes: Engagement scores and turnover intent

Key Features Demonstrated

  • Ordinal Analysis: Satisfaction scales (1-5 ratings)

  • Group Comparisons: Satisfaction by department

  • Causal Analysis: Factors affecting turnover intent

  • Missing Data: Realistic patterns of non-response

Code Example

# Bivariate analysis: Job satisfaction by department
biv_analyzer = BivariateAnalyzer()
result = biv_analyzer.analyze(
    data,
    codebook.variables['job_satisfaction'],
    codebook.variables['department']
)

# Results include chi-square test and effect size
print(f"Chi-square p-value: {result['p_value']:.4f}")
print(f"Effect size (Cramer's V): {result['effect_size']:.3f}")

Generated Insights

Q: How does job satisfaction vary by department?
A: **Job Satisfaction by Department**: Engineering shows highest satisfaction (mean=4.2),
   while Sales shows lowest (mean=3.1). Chi-square test indicates significant differences
   (p<0.001, Cramer's V=0.34).

Files in Example

  • employee_survey.csv: 1,000 employee responses

  • survey_codebook.csv: Variable definitions with ordinal scales

  • analysis.py: Multi-level analysis script

  • results/: Department comparisons and satisfaction plots