← Upstart Interview Insights

Upstart·Data Scientist·Technical Phone Screen·Intermediate

Intermediate
Apr 2026

Summary

Upstart data science interview with a probability/statistics brain teaser that looks straightforward until you actually think about the sampling mechanism. The whole thing hinges on whether you catch the bias in how the kids were selected.

Questions Asked (1)

Q1

You survey 100 kids at a school and ask how many children are in their family. 50 say 1, 20 say 2, 30 say 3. You then knock on a random house in town and ask how many kids live there. What's your best estimate of the probability that house has exactly 1 child? State any assumptions you need about how the kids were sampled.

Product Analytics & MetricsAlgorithms & Data Structures
Author's notes

This one got me.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

First, clarify the sampling assumptions: the 100 kids are a random sample of children, so the distribution of family sizes is size-biased when sampling households. Then, compute the probability of selecting a household with exactly 1 child using the formula P(1 child) = (number of children in 1-child families) / (total number of children), which is 50 / (50*1 + 20*2 + 30*3) = 50/180 = 5/18 ≈ 0.278. Finally, discuss potential biases and whether the school sample is representative of the town.

Pro tip: Explicitly state that you're assuming the 100 kids are a random sample of all children in the town and that family sizes are stable; this shows you understand the size-biased sampling issue and avoids overcomplicating with unnecessary assumptions.

1. Clarify the sampling process

State that the 100 kids are assumed to be a simple random sample of children from the town, and that each child's response accurately reflects their family size.

2. Recognize size-biased sampling

Explain that when you knock on a random house, you're more likely to land on a family with more children because larger families occupy more houses (if each family has one house) or have more children to be sampled.

3. Compute the probability

Calculate the probability as the proportion of children who live in 1-child families: 50 / (50*1 + 20*2 + 30*3) = 50/180 = 5/18 ≈ 0.278.

4. Discuss assumptions and limitations

Mention that this assumes each family occupies exactly one house and that the school sample is representative of the town; otherwise, the estimate may be biased.

Key Points to Mention

  • Size-biased sampling: households with more children are more likely to be selected when sampling children.
  • The correct probability is the proportion of children in 1-child families, not the proportion of families.
  • Calculation: 50 / (50*1 + 20*2 + 30*3) = 50/180 = 5/18 ≈ 0.278.
  • Assumption: the 100 kids are a random sample of all children in the town.
  • Assumption: each family lives in exactly one house (no multiple families per house or multiple houses per family).
  • Potential bias if the school sample is not representative of the town's families.

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