← Amazon Interview Insights

Amazon·Data Scientist·Technical Phone Screen·Intermediate

IntermediatePrefer not to say
Apr 2026

Summary

Got a stats question at Amazon that felt deceptively simple but I definitely fumbled parts of the explanation.

Questions Asked (1)

Q1

How do you determine the required sample size given a desired statistical power?

A/B Testing & ExperimentationProduct Analytics & Metrics
Author's notes

I knew the general idea, effect size, significance level, power, but when I tried to walk through it out loud I got tangled up in the direction of the relationship between sample size and power.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining the key inputs: significance level (α), power (1-β), effect size, and variance. Then describe how to use the appropriate formula or simulation to compute sample size, and emphasize the importance of estimating effect size from historical data or a pilot. Finally, discuss practical considerations like traffic constraints and business impact.

Pro tip: At Amazon, always tie sample size to business impact: a smaller effect size may require an infeasibly large sample, so discuss trade-offs between power, effect size, and duration. Also, mention sequential testing or Bayesian methods if applicable, as Amazon often uses them.

1. Define parameters

Specify significance level (α), desired power (1-β), minimum detectable effect (MDE), and variance (or baseline conversion rate).

2. Choose method

Select formula-based (e.g., for proportions or means) or simulation-based approach, depending on metric and complexity.

3. Estimate effect size and variance

Use historical data, pilot studies, or domain knowledge to estimate MDE and variance; be conservative if uncertain.

4. Compute sample size

Apply the formula or run simulations to calculate required sample size per variant, adjusting for unequal allocation if needed.

5. Validate and adjust

Check assumptions (e.g., normality), consider practical constraints (traffic, duration), and possibly adjust for multiple comparisons or sequential testing.

Key Points to Mention

  • Significance level (α) and power (1-β) trade-offs
  • Minimum detectable effect (MDE) and its business relevance
  • Variance estimation and baseline conversion rate
  • Formula for sample size (e.g., n = 2*(Zα/2 + Zβ)^2 * σ^2 / Δ^2 for means)
  • Simulation or bootstrapping for complex metrics
  • Adjustments for multiple testing, sequential analysis, or unequal allocation

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.