quant-two-group-indlisted

# quant-two-group-ind — Two Independent Groups Comparison ## Objective Test whether two independent groups differ on a continuous or ordinal outcome variable. The primary method is the permutation Welch t-test — it makes no assumptions about equal variances or normality of residuals, making it suitable for most market research data. --- ## Pre-flight **Required inputs:** - `output/data/cleaned_data.csv` - `output/data/column_metadata.json` **Minimum requirements:** - Outcome variable: ordinal or continuous (from metadata) - Grouping variable: exactly 2 unique non-NA values - Minimum n per group: 10 (warn), 20 (preferred) - If grouping variable has >2 unique values: redirect to `/quant-multi-group-ind` --- ## Pipeline ### Step 1 — Load data and metadata ```python # See references/pipeline.py → load_prepared_data() ``` Identify candidate outcome and grouping variables from metadata. --- ### Step 2 — AskUserQuestion: select variables ``` AskUserQuestion: question: "Which variable would you like to compare between groups?" header: "Outcome variable" options: [list all ordinal/continuous columns with their labels] ``` ``` AskUserQuestion: question: "Which variable defines the two groups?" header: "Grouping variable" options: [list all binary/nominal columns with their labels + unique value preview] ``` If the grouping column has >2 unique values, ask the user to either: - Filter to 2 specific groups - Or redirect to `/quant-multi-group-ind` --- ### Ste