How it Works

The expect_column_median_to_be_between test from the dbt-expectations package validates that the median (50th percentile) of a numeric column's values falls within a defined range. The median is more robust to outliers than the mean, making this test preferable when data may contain extreme values that would distort a mean-based check.

Optional arguments include group_by, row_condition, and strictly.

Steps and Conditions

1. Column Selection: Choose the numeric column to evaluate.

2. Define Range: Set min_value and/or max_value for acceptable median values.

3. Optional Config: Apply group_by, row_condition, and strictly as needed.

4. Execution: The median is computed and compared to the range.

5. Outcome: Pass if within range; fail otherwise.

Example Usage: Real Estate Analytics

A real estate analytics platform tracks property valuations and wants to ensure the median property price stays within an expected market range.

models:
  - name: property_listings
    columns:
      - name: listing_price_usd
        tests:
          - dbt_expectations.expect_column_median_to_be_between:
              min_value: 150000
              max_value: 2000000

models:
  - name: property_listings
    columns:
      - name: listing_price_usd
        tests:
          - dbt_expectations.expect_column_median_to_be_between:
              min_value: 150000
              max_value: 2000000

models:
  - name: property_listings
    columns:
      - name: listing_price_usd
        tests:
          - dbt_expectations.expect_column_median_to_be_between:
              min_value: 150000
              max_value: 2000000

A median outside this range could indicate a data ingestion error, a market segment misclassification, or a currency conversion issue.