Which output format is requested?

• A. Plain text
• B. Raw JSON matching the schema
• C. XML
• D. CSV

Answer: (B)

Output formats for automated grading rely on strict structure so a validator can parse data unambiguously. The choice that uses raw JSON matching the schema provides a precise, named-field structure with defined types that the schema requires. This makes automated checks straightforward: the validator can verify each field, its type, and its presence, and extract scores or feedback reliably. Plain text lacks a defined layout, so the validator can’t rely on consistent parsing. XML is structured but unless it exactly mirrors the required schema, it may fail validation and adds parsing variability. CSV is suited for flat tabular data and can’t easily represent nested or optional fields the schema often demands. For reliable, exact conformance to the predefined schema, raw JSON is the best fit.