Full-Stack Development Research & Evidence-Based Practice 2

Question 1

When evaluating a third-party library for production use, which metric is MOST important as a primary signal of long-term viability?

Accepted Answer

Frequency of recent commits and maintained changelog

Answer

Active maintenance—evidenced by recent commits and a maintained changelog—best predicts a library's long-term viability and security patch cadence.

Question 2

A developer benchmarks two sorting algorithms and finds Algorithm A is faster on their test machine. What threat to internal validity should they address first?

Accepted Answer

Confounding variables such as dataset size and hardware warm-up

Answer

Confounding variables like dataset characteristics and CPU cache warm-up can invalidate benchmark results without proper controls.

Question 3

Which practice best supports reproducible research in software experiments?

Accepted Answer

Publishing code, datasets, and environment specifications together

Answer

Reproducibility requires sharing all artifacts—code, data, and environment—so others can independently verify results.

Question 4

A team wants to determine if a new caching strategy reduces API latency. What is the correct null hypothesis?

Accepted Answer

The new caching strategy has no effect on API latency

Answer

The null hypothesis always states there is no effect; the experiment attempts to reject it with sufficient evidence.

Question 5

When reading a case study claiming a microservices migration reduced deployment time by 60%, what is the strongest caveat to consider?

Accepted Answer

Results may not generalize because case studies reflect a single organization's context

Answer

Case studies provide rich detail but low external validity—organizational context, team size, and tooling heavily influence outcomes.

Full-Stack Development Practice Test

Full-Stack Development Practice Test

Full-Stack Development Research & Evidence-Based Practice 2