Full-Stack Development Research & Evidence-Based Practice 3 — Questions and Answers
Question 1: Which type of study provides the strongest evidence when deciding to adopt a new frontend framework?
- A single developer's blog post with code examples
- A systematic review or meta-analysis of multiple controlled studies (Correct answer)
- An official framework tutorial
- A Twitter thread by a popular engineer
Correct answer: A systematic review or meta-analysis of multiple controlled studies
Systematic reviews aggregate evidence across multiple studies, reducing the bias and noise of individual reports.
Question 2: In an A/B test for a checkout UI change, the experiment runs for two days and shows 95% confidence. Why might a developer wait longer?
- Statistical confidence always increases with time
- Short windows miss weekly traffic cycles and novelty effects that inflate early results (Correct answer)
- Two days is the standard minimum required by law
- A/B tests are invalid under 95% confidence
Correct answer: Short windows miss weekly traffic cycles and novelty effects that inflate early results
Traffic patterns vary by day of week, and novelty effects can make new UIs appear better initially; longer windows produce more reliable estimates.
Question 3: A developer uses feature flags to roll out a backend change to 5% of traffic. This practice is best described as:
- Canary deployment used as a controlled experiment (Correct answer)
- Blue-green deployment
- Shadow testing
- Chaos engineering
Correct answer: Canary deployment used as a controlled experiment
Rolling a change to a small traffic slice (canary) allows real-world validation with limited blast radius before full rollout.
Question 4: When conducting a usability study, a researcher notices users struggle with a form but does not intervene. This approach is called:
- Participatory design
- Think-aloud protocol
- Formative evaluation
- Summative evaluation with observer effect control (Correct answer)
Correct answer: Summative evaluation with observer effect control
Refraining from intervening during observation is critical to avoid the observer effect distorting how users naturally interact with the interface.
Question 5: Which RFC (Request for Comments) process principle is most relevant when a team debates adopting a new internal API standard?
- Standards should be adopted only after a vendor endorses them
- A formal proposal with stated motivation, trade-offs, and review period builds consensus and documents decisions (Correct answer)
- RFCs are only for internet protocol standards
- The most senior engineer's opinion supersedes the RFC
Correct answer: A formal proposal with stated motivation, trade-offs, and review period builds consensus and documents decisions
RFC-style processes surface trade-offs, collect stakeholder feedback, and create a durable record of why decisions were made.
Question 6: A developer reads that GraphQL always outperforms REST. What research principle should guide their response?
- Accept the claim because GraphQL is newer
- Seek comparative studies under conditions matching their own use case before concluding (Correct answer)
- Reject the claim because REST is the industry standard
- Adopt GraphQL immediately to stay current
Correct answer: Seek comparative studies under conditions matching their own use case before concluding
Absolute performance claims are context-dependent; evidence should be evaluated relative to your specific query patterns, payload sizes, and client needs.
Question 7: What is the primary purpose of maintaining an Architecture Decision Record (ADR)?
- To satisfy compliance auditors
- To document the context, decision, and consequences so future developers understand why choices were made (Correct answer)
- To replace code comments
- To track sprint velocity
Correct answer: To document the context, decision, and consequences so future developers understand why choices were made
ADRs preserve the reasoning behind architectural decisions, preventing future teams from revisiting settled debates without new evidence.
Full-Stack Development Research & Evidence-Based Practice 3
Which type of study provides the strongest evidence when deciding to adopt a new frontend framework?