Data Analyst Interview Questions

Group activity by week and count distinct users.

A strong answer defines the activity event, handles timezone and deduplication, then uses a weekly date bucket with COUNT(DISTINCT user_id).

Use it when you need row-level output plus aggregate context.

Window functions are useful for ranking, running totals, cohort calculations, lag comparisons, and moving averages without collapsing rows.

Inspect the plan, filter early, index join keys, and reduce scanned data.

Start with EXPLAIN, check joins, cardinality, missing indexes, large sorts, and whether predicates can use partitions or indexes.

It means the observed result is unlikely under the null hypothesis.

Interviewers expect you to separate p-value, confidence, sample size, effect size, and practical business importance.

They show a plausible range for an estimated value.

A good explanation avoids saying there is a fixed probability the true value is inside one computed interval; instead, describe repeated sampling behavior.

It happens when sampled data does not represent the target population.

Bias can come from selection, survivorship, non-response, instrumentation, or product exposure differences.

Validate the data, segment the metric, and compare against recent changes.

Check pipeline health first, then slice by time, geography, platform, cohort, channel, and product release to isolate the driver.

It connects a finding to a decision or next step.

Actionable analysis includes context, magnitude, confidence, tradeoffs, and a recommendation the team can evaluate.

Measure it, understand why it is missing, then choose a transparent treatment.

Depending on the cause, you may filter, impute, label missingness, or redesign collection. The key is documenting impact.

Match the chart to the comparison you need to show.

Use lines for trends, bars for category comparison, scatter plots for relationships, and tables for precise lookup.

It answers a recurring decision with clear metrics and context.

A strong dashboard has an audience, a refresh cadence, definitions, targets, owners, and minimal clutter.

Use honest scales, labels, baselines, and context.

Avoid truncated axes, overloaded colors, hidden denominators, and charts that emphasize noise over signal.

Normalization organizes tables to reduce redundancy and update anomalies.

Explain entities, keys, relationships, and tradeoffs. In analytics, denormalization may be acceptable for performance and usability.

A transaction is a unit of work that should satisfy ACID properties.

Good answers cover atomicity, consistency, isolation, durability, plus examples like money transfer or inventory update.

Indexes let the database find rows without scanning the whole table.

They improve lookup and join performance but add write overhead and storage cost, so they should match query patterns.