What primary teaching interviews actually test

The day is the assessment, not the interview

If you have been shortlisted for a class teacher post and are preparing by rehearsing answers to "what is your teaching philosophy", you are preparing for about a fifth of the day.

Most primary interviews in England are a full day or most of one. You arrive at half eight, you are shown round by two Year 6 pupils, you teach a twenty to thirty minute lesson to a class you have never met, you eat lunch in a staffroom where people are watching how you talk to them, you may sit with a pupil panel, you may mark a set of books or write a response to a piece of data, and then you sit in front of a panel for forty minutes. The panel decision is usually made in the light of everything that happened before it. Heads talk to the class teacher whose room you borrowed. They talk to the TA. They ask the office staff how you were when you arrived.

This matters because a candidate who prepares only for the formal questions is competing on the smallest part of the evidence. And because the day is designed to be hard to fake: you cannot rehearse how you respond when a child in the observed lesson says something you were not expecting.

Who is in the room

For a mainstream class teacher post in a maintained primary, expect the headteacher and usually the deputy or a phase leader. If the post has a subject responsibility, the current subject lead or a member of SLT with that remit may be there. A governor sits on most panels, and is required by many governing boards for appointments; they often ask the safeguarding question and one about why this school. In a multi-academy trust you may also get a trust-level director of education or a primary lead from another school in the trust, which changes the flavour of the questions — more about curriculum consistency across schools, more about willingness to teach a trust-wide scheme rather than your own planning.

The headteacher is assessing whether you can be left alone with thirty children in September. The phase leader is assessing whether you will be a burden or a help to the year group team — whether they will be planning for you. The governor is often assessing safeguarding and fit. They are all, quietly, assessing whether you will still be there in two years.

The person whose opinion you might not have considered is the class teacher whose class you taught. They saw the lesson from the back. They will be asked.

The observed lesson

You will normally get the brief a few days ahead: a year group, a subject, a time limit, sometimes an objective, sometimes a lot of latitude. Twenty to thirty minutes is typical. You will not know the children. You may get a seating plan and a note of who has EHCPs or who is EAL. You may get nothing.

What the observer is looking for is narrower than most candidates assume. They are not scoring the resources. In half an hour, with children you have never met, nobody expects to see mastery of a full sequence. They are watching for a handful of things:

Whether you notice. The single most common reason a lesson reads as weak is that the candidate delivers the plan while a third of the class are lost. Someone is not writing. Two children at the back have stopped. Do you see it? Do you do anything? A candidate who abandons the third activity because the misconception on the second one needed another five minutes is doing the job. A candidate who gets to the plenary on time with a room full of confused children is not.

How you handle wrong answers. Not whether you praise, but what you do next. "Good try, who else?" moves past the thinking. "Say more about why you got 43" goes towards it. This is the moment an experienced observer watches hardest, because it cannot be planned and it tells them exactly how you treat error in your own classroom.

Whether the adults in the room are used. If there is a TA and you walk in without speaking to them, and they spend thirty minutes standing by the sink, that is noted. A thirty-second conversation before you start — what you are doing, who you would like them to sit with, what you want them to watch for — costs nothing and reads as someone who has worked in a real classroom.

Behaviour, minimally. They know the class is excitable because a stranger is teaching. They are watching whether your expectations are visible and calm, whether you wait for silence rather than talking over it, whether you know the school's language — if the corridor is covered in a particular set of rules, using them shows you looked.

On subject: if you are given free choice, do not pick your flashiest lesson. Pick something with visible pupil thinking in it. A maths lesson where you have to respond to methods you did not anticipate shows more than a highly produced topic lesson where the children cut things out.

The pupil panel

Many primaries now put shortlisted candidates in front of a group of children, usually Year 5 and 6, sometimes school council. They ask their own questions, often written with a teacher's help, often the same five for every candidate. Common ones: what is your favourite book, what would you do if someone was being unkind, what is the funnest lesson you have taught, what were you like at school, do you have a pet.

The children's feedback is usually passed to the head and is usually not decisive on its own, but it can break a tie, and it can sink someone. What sinks people is talking down. Squeaky voice, over-simplified vocabulary, treating it as a game. Children notice immediately and they tell the teacher who collected them. Answer as you would answer an adult, with shorter sentences and a real answer. If you name a favourite book, name one you have actually read aloud and say what happened when you read it.

The questions that are actually testing something

Most panels ask between six and ten questions with a strict structure, because in a competitive appointment they need comparable evidence. A few carry far more weight than the rest.

"A child in your class discloses something to you. What do you do?" This is not conversation. Every panel asks a version of it and the answer is either right or it is not. The right shape: you listen, you do not promise confidentiality, you do not ask leading questions, you let the child use their own words, you write it down as soon as you can in the child's words with the date and time, and you report it to the Designated Safeguarding Lead the same day — immediately if there is a risk of harm. You do not investigate. You do not tell the parent. You may be asked what you would do if you reported it and nothing appeared to happen; the answer is that you escalate, to the head, and if necessary beyond the school, and that you know the school's policy tells you how.

A shallow answer here sounds like: "I'd reassure them and then I'd tell the safeguarding lead." It is not wrong, it is thin. It does not mention recording, it does not mention not promising secrecy, it does not use the language of KCSIE, and it gives the panel no evidence you have ever actually done this. If you have — and most teachers have — describe the mechanics without identifying details. That single change makes the answer unmistakably real.

"You have a child working two years below age-related expectations and one working well above, in the same maths lesson. Talk me through the lesson." The test is whether you can talk about adaptive teaching without reaching for three worksheets. What they want to hear is same objective, different scaffolds: concrete resources still on the table for the child who needs them, a pre-teach the day before, a deliberately harder version of the same problem rather than more of it, and how you would know by the end of the lesson who had understood. Naming specific things — Numicon, bar models, a manipulative you actually use, the school's calculation policy — separates you instantly from someone describing differentiation in the abstract.

A shallow answer sounds like: "I'd differentiate three ways and make sure everyone is challenged." That sentence could be said by someone who has never taught. It names no resource, no year group, no misconception.

"How do you know your teaching is working?" Or: talk me through your assessment. They are listening for whether you can distinguish what you do inside the lesson from what you do at the end of a unit. Mini-whiteboards, targeted questioning and the decision to reteach on Thursday because Tuesday's exit tickets were poor — that is formative. End-of-unit assessments, phonics screening data, the trackers that feed pupil progress meetings — that is a different conversation. Candidates who conflate the two tend to be the ones who collect data and do not use it.

"Tell me about a difficult parent conversation." Not a test of whether you win. A test of whether you stay professional, whether you know when to bring in the phase leader or the head, whether you follow up in writing, and whether you can describe a parent as anxious rather than difficult.

Early years and phonics. If the post is EYFS or KS1, expect direct questions on your phonics scheme and how you teach it. Schools have adopted specific validated SSP programmes — Little Wandle, Read Write Inc, Sounds-Write and others — and the panel will want to know whether you have taught the school's one, or can say plainly that you have not but have taught a different one and know that fidelity to the scheme matters. Bluffing here is transparent within two sentences, because the schemes have distinct routines and vocabulary and anyone who has taught one knows the difference.

"Why this school?" Lightweight-sounding, heavily weighted. Heads are appointing someone who will stay. "I want to work in a school with a strong community feel" is a sentence that fits any school in the country. What lands is something from the day you are standing in: a display you looked at, a thing a child said on the tour, the curriculum on the website, the Ofsted report's specific wording, the fact that they are two-form entry and you have always worked in one-form entry and want the year group team.

The things that decide it that nobody lists

The staffroom at lunch. Whether you asked the TA's name. Whether you tidied the room you borrowed. Whether, when the head asked how you thought the lesson went, you gave an honest self-evaluation or a defence. That last one is close to a hidden criterion: heads ask it to find out whether you are coachable, and a candidate who says "the second half of the input was too long, I'd have got them onto the task five minutes earlier and used the time on the misconception with the tens" has just demonstrated more about their practice than the lesson itself did.

Also: most primary schools tell you the same day. They often expect an answer on the spot. Decide in advance whether you would take it, because "can I have until tomorrow" is read, fairly or not, as ambivalence.

What to do before your next one

Read the school's latest Ofsted report and its curriculum pages, and write down three specific things you can refer to.

Write out your safeguarding answer in full, once, longhand. Include recording, not promising confidentiality, reporting to the DSL, and escalation. Say it aloud until it is a sequence rather than a list.

Pick a real lesson that went wrong and work out the sentence that explains what you changed. You will be asked something it answers.

For the observed lesson, plan the misconception you expect, and plan what you will do when you see it. That is the thing you will be judged on.

And find out the year group team's structure before the day — one-form, two-form, mixed-age — because half the questions about planning and workload assume you know.

If the harder problem is that you are doing all of this for posts that were never a fit — the wrong key stage, a subject lead responsibility you do not want, a trust whose model does not suit you — jobmarket.pro reads each advert in full and tells a candidate where they fit and where they do not before they apply.