Behavioral Interview Questions: A Guide for Interviewers (Not Candidates)

Search "behavioral interview questions" and almost every result tells the candidate how to answer. This guide does the opposite. It's written for the interviewer who has to ask, score, and compare those answers, often three candidates deep, weeks apart, with no time to prep. The intent here is informational and practical: by the end you'll have a question bank, a scoring rubric, and a repeatable way to decide who actually demonstrated the skill versus who just told a good story.

Here's the contrarian frame the candidate-prep articles miss: a behavioral interview question is only as good as the rubric you score it against. The question does almost none of the work. The rubric does. Two interviewers can ask the exact same "tell me about a time you handled conflict" and reach opposite conclusions, because one is scoring the signal and the other is scoring the vibe. Asked exists to close that gap, and this guide explains the thinking behind it.

Why the Question Matters Less Than the Score

The interviewer side of the table has a measurement problem, not a question problem. You can ask a flawless behavioral question and still hire badly if every interviewer rates the answer on a private, shifting scale in their head. The fix is structure, and the research is unusually direct about how much it helps. Schmidt and Hunter's 1998 meta-analysis of 85 years of selection research put the predictive validity of structured interviews at .51, against .38 for unstructured ones (McDaniel et al. / Schmidt & Hunter meta-analysis). That gap is the entire argument for scoring deliberately.

Structure also fixes a problem you probably don't notice yourself doing. Jason Dana, Robyn Dawes, and Nathanial Peterson studied why people keep trusting unstructured interviews even after seeing them fail. Their answer is uncomfortable. As they put it, interviewers "have too many degrees of freedom to build a coherent story" out of whatever the candidate says (Dana, Dawes & Peterson, 2013, Judgment and Decision Making). You build the story, then you believe it. A rubric takes those degrees of freedom away on purpose.

So the rubric is the product, not the question. Across the interviews we have processed at Asked, the pattern is consistent: when two interviewers ask the same behavioral question without a shared rubric, their independent ratings agree far less often than either interviewer expects. The disagreement isn't about the candidate. It's about what each interviewer silently decided "good" meant. That silent definition is the thing this guide makes explicit, starting with the questions worth asking.

The Interviewer's Behavioral Question Bank

Build the question bank from the role, not from a generic list. Levashina, Hartwell, Morgeson, and Campion (2014), in their Personnel Psychology review, listed basing questions on a job analysis as the first structuring component that makes interviews work, ahead of the questions themselves (Levashina et al., 2014). Start by naming the three or four competencies the role actually needs, then attach two questions to each. That keeps the bank short, comparable, and defensible. If you want prompts grouped by competency to draw from, our 40 STAR Interview Questions for Interviewers (Organized by Competency) covers the durable ones, and our Interview Question Bank: 100 Structured Questions Ready to Use gives you a governed set to pull from across roles.

These are durable behavioral prompts you can map to most roles. Adapt the wording, keep the same structure for every candidate:

Conflict and disagreement: "Tell me about a time you disagreed with a decision your manager made. What did you do?"
Ownership under failure: "Describe a project that missed its goal. What was your specific role in the outcome?"
Prioritization under load: "Walk me through a week when you had more committed than you could finish. How did you decide what slipped?"
Influence without authority: "Tell me about a time you got a team to change direction when you weren't their manager."
Learning from feedback: "Describe a piece of critical feedback that changed how you work. What specifically changed?"

Notice that each prompt asks for a past event, not a hypothesis. Levashina and colleagues catalogued as many as 18 structuring elements, with about six used in a typical structured interview, and "ask every applicant the same questions" is one of the cheapest to adopt. The same wording for every candidate is what makes their answers comparable later, which is exactly what the next piece, the rubric, depends on.

How to Score a STAR Answer: A Step-by-Step Rubric

STAR is the scoring lens, not just a candidate trick. The four parts (Situation, Task, Action, Result) give you four places to check whether the candidate actually did the thing or just stood near it (The Predictive Index on STAR rating). For the weighted, component-by-component version with a worked example, see How to Score STAR Method Answers: A Rubric for Interviewers. Here is the procedure we recommend interviewers run on every behavioral answer:

Capture the answer in STAR slots first, judge second. As the candidate talks, sort what they say into Situation, Task, Action, and Result. Do not rate anything yet. Most weak answers reveal themselves here because the Action slot stays empty: lots of context, no first-person verbs.
Probe the gaps before scoring. If the Action or Result is thin, ask "what did you specifically do?" or "how did you know it worked?" A candidate who can fill the gap on demand scores differently from one who can't.
Score against a behaviorally anchored scale, not a feeling. Use a fixed 1 to 5 scale where each number has a written description, then circle the number whose description matches the answer.
Write one sentence of evidence next to the score. The sentence must quote or paraphrase what the candidate actually said. A score with no evidence sentence is a vibe, and vibes don't survive a hiring debate.
Score immediately, before the next candidate. Recency and halo bias both feed on delay. The scorecard finished in the room beats the one written after hours.

Here is the anchored scale itself. Print it. Every interviewer uses the same one:

Score	Anchor	What the answer shows
5	Exceptional	Complex situation, clear first-person actions, measurable result, and honest self-reflection on what they'd change
4	Strong	Well-defined situation, appropriate actions, a positive result, and some learning
3	Acceptable	Clear situation and logical actions, satisfactory result, limited reflection
2	Developing	Vague situation, few concrete actions, unclear or borrowed result
1	Inadequate	No relevant example, or a hypothetical instead of something they actually did

Behaviorally anchored rating scales aren't decoration. Levashina's review found they tend to raise the reliability and predictive validity of interview scores and may reduce bias against protected groups by pulling interviewer attention back to job-related information (Levashina et al., 2014, Personnel Psychology). The anchor descriptions do the disciplining. That discipline is what lets you compare two candidates you met two weeks apart, which is the comparison most interview processes get wrong.

Comparing Candidates Without Trusting Your Memory

Memory is the weakest tool in the room, and most processes lean on it hardest. By the time a panel sits down to decide, the strongest single moment from one candidate is competing against a faded summary of another. The fix is to compare scores and evidence sentences, not impressions. Because every interviewer scored the same competencies on the same anchored scale, you can lay the numbers side by side and argue about the evidence instead of the personalities.

Run a calibration session before anyone declares a winner. Each interviewer brings their independent scores, the panel surfaces where scores diverge, and the group resolves the gap by going back to what the candidate actually said. Making calibration a repeatable habit rather than a one-off is its own discipline, which our guide on Interviewer Training: How to Build Consistent Evaluators walks through in full. We have watched calibration flip a decision: a candidate two interviewers rated a 4 dropped to a 3 once a third interviewer pointed out the Result was a team result the candidate had described in the first person. The evidence sentence settled it, not seniority in the room.

This is also where combining methods pays off. Schmidt and Hunter found that pairing a structured interview with a general mental ability test reached a composite validity around .63, higher than either alone. The behavioral interview is one strong instrument, not the only one, and treating it that way keeps you honest about what a single conversation can and can't tell you. Honesty about limits leads straight to the objections most interviewers raise next.

Common Objections, Answered

"Structured scoring feels robotic and kills rapport." It doesn't have to. Structure governs how you score, not how you talk. You can be warm, curious, and human in the room and still drop the answer into STAR slots and an anchored scale afterward. The candidate experiences a good conversation. You experience a scorable one.

"My team won't follow a rubric." Then start with one competency and one shared question, not a full rewrite. Adoption fails when you hand interviewers a 12-page packet between two meetings. It succeeds when the structure fits the 30 minutes they actually have, which is the entire design constraint Asked was built around.

"We hire on instinct and it's worked." Maybe. But instinct is exactly what Dana and colleagues warned about: the coherent story you build feels like insight and is often just narrative fit. Keep the instinct as a tiebreaker. Make the rubric the primary record. That ordering, evidence first and gut second, is the practical core of the next section.

Quick-Start Interviewer Checklist

Use this before your next behavioral interview. It takes about ten minutes to set up and saves the scorecard from becoming an after-hours chore:

Pick three or four competencies tied to the actual role, not a generic list
Attach two behavioral questions to each competency, asking for past events not hypotheticals
Use the same question wording for every candidate
Print the 1 to 5 anchored scale so every interviewer scores against the same descriptions
Sort answers into STAR slots while the candidate talks, score after
Probe thin Action or Result slots before assigning a number
Write one evidence sentence next to every score
Score in the room, not days later
Hold a calibration session before anyone names a favorite

Frequently Asked Questions

How many behavioral interview questions should I ask in one interview?

Plan for three to five, one per core competency, with room to probe each. More than that and you trade depth for coverage, and the answers you can actually score well drop off. Asking the same set of every candidate matters more than asking many.

What's the difference between behavioral and situational interview questions?

Behavioral questions ask what the candidate actually did in a past event ("tell me about a time..."). Situational questions ask what they would do in a hypothetical. Levashina and colleagues studied both, and behavioral past-event questions give you real evidence to score rather than a rehearsed projection.

How do I stop one strong answer from skewing the whole evaluation?

Score each competency separately and immediately, and write an evidence sentence for each. Halo bias works by letting one impressive moment color unrelated scores. Separate, anchored, in-the-room scoring breaks that spread, and a calibration session catches whatever slips through.

Do This Next

Don't wait for a hiring committee to bless a new process. Pick one open role today and write down its three core competencies. Attach two behavioral questions to each, ask for past events, and keep the wording identical for every candidate. Print the anchored 1 to 5 scale and put it next to your notes for the very next interview. Score in the room, write one evidence sentence per number, and run a five-minute calibration before anyone declares a favorite. Then start a free trial of Asked and let the interview agent capture the transcript and draft the scorecard while you stay in the conversation, so you hire on what they said, not how you felt.