# Representation Gap Analysis
import pandas as pd
import numpy as np

def calculate_representation_gaps(df, benchmark_col='market_availability'):
    """
    Calculate representation gaps against external benchmarks
    """
    gaps = {}
    for demographic in df['demographic_group'].unique():
        current_rep = df[df['demographic_group'] == demographic]['current_percentage'].iloc[0]
        benchmark = df[df['demographic_group'] == demographic][benchmark_col].iloc[0]
        gaps[demographic] = {
            'gap_percentage': benchmark - current_rep,
            'gap_ratio': current_rep / benchmark if benchmark > 0 else 0,
            'status': 'above_benchmark' if current_rep >= benchmark else 'below_benchmark'
        }
    return gaps

# Promotion Rate Analysis
def analyze_promotion_equity(promotions_df):
    """
    Analyze promotion rates across demographic groups
    """
    promotion_rates = promotions_df.groupby(['demographic_group']).agg({
        'promoted': 'sum',
        'eligible': 'sum'
    })
    promotion_rates['promotion_rate'] = promotion_rates['promoted'] / promotion_rates['eligible']

    # Calculate statistical significance
    overall_rate = promotion_rates['promoted'].sum() / promotion_rates['eligible'].sum()
    promotion_rates['variance_from_overall'] = promotion_rates['promotion_rate'] - overall_rate

    return promotion_rates

# Inclusion Survey Framework
inclusion_survey:
  psychological_safety:
    - "I feel comfortable expressing my authentic self at work"
    - "I can voice dissenting opinions without fear of negative consequences"
    - "My colleagues actively seek out my perspectives"

  belonging:
    - "I feel valued as a member of my team"
    - "My background and experiences are seen as assets"
    - "I have strong relationships with colleagues across different backgrounds"

  growth_opportunity:
    - "I have equal access to stretch assignments"
    - "I receive constructive feedback that helps me grow"
    - "Leadership actively supports my career development"

  organizational_commitment:
    - "Leadership demonstrates commitment to DEI through actions, not just words"
    - "DEI efforts feel authentic rather than performative"
    - "I see diverse role models in leadership positions"

scoring:
  scale: 1-5 (Strongly Disagree to Strongly Agree)
  benchmark_targets:
    psychological_safety: 4.2
    belonging: 4.0
    growth_opportunity: 3.8
    organizational_commitment: 3.5

# Retention Risk Model
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split

def build_retention_model(employee_df):
    """
    Build predictive model for retention risk by demographic group
    """
    features = ['tenure', 'promotion_count', 'inclusion_score', 'manager_support_score',
               'career_development_rating', 'compensation_percentile']

    X = employee_df[features]
    y = employee_df['retained_12_months']

    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y)

    model = RandomForestClassifier(n_estimators=100, random_state=42)
    model.fit(X_train, y_train)

    # Feature importance analysis
    feature_importance = pd.DataFrame({
        'feature': features,
        'importance': model.feature_importances_
    }).sort_values('importance', ascending=False)

    return model, feature_importance

# Intersectionality Analysis
def intersectionality_analysis(df, primary_dim, secondary_dim, metric):
    """
    Analyze metrics across intersecting identity dimensions
    """
    pivot_table = df.pivot_table(
        values=metric,
        index=primary_dim,
        columns=secondary_dim,
        aggfunc='mean'
    )

    # Calculate variance within and between groups
    overall_mean = df[metric].mean()
    within_group_variance = df.groupby([primary_dim, secondary_dim])[metric].var().mean()
    between_group_variance = pivot_table.var().mean()

    return {
        'pivot_table': pivot_table,
        'within_group_variance': within_group_variance,
        'between_group_variance': between_group_variance,
        'variance_ratio': between_group_variance / within_group_variance
    }

{
  "executive_dei_scorecard": {
    "representation_health": {
      "overall_diversity_index": {
        "current": 0.73,
        "target": 0.80,
        "trend": "improving"
      },
      "leadership_representation": {
        "vp_plus_diversity": 32,
        "target_percentage": 40,
        "yoy_change": "+5%"
      }
    },
    "inclusion_experience": {
      "belonging_score": {
        "overall": 3.8,
        "target": 4.0,
        "lowest_scoring_group": "asian_women",
        "gap_size": 0.4
      },
      "psychological_safety": {
        "company_average": 4.1,
        "benchmark": 4.2,
        "departments_below_benchmark": ["engineering", "sales"]
      }
    },
    "equity_outcomes": {
      "pay_equity": {
        "adjusted_pay_gap": "2.1%",
        "target": "<2%",
        "status": "needs_attention"
      },
      "promotion_equity": {
        "parity_index": 0.94,
        "target": 1.0,
        "most_impacted_group": "black_employees"
      }
    }
  }
}