Data Analyst interview questions

Sample answer:

INNER JOIN returns only rows with matching values in both tables. LEFT JOIN returns all rows from the left table plus matched rows from the right table, filling unmatched columns with NULL.

Common mistake: Confusing which table is 'left' when the query has multiple joins.

Tip: Draw or describe a small example with two tables - it's the fastest way to prove understanding.

Sample answer:

It depends on why data is missing. I might drop rows if missingness is small and random, impute with mean, median or a model-based estimate, or flag missingness as its own category if it's meaningful.

Common mistake: Saying you'd just delete the rows without considering whether that biases the analysis.

Tip: Mention checking whether data is missing at random versus systematically.

Sample answer:

Correlation means two variables move together statistically. Causation means one variable directly influences the other. Correlation alone never proves causation - a third variable or coincidence could explain it.

Common mistake: Not providing a concrete example to illustrate the distinction.

Tip: Have one memorable example ready, like ice cream sales and drowning rates both rising in summer.

Sample answer:

A p-value is the probability of observing results as extreme as yours if the null hypothesis were true. A small p-value, commonly below 0.05, suggests the observed effect is unlikely to be due to chance alone.

Common mistake: Saying a p-value tells you the probability the hypothesis is true - a common misinterpretation.

Tip: Be precise about what a p-value does and does not tell you - interviewers often probe this.

Sample answer:

I would visualize the data with box plots or scatter plots, or use the IQR rule or z-scores to flag extreme values, then investigate whether they are errors or genuine rare events before deciding to remove, cap, or keep them.

Common mistake: Automatically removing every outlier without checking whether it is a real, meaningful data point.

Tip: Emphasize investigating the cause before deciding how to treat an outlier.

Sample answer:

A right-skewed distribution has a longer tail on the right, with the mean typically greater than the median. A left-skewed distribution has a longer tail on the left, with the mean typically less than the median.

Tip: Give a real example: income distribution is typically right-skewed due to a small number of very high earners.

Sample answer:

I would lead with the business implication first, use visuals over tables, avoid jargon, and tie the finding back to a decision they need to make.

Common mistake: Walking through your full methodology before getting to the takeaway.

Tip: Practice a 'headline first' structure - most stakeholders only need your conclusion.

Sample answer:

I would check the shape and data types, look for missing or duplicate values, review summary statistics, visualize key distributions, and clarify with stakeholders what questions the data should answer.

Common mistake: Jumping straight into modeling before doing basic data exploration.

Tip: Mention this as a repeatable checklist - it shows structured thinking, not guesswork.

Sample answer:

Summarize your analytics background, a project or two that show real business impact, and why this role interests you - under two minutes.

Common mistake: Listing every tool you know instead of telling a coherent, relevant story.

Tip: Pick one or two achievements most relevant to this specific job.

Sample answer:

Explain the business question, your approach, the key finding, and most importantly what actually changed as a result.

Common mistake: Focusing on technical details while skipping the actual business impact.

Tip: End with the outcome in business terms, like revenue, cost, or efficiency.

Sample answer:

Be honest about a real mistake, how you or someone else caught it, how you corrected it and communicated the correction, and what you changed afterward.

Common mistake: Claiming you've never made a significant analytical mistake, which reads as lacking self-awareness.

Tip: Interviewers care more about how you handled the mistake than the mistake itself.

Sample answer:

I clarify business urgency and impact with stakeholders, communicate realistic timelines, and flag trade-offs early rather than silently falling behind.

Tip: Mention proactively communicating trade-offs - that separates a strong analyst from someone who says yes to everything.

Sample answer:

Describe the specific data quality issues, the steps you took to clean or validate the data, and how you communicated the resulting confidence level to stakeholders.

Common mistake: Presenting cleaned results without mentioning data quality caveats.

Tip: Transparency about limitations builds more trust than pretending the data was perfect.

Sample answer:

I sanity-check results against known benchmarks, look at the data from more than one angle, and ideally have someone else review key numbers.

Tip: Mention a specific habit, like re-deriving a number a second way to confirm it.

Sample answer:

Explain the request, why it wouldn't answer the real business question, and how you proposed an alternative that still met their underlying need.

Common mistake: Framing it as simply refusing instead of offering an alternative.

Tip: Always pair a 'no' with an alternative.

Sample answer:

Define the hypothesis clearly, identify the right metric and comparison group, pull and clean the data, run the appropriate test, and interpret results with attention to confounding factors.

Common mistake: Skipping the step of clearly defining the hypothesis before pulling data.

Tip: Mention confounding variables specifically - a strong signal of statistical maturity.

Sample answer:

I'd explain specifically why the design could mislead, propose an alternative representation, and if they still want their version, add context or caveats directly in the dashboard.

Common mistake: Building exactly what was asked for without raising the concern at all.

Tip: Frame your pushback around the risk to their decision-making, not just best practices.

Sample answer:

Notify the relevant stakeholders immediately with a clear explanation and corrected numbers, rather than waiting or quietly fixing it for next time.

Common mistake: Delaying disclosure out of concern about how it will look.

Tip: Speed and transparency matter more here than looking perfect.

Sample answer:

I'd clarify both requests' urgency with their respective owners, and involve my manager to help set priority if needed rather than silently picking one myself.

Tip: Mention involving a manager when priorities genuinely conflict.

Sample answer:

Tie specific things about their data, product, or business model to your own analytical interests and experience.

Common mistake: Giving a generic answer about 'working with data' that could apply to any company.

Tip: Mention a specific product feature or business metric you'd be excited to analyze.