TikTok·Data Scientist·Technical Phone Screen·Intermediate

Intermediate

Apr 2026

Summary

TikTok data scientist interview with a stats-heavy question about hypothesis testing on a billboard campaign. Pretty much one focused problem but it covered a lot of ground fast.

Questions Asked (1)

You have a billboard campaign where a sample of 100 users shows a 65% conversion rate. Set up the null and alternative hypotheses for a one-sided test against 60%, then compute the standard error, margin of error at 95% confidence, and the resulting confidence interval. Use the interval to decide whether to reject the null at the 0.05 significance level and explain what that means.

A/B Testing & ExperimentationProduct Analytics & Metrics

Author's notes

I knew the formula for SE but fumbled a bit explaining why we use the sample proportion in it rather than the null value.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

First, clearly state the null and alternative hypotheses for a one-sided test. Then, compute the standard error using the sample proportion, find the margin of error for 95% confidence, construct the confidence interval, and compare it to the null value to make a decision. Finally, interpret the result in the context of the billboard campaign.

Pro tip: When interpreting the confidence interval, emphasize that it provides a range of plausible values for the true conversion rate; if the null value falls outside this range, we reject the null. Also, mention that the one-sided test is appropriate if we only care about whether the conversion rate is greater than 60%.

1. State the hypotheses

Define the null hypothesis (H0: p = 0.60) and the alternative hypothesis (H1: p > 0.60) for a one-sided test.

2. Compute the standard error

Calculate the standard error using the sample proportion (0.65) and sample size (100): SE = sqrt(p_hat * (1 - p_hat) / n).

3. Determine the margin of error and confidence interval

For a 95% confidence level, the critical z-value is 1.645 for a one-sided test. Margin of error = 1.645 * SE. The confidence interval is p_hat ± margin of error.

4. Make a decision and interpret

Check if the null value (0.60) falls outside the confidence interval. If it does, reject H0 at the 0.05 significance level. Explain that this means there is sufficient evidence to conclude the true conversion rate is greater than 60%.

Key Points to Mention

Null hypothesis: p = 0.60; Alternative hypothesis: p > 0.60 (one-sided).
Standard error formula: SE = sqrt(p_hat * (1 - p_hat) / n).
For a one-sided 95% confidence interval, use z = 1.645 (not 1.96).
Margin of error = z * SE; confidence interval = p_hat ± margin of error.
Decision rule: If the null value (0.60) is not in the interval, reject H0.
Interpretation: Rejecting H0 means the campaign's conversion rate is statistically significantly greater than 60% at the 5% level.

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