Which of the following is NOT a characteristic of good test data and test design?

Introduction:
Effective test data ensures quality by exercising functionality, boundaries, and error handling. Collaboration is also key. This question asks you to spot an anti-pattern that contradicts good testing principles.

Given Data / Assumptions:

• We consider unit, integration, and user acceptance test perspectives.
• Coverage, oracle definition, and realistic scenarios are important.
• User involvement is beneficial where business validation matters.

Concept / Approach:
Good test design seeks breadth (normal and edge inputs) and depth (branch and condition coverage). Expected outputs are defined in advance to enable objective pass/fail decisions. Representative user scenarios require stakeholder engagement even early in planning for UAT.

Step-by-Step Solution:
1) Identify best practices: comprehensive cases, coverage, predefined oracles. 2) Compare options to these practices. 3) Excluding users early reduces realism and acceptance; therefore it is not a good characteristic.

Verification / Alternative check:
Standards like ISTQB emphasize stakeholder collaboration and clearly specified expected results before execution.

Why Other Options Are Wrong:
Option A: Breadth is essential to uncover defects.
Option C: Code coverage supports structural adequacy.
Option D: Oracles allow objective verification of results.
Option E: Negative testing ensures robustness.

Common Pitfalls:
Mistaking “preliminary stage” as purely technical. Early user input helps shape realistic scenarios and acceptance criteria.

Final Answer:
Users do not participate at the preliminary stage