What Interview Boards Actually Score: The Evidence Behind Structured Interviews
Most candidates prepare for an interview as though it were a conversation to be won. Most formal boards, in education and in the wider public and corporate sectors, are running something closer to a measuring instrument. Understanding the difference changes how you prepare, and the research on why it works is worth knowing.
Structured and Unstructured Are Genuinely Different Instruments
An unstructured interview is a conversation. Questions vary between candidates, follow-up is improvised, and the panel forms an overall impression which is then justified after the fact. A structured interview is a standardised assessment. Every candidate faces the same core questions, in the same order, mapped to a defined set of competencies, and each answer is scored against a rating scale before the panel moves on.
The consistent finding across decades of personnel selection research is that the structured version predicts subsequent job performance substantially better than the unstructured version. That finding is not seriously disputed. What has been disputed, and revised, is exactly how large the difference is.
The Classic Citation, and Why It Needs a Caveat
The paper most often quoted is Schmidt and Hunter's 1998 meta-analysis in Psychological Bulletin, which placed structured interviews at the top of the validity hierarchy alongside general mental ability tests. For twenty years it was the standard reference, and its numbers were repeated in training materials and recruitment textbooks as settled fact.
They are not settled. In 2022, Sackett, Zhang, Berry and Lievens published a re-analysis in the Journal of Applied Psychology which identified a systematic overcorrection for range restriction in the earlier meta-analytic work. Correcting for it produced materially lower operational validity estimates across most selection methods. In the revised hierarchy, structured interviews came out at roughly .42 and cognitive ability tests at roughly .31, which reordered the two rather than simply shrinking both.
Two honest conclusions follow. First, anyone still quoting the 1998 figures as current is working from a superseded source. Second, and more importantly for candidates, the practical direction of travel did not change: structured interviewing remained the strongest single predictor examined. The correction reduced the confidence with which the specific coefficients should be quoted. It did not overturn the underlying case for structure.
It is also worth saying plainly that these are correlations across large samples, not guarantees about any individual hiring decision. A validity coefficient in this range describes a real but modest predictive relationship. Boards are working with a better-than-nothing instrument, not a precision one, and good boards know it.
What a Scoring Rubric Actually Looks Like
The mechanism that makes a structured interview work is the rating scale. In its most developed form this is a behaviourally anchored rating scale, usually shortened to BARS. Each competency is broken into proficiency levels, and each level carries a written behavioural anchor describing what an answer at that level actually sounds like.
In practice, a panel member sitting in front of you typically has:
- A named list of competencies drawn from a published framework, often four to six of them.
- A fixed set of questions, each mapped to one or more of those competencies.
- A numeric scale per competency, commonly running from below expectation through fully meets to exceeds.
- Written anchors describing the behaviour that earns each score.
- Space for verbatim evidence, because the anchor has to be justified against something you said.
The research literature on BARS associates their use with higher predictive validity and reliability, and with reduced bias, and rater training is treated as part of the method rather than an optional extra. This is why well-run boards score each answer before moving on, and why they sometimes seem to write more than you expect during a short answer.
What This Means for How You Answer
If the panel is filling in evidence boxes against anchors, several things follow directly.
Give them evidence they can write down. An anchor cannot be justified against a statement of belief. It can be justified against a specific action you took and its outcome. This is the real reason structured frameworks such as STAR are recommended: not because the acronym is magic, but because it forces you to produce scoreable material.
Answer the competency, not just the question. Questions are proxies. If a question is mapped to managing conflict, an answer that is warm, likeable and entirely about something else scores nothing on that line.
Do not assume credit carries across questions. Each competency is scored separately. Evidence you gave under one question does not automatically populate another box. If a strength is relevant twice, use it twice, framed differently.
Expect the same questions as everyone else. Standardisation is the point. The panel is not warming to you or cooling to you across the session, and a friendly opening question is being scored exactly like the rest.
Use I rather than we. The scale measures your competence. Collective language is one of the most common reasons a strong example scores poorly.
Where the Evidence Is Weaker Than People Claim
Be sceptical of confident precision in this field. The corrections applied by Sackett and colleagues are themselves debated, and the meta-analytic literature depends on primary studies of varying quality across very different jobs and eras. Validity estimates are averages across contexts, and they say little about any specific board's competence in applying its own rubric. Panels vary in rater training, and a well-designed rubric operated badly does not deliver the validity the research describes.
What survives all of that caution is worth acting on: structure helps, evidence beats assertion, and the candidate who understands the instrument prepares differently from the candidate who prepares to have a good chat.
Our Get the Job series applies this to UK competency interviews, with the flagship title plus dedicated books for Project Manager, Accounts and Finance, Nursing and HR roles. For Irish school leadership boards, the To Lead series works through the same scoring logic against the competency frameworks used for Deputy Principal, Principal and Assistant Principal posts.