Canada Flag 🇨🇦 Wellness Metrics

Data Analysis Workflow

How raw Canadian health survey data was acquired, audited, cleaned, and transformed into actionable population intelligence and decision-support tools.

View Power BI Report→Insight→
PIPELINE STEPS (7)Hover or click to inspect
  1. 01
    Phase 1Inspecting

    Data Understanding & Ingestion

    Profile raw Canadian health survey tables, examine multi-year survey cycles, and establish suppression rules.

  2. 02
    Phase 2

    Data Cleaning & Standardization

    Standardize StatCan long format, clean metadata columns, and unpivot CIHI chart definitions into tidy rows.

  3. 03
    Phase 3

    Exploratory Data Analysis (EDA)

    Analyze provincial variance, sex disparities (gender paradox), age gradients, and COVID-19 cycle breaks.

  4. 04
    Phase 4

    SQL Analytics & KPI Development

    Execute SQL analytical queries to compute high-level national and provincial KPIs for executive decision support.

  5. 05
    Phase 5

    Machine Learning Synthetic Sandbox

    Experiment with ML predictive algorithms in a strict sandbox environment with zero individual-level prediction.

  6. 06
    Phase 6

    Live StatCan API & Trend Forecasting

    Deploy a real-time FastAPI microservice fetching live Statistics Canada vectors and predicting indicator trajectories.

  7. 07
    Phase 7

    Native Power BI Rebuild & Web Dashboard

    Reconstruct the 4-page Power BI report natively into Next.js/Recharts with interactive choropleths and slide decks.

01
Phase 1: Ingestion & Profiling

Data Understanding & Ingestion

4 Core Public Sources · 10+ Survey Cycles · Strict 'F/x' Suppression Handling
Public Data Ingestion MatrixZero Imputation Protocol
StatCan 13-10-0972182K+
CCHS AnnualA/B/C
StatCan 13-10-04651.7M+
CCHS-MH DisordersA/B
StatCan 13-10-0802466K+
Stress & CopingA/B/E
CIHI Child & Youth4.3K+
ED & InpatientTidy Rows
Suppression Rule: Data cells flagged as F (too unreliable) or x (confidential) are retained as null to prevent synthetic distortion.

Analytical Objective

Audit public Statistics Canada (CCHS) and CIHI tables to document data grain, cycles, missingness, and data quality flags prior to processing.

Key Analytical Steps & Methods

  • ✓Audited StatCan CCHS tables (13-10-0972, 13-10-0465, 13-10-0802) across 2-year survey cycles (2015–2022).
  • ✓Catalogued data quality flags: A (excellent), B (very good), C (good), E (use with caution), and F / x (suppressed for confidentiality).
  • ✓Verified geographic coverage across 10 provinces and 3 territories, identifying territory sampling exclusions.
  • ✓Established strict data governance: zero imputation for suppressed cells (F/x) to avoid artificial variance.