← Google Interview Insights

Google·Data Scientist·Technical Phone Screen·Senior

Senior
May 2026

Summary

Google data scientist interview with a heavy stats focus. The whole session basically revolved around one meaty confidence interval question with several sub-parts, which felt more like a stats oral exam than a typical DS interview.

Questions Asked (3)

Q1

You have a sample of n=100 with a mean of 100 and a 95% CI of ±10. If you increase n to 10,000 while keeping the sample standard deviation the same, what is the new CI half-width and why does it scale the way it does?

Product Analytics & MetricsA/B Testing & Experimentation
Author's notes

The math itself is straightforward once you remember the half-width is z times s over root n.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

First, recognize that the CI half-width is proportional to 1/√n. Then, compute the scaling factor: √(100/10000) = √(1/100) = 1/10. So the new half-width is 10 * (1/10) = 1. Explain that this occurs because the standard error of the mean decreases with the square root of the sample size, reflecting the averaging of independent observations.

Pro tip: Emphasize that this scaling assumes independent observations and constant population standard deviation; in practice, with large samples, you might also mention the central limit theorem ensuring normality of the sample mean.

1. Identify the relationship

State that the confidence interval half-width is proportional to the standard error, which is σ/√n. Thus, the half-width scales as 1/√n.

2. Compute the scaling factor

Calculate the ratio of the new sample size to the old: 10000/100 = 100. The square root of this ratio is 10, so the new half-width is the old half-width divided by 10.

3. Calculate the new half-width

Apply the scaling: old half-width = 10, so new half-width = 10 / 10 = 1.

4. Explain the intuition

Discuss that increasing sample size reduces variability of the sample mean because more data provides more precise estimates. The square root scaling reflects the diminishing returns of adding more samples.

5. State assumptions

Mention that this holds if the sample standard deviation remains the same and observations are independent. Also note that for large n, the t-distribution approaches the normal, so the critical value (≈1.96) remains essentially unchanged.

Key Points to Mention

  • Standard error formula: SE = s/√n
  • Confidence interval half-width = critical value * SE
  • Scaling proportional to 1/√n
  • Calculation: √(100/10000) = 1/10
  • New half-width = 1
  • Assumption of independence and constant standard deviation
  • Central limit theorem for large n

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

Q2

Without increasing sample size, what are at least four ways you could narrow the confidence interval, and what are the trade-offs of each?

A/B Testing & ExperimentationTechnical Trade-offsProduct Analytics & Metrics
Author's notes

This is where it got interesting.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying that narrowing the confidence interval (CI) without increasing sample size means reducing variance or increasing the signal-to-noise ratio. Then systematically present at least four methods, explaining the statistical mechanism and the practical trade-offs for each, ideally with examples relevant to A/B testing at Google.

Pro tip: Emphasize that these methods often involve trade-offs between statistical power, bias, and practical constraints, and that the best approach depends on the specific experiment and business context. Mention that some methods (like CUPED) are particularly effective in reducing variance without introducing bias, which is highly valued in tech companies.

1. Clarify the goal and constraints

Restate the question to ensure understanding: narrowing the CI without increasing sample size means reducing variance or increasing effect size. Acknowledge that sample size is fixed, so we must focus on other levers.

2. Brainstorm methods to narrow CI

List at least four methods: (1) reduce variance through better experimental design (e.g., stratification, blocking, paired designs), (2) use variance reduction techniques like CUPED or regression adjustment, (3) increase effect size by targeting a more responsive subpopulation or using a more sensitive metric, (4) improve measurement precision (e.g., reduce noise in data collection), (5) use one-sided tests or adjust confidence level (though this changes coverage).

3. Explain each method's mechanism

For each method, briefly explain how it narrows the CI: e.g., stratification removes between-strata variance, CUPED uses pre-experiment data to explain post-experiment variance, targeting increases the true effect size, better measurement reduces noise.

4. Discuss trade-offs

For each method, articulate the trade-offs: e.g., stratification requires known covariates and may complicate analysis; CUPED requires pre-period data and may introduce bias if not applied correctly; targeting reduces generalizability; one-sided tests increase Type I error risk; changing confidence level affects interpretation.

5. Conclude with recommendation

Summarize that the best approach depends on context, and often a combination of methods (e.g., CUPED + stratification) is used. Highlight that at Google, techniques like CUPED are standard for variance reduction.

Key Points to Mention

  • Stratification/blocking: reduces variance by accounting for known covariates, but requires pre-experiment data and can complicate analysis if many strata.
  • CUPED (Controlled-experiment Using Pre-Experiment Data): uses pre-period data to reduce variance, but requires pre-period data and careful implementation to avoid bias.
  • Targeting a more responsive subpopulation: increases effect size, but reduces generalizability and may introduce selection bias.
  • Using a more sensitive metric: e.g., continuous metrics instead of binary, but may not align with business goals.
  • Improving measurement precision: reduces noise, but may require better instrumentation or data cleaning, which can be costly.
  • Adjusting confidence level or using one-sided tests: narrows CI but changes error rates and interpretation.

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

Q3

Under what conditions is narrowing the CI by simply raising alpha misleading, and how would you explain that risk to someone without a statistics background?

A/B Testing & ExperimentationStakeholder ManagementTechnical Trade-offs
Author's notes

Blanked for a second here.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

First, clarify that raising alpha increases Type I error, so any 'narrower' CI is an artifact of a higher false positive rate, not more precision. Then explain the conditions where this is misleading—multiple testing, peeking, or when stakeholders misinterpret the CI as a precision measure—and use a non-technical analogy like a smoke detector to convey the trade-off.

Pro tip: Emphasize that the real risk is not just statistical but organizational: a wider CI with a lower alpha protects against costly false launches, while a narrower CI from higher alpha can erode trust when results fail to replicate.

1. Define the trade-off

Explain that alpha is the false positive rate; raising it makes CIs narrower but increases the chance of declaring a false effect.

2. Identify misleading conditions

List scenarios: multiple comparisons without correction, optional stopping (peeking), and when the CI is interpreted as a precision measure rather than a decision threshold.

3. Use a non-technical analogy

Compare to a smoke detector: lowering the threshold (raising alpha) makes it beep more often, but many beeps are false alarms—so you can't trust a 'quieter' alarm as evidence of fewer fires.

4. Connect to business impact

Explain that a narrower CI from higher alpha may lead to launching a harmful feature, wasting resources, or damaging user trust when the effect doesn't replicate.

5. Recommend alternatives

Suggest keeping alpha low, using sequential testing or corrections, and focusing on effect sizes and practical significance rather than CI width alone.

Key Points to Mention

  • Type I error rate and its relationship to confidence level
  • Multiple testing problem and family-wise error rate
  • Optional stopping / peeking and its effect on false positives
  • Misinterpretation of confidence intervals as precision measures
  • Business consequences of false positives (cost, trust, replication)
  • Alternatives: sequential testing, alpha spending, Bayesian methods

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