60 Market Research Assistant interview questions to ask job applicants

A market research assistant rarely gets caught making a mistake. They get caught six months later, when a product decision built on their survey turns out to rest on 40 responses from the wrong segment.

The best market research assistant interview questions test four things: whether the candidate understands method well enough to pick the right one, whether they check data before anyone analyses it, whether they can state a finding without overclaiming, and whether they keep several studies moving without losing track.

The 25 below market research assistant interview questions cover all four areas; each comes with what a strong answer looks like and the red flags that can indicate gaps in the candidate’s skills, experience, or judgment.

TL;DR

• Hire for method and accuracy, not enthusiasm about insights. An assistant who cannot spot a leading question produces confident, useless data.
• The 25 questions split across five areas: research fundamentals, data handling, analysis, fieldwork, and working style.
• Ask for a study they worked on and follow what they personally did. Most candidates at this level supported research rather than owning it, which is fine if they say so.
• Score the correlation question hardest. Overclaiming cause is the most common error at this level and the one that reaches a slide deck.
• Give a short data exercise first. Cleaning and checking are easy to measure and almost impossible to assess in conversation.
• Treat this as a feeder role. The assistant you hire well becomes your analyst in two years.

What does a market research assistant do?

A market research assistant supports studies rather than designing them. The daily work is desk research, survey programming, fieldwork chasing, data cleaning, chart building and a first pass at summarising what the numbers say.

The role is a genuine pipeline into analyst work, which matters for how you pitch it. The US Bureau of Labor Statistics reports a median annual wage of $78,760 for market research analysts in May 2025, across 952,700 jobs, with 7% projected growth through 2035 and roughly 82,000 openings a year.

Read that as a retention argument rather than a salary one. Candidates at assistant level can see the analyst number, so a role with no visible route upward loses people in month twelve to employers who drew the path.

Structure matters more at this level, because assistants rarely have enough history to separate themselves on experience. The US Office of Personnel Management defines a structured interview as one asking every candidate the same questions in the same order, with every answer rated against the same scale.

Research fundamentals questions

1. Walk us through a study you worked on. What was your part?

What a strong answer looks like: strong candidates name the objective, the method, the sample and what they personally did, separating their work from the team's.

Red flags: the candidate says "we" throughout and cannot name one task they owned. Supporting a study is fine, but they should know which parts were theirs.

2. What separates primary from secondary research, and when do you reach for each?

What a strong answer looks like: primary research collects new data directly, and secondary research uses data someone else already gathered. Strong candidates start with secondary to frame the question cheaply, then run primary only for what the existing sources cannot answer.

Red flags: the candidate reaches for a survey first every time. Running fieldwork to learn something a published report already covers wastes weeks.

3. How do you decide on a sample size for a study?

What a strong answer looks like: strong candidates connect sample size to the confidence needed, the size of the difference worth detecting, and how many subgroups they plan to analyse separately.

Red flags: the candidate quotes a round number with no reasoning. A sample of 400 means nothing until you know how many ways you cut it.

4. Explain qualitative and quantitative methods to someone who thinks they are interchangeable.

What a strong answer looks like: strong candidates explain that qualitative work tells you why and what language people use, and quantitative work tells you how many and how much. The best answers describe running them in sequence.

Red flags: the candidate treats qualitative research as a small survey. That misunderstanding produces focus groups analysed as if they were surveys.

5. How do you write a survey question that does not lead the respondent?

What a strong answer looks like: strong candidates name concrete faults: loaded adjectives, double-barrelled questions, unbalanced scales, and assumptions buried in the stem. The best answers mention piloting with a handful of people first.

Red flags: the candidate has never spotted a leading question. It is the most common defect in a draft questionnaire and the easiest to fix.

Data handling and accuracy questions

6. A survey comes back with 800 responses. What do you check before anyone analyses it?

What a strong answer looks like: strong candidates check completion rates, speeders, duplicate entries, screener logic failures and whether the achieved sample matches the quota they set.

Red flags: the candidate opens the file and starts charting. Everything downstream inherits whatever is wrong in that export.

7. How do you spot a respondent who is straight-lining, and what do you do about it?

What a strong answer looks like: straight-lining means selecting the same scale point down a whole grid. Strong candidates flag it alongside impossibly fast completion, set a rule before fieldwork, and document how many responses they removed.

Red flags: the candidate removes responses by instinct after seeing results. Choosing exclusions once you know the outcome shapes data to fit.

8. What do you do with outliers in a dataset?

What a strong answer looks like: strong candidates investigate before deleting, separating a data entry error from a real extreme value, and they report what they removed and why.

Red flags: the candidate deletes anything unusual. Some outliers are the finding, especially in spend and usage data.

9. How do you judge whether a secondary source is reliable enough to cite?

What a strong answer looks like: strong candidates check who funded it, who they sampled, when the fieldwork ran, and whether the number traces back to a primary source rather than another blog post.

Red flags: the candidate trusts anything in a search result. Circular citation is rife in market sizing, and assistants repeat it most often.

10. How do you document a piece of analysis so someone else can reproduce it?

What a strong answer looks like: strong candidates keep the raw file untouched, record every cleaning step and filter, note which version produced which chart, and name the file so a colleague can follow it.

Red flags: the candidate works in the master file with no record of changes. Nobody can check the number three weeks later, including them.

11. Which tools have you analysed data in, and what did you do in them yourself?

What a strong answer looks like: strong candidates distinguish tools they used from tools they watched, and they describe something they actually built, such as a crosstab, a pivot or a weighted summary.

Red flags: the candidate lists six platforms and cannot describe one output. Test this rather than taking the claim.

Analysis and insight questions

12. Your data shows a correlation. How do you talk about it without claiming cause?

What a strong answer looks like: strong candidates state the association plainly, name what else could explain it, and say what evidence a causal claim would need.

Red flags: the candidate presents correlation as cause. This is the most expensive error at assistant level, because it travels into a slide deck and then into a decision.

13. How do you turn a finding into a recommendation?

What a strong answer looks like: strong candidates connect the number to a decision someone has to make, state what they would do, and name the confidence behind it.

Red flags: the candidate stops at describing the data. A chart with no implication leaves the stakeholder to analyse it themselves.

14. The result contradicts what the stakeholder expected. What do you do?

What a strong answer looks like: strong candidates check their own work first, then report the finding plainly, show the method, and name the weakest point in their own conclusion before anyone else does.

Red flags: the candidate softens the result or buries it in an appendix. Research that only confirms expectations is not worth running.

15. How do you choose a chart for a finding?

What a strong answer looks like: strong candidates pick by what the reader needs to compare, use bars for categories and lines for change over time, and start a bar axis at zero.

Red flags: the candidate picks by what looks impressive. A truncated axis is the visual equivalent of overclaiming.

16. What belongs in a research summary, and what do you leave out?

What a strong answer looks like: strong candidates lead with the answer to the original question, support it with the two or three findings that carry it, and put method and full tables behind that.

Red flags: the candidate walks through the research chronologically. The person reading has ten minutes and needs the conclusion first.

Fieldwork and project handling questions

17. Recruitment for a study is running behind. What do you do?

What a strong answer looks like: strong candidates diagnose the cause, usually a screener that is too tight or an incentive that is too low, then propose a specific change and flag the timeline impact early.

Red flags: the candidate waits and hopes. Fieldwork rarely recovers on its own.

18. How do you keep several studies moving at once?

What a strong answer looks like: strong candidates name one tracking system and show it survived a busy month, usually a list organised by study with the next action and owner against each.

Red flags: the candidate works from their inbox. Fieldwork has dependencies that do not wait for someone to scroll back.

19. How do you handle sensitive or confidential research data?

What a strong answer looks like: strong candidates separate respondent identifiers from responses, restrict file access, report findings in aggregate, and know what they promised respondents in the consent wording.

Red flags: the candidate treats confidentiality as a file storage question. The promise made at the start of the survey is the binding part.

20. A stakeholder changes the research question mid-study. What now?

What a strong answer looks like: strong candidates explain what the current design can and cannot answer, offer the closest achievable alternative, and name the cost in time and money before agreeing.

Red flags: the candidate absorbs the change quietly. A questionnaire already in field cannot answer a new question.

21. How do you brief a panel provider or fieldwork vendor?

What a strong answer looks like: strong candidates give the screening criteria, quotas, incentive, timeline and what counts as a valid complete, then check the first returns before full launch.

Red flags: the candidate sends the questionnaire and waits. Soft launch checks stop a broken screener burning the whole sample.

Working style and fit questions

22. Describe a mistake you made on a research project.

What a strong answer looks like: strong candidates name the error, who they told and when, and the check they added afterwards rather than a promise to be more careful.

Red flags: the candidate cannot recall one, or fixed it quietly. Both answers tell you how the next error will surface.

23. How do you handle feedback on your analysis?

What a strong answer looks like: strong candidates want it early and specifically, and they can describe a piece of feedback that changed how they present findings.

Red flags: the candidate defends the first version. Analysis improves through challenge, and this role gets challenged weekly.

24. Which part of this work would bore you first?

What a strong answer looks like: honest candidates name data cleaning, chasing fieldwork or formatting decks, and they describe getting through it anyway.

Red flags: the candidate finds every part fascinating. Much of this job is repetitive, and pretending otherwise predicts a short stay.

25. Where do you want to be in two years?

What a strong answer looks like: strong candidates name a direction, usually analyst work, a method specialism or a category they want to know deeply.

Red flags: the candidate wants a path your team cannot offer. Naming that now beats discovering it mid-project.

Which questions fit each hiring stage?

Give each stage one job rather than running all 25 in a single sitting.

The data exercise does more work than any interview round here. Send a file with duplicates, a straight-liner and one impossible value, then ask what they would fix and flag.

How do you score the answers?

Rate every answer from 1 to 5 against written anchors, with one line of evidence beside each number.

Score independently before the panel talks, because the first opinion spoken anchors everyone else. Resumes barely separate candidates at this level, so the answers and the exercise carry the decision.

When should you test before interviewing?

Whenever the skill is measurable and you have more than a handful of applicants. For this role that means data handling, written clarity and structured reasoning, which an interview reads badly.

The market research test covers methodology and trend identification, the data analyst test covers the handling side, and a communications test shows whether someone can write a finding plainly under time pressure. For roles that lean toward problem framing, add a problem solving test.

If you would rather run a structured first round, then the market research methods interview covers qualitative and quantitative techniques, analysis and research ethics with scoring built in. Furthermore our guide to evaluating market research skills covers what to weight for each seniority.

Final thoughts

This hire fails quietly. A market research assistant who misunderstands method does not produce visibly bad work, they produce confident work that nobody questions until a decision built on it goes wrong.

The 25 questions above surface that gap early. Calibrate your interviewers on the strong answer profiles first, then use the red flags to stop a fluent answer from passing as a correct one.

Then do the cheap thing most teams skip. Send a messy dataset before the first conversation, and rate every answer against written anchors. Browse the test library to build the screen, or book a free demo.

Frequently asked questions (FAQs)