# Materials Science interview questions: turn every choice into a trade-off

> Ten original practice problems for Imperial’s Materials Science applicant day, where a 30 minute talk and a team task carry the selection. No admissions test.

[Formatted version of this page](https://www.frontiervue.com/interviews/questions/materials-science)

The Materials Science and Engineering MEng page is blunt: **this department does not use a test as part of its selection process**. The only live contact you get is the **applicant day**, a Wednesday between November and March, 10.30 to 15.30, holding a 30 minute one-to-one and a 45 minute engineering team task. With no test score speaking for you, your statement and your behaviour on the day are read closely. The ten problems below are **original practice written for this site**.

## Key takeaways
- This department uses no admissions test, so the applicant day is the only live evidence you give them, and it should not be prepared last.
- The one-to-one is described as “more of a conversation” and is explicitly two-way, so the questions you bring count as content.
- The group task rewards the person whose idea others can pick up, not the first one to an answer.
- Materials answers are mostly “it depends”. Name the property you are optimising for, then the condition under which you would switch.
- Every problem here is original and none of them predict the day. What you can train is starting to talk, showing the working, and turning under a follow-up.

### The Materials applicant day at a glance

| Item | Value |
| --- | --- |
| Admissions test | No test is used in selection |
| Format | Applicant day, a Wednesday from November to March |
| Length | 10.30 to 15.30, attending the whole day |
| One-to-one | 30 minutes with a lecturer, meant to be a conversation |
| Group task | 45 minute engineering challenge, watching how you team-work |

> Invitations and dates are set by the department, which contacts you at the email address on your UCAS form. The ten problems here are original practice for this site and are not what you will be asked.

> What the format looks like: The course page says you will be invited to an applicant day if your UCAS application indicates you are likely to satisfy the requirements. Applicant days run on Wednesdays between November and March from 10.30 to 15.30, and you are expected to stay the whole day: talks from the Director of Undergraduate Studies and the Admission Tutor, then a 30 minute one-to-one interview, a 45 minute engineering group task and a 45 minute Q&A with student ambassadors. Imperial’s central page adds that interviews usually fall between November and February, may be virtual or in person, and that the department contacts you itself. Whether you are invited, when and in what form is the department’s decision; the ten problems below are original practice, not the questions of that day.

## Contents
1. What the applicant day actually is
2. With no test, what the day has to carry
3. What a lecturer is judging in thirty minutes
4. How to rehearse a 45 minute group task
5. The numbers, and how this differs from Oxbridge
6. How to use these ten problems, and the official checklists

## 1. What the applicant day actually is

It is called an applicant day, not an interview day: after talks from the Director of Undergraduate Studies and the Admission Tutor come **a 30 minute one-to-one with a lecturer**, **a 45 minute group task**, “an engineering-based problem-solving challenge to see your team working skills”, and **a 45 minute Q&A with student ambassadors**, on a Wednesday from 10.30 to 15.30.

Two things follow. You are under observation far longer than thirty minutes, since the group task and the student session count too. And the day runs both ways: the page says the one-to-one is as much for you to decide whether Imperial suits you as for them to judge you.

1. **Two talks** (Listen with a pen)
These talks are not background noise; every question you ask later stands on them.

2. **The 30 minute one-to-one**
It opens on your statement and interests, then follows your answers instead of a card of set questions.

3. **The 45 minute group task** (The longest observable block)
Nobody is credited for solving it alone; what is noticed is who fixed the brief and who built on another’s idea.

4. **The 45 minute student Q&A**
An open session with ambassadors, where you settle what the four years actually involve.

## 2. With no test, what the day has to carry

Imperial says most departments interview applicants who show potential in their UCAS application “and in an admissions test (where this applies)”. Materials sits in the does-not-apply column, so what is left is the form itself, results gained and predicted, the referee’s report, and what you do on the day.

Three consequences. Your statement carries more weight, which is why the first published tip is to reread it. Nothing compensates for a shaky corner of your foundation, and nothing has to explain it either. And the group task is fresh evidence that appears in no standardised document. In our sessions a two hour rehearsal is split forty minutes one-to-one, thirty minutes on the team task, the rest reviewing the recording.

> The ten problems here are **original interview practice written by Frontier Academy for the applicant day**. What you are asked depends on how you answer, so prediction is useless; what you can train is starting to speak, showing your working and using the follow-up.

## 3. What a lecturer is judging in thirty minutes

“More of a conversation” is about format, not politeness: there is no shared question card, and the lecturer follows wherever your answer goes. In our sessions we sort what can be observed into four things. **Property language**: can you trade “use aluminium here” for “stiffness per unit mass dominates, because self-weight sets the life”. **Visible assumptions**: do you read out your numbers before the conclusion. **Redirection**: a condition gets added, hotter, more cyclic, cheaper, weldable, and the signal is whether you re-run the reasoning. **Why you**: whether you are taken by the word “materials” or chased by “why did this part fail”.

- **Course fit:** Can you name the kind of problem you want to work on
- **Visible method:** Property and assumption first, conclusion second
- **Coachability:** Do you change course when a condition changes
- **Two-way judgement:** The questions you ask are evidence too

## 4. How to rehearse a 45 minute group task

Group tasks fail in a few familiar ways, rarely for lack of skill. The commonest is four people calculating in silence until the clock runs out with two incompatible answers. Next is one person talking throughout while the rest become an audience, which reads badly when the stated purpose is team working. Third is discussion that never lands, since a proposal needs a sketch, a few numbers and a reason.

Our rehearsal rules are three: write the requirements on one line before anyone calculates; circulate one sheet, with each idea stated in a sentence and repeated back by someone else; and stop in the last five minutes to settle who presents and how you answer “why this”.

### Opening with a solution

- Wrong: “Let us just use carbon fibre.”

- Right: “Three requirements first, then two candidates against them.”

- How to spot it: Nobody can build on your idea because the target was never written down.

### Finishing every sentence

- Wrong: “I have finished, do it my way.”

- Right: Push the paper to the middle and ask where everyone else got to.

- How to spot it: At the end the group has only your answer to present.

### Calculating in silence

- Wrong: Four heads down for forty minutes.

- Right: Look up halfway and read the interim result so the group re-syncs.

- How to spot it: Whoever presents cannot say where the number came from.

### Agreement that is not agreement

- Wrong: “Whatever, yours is fine.”

- Right: “I am unsure about that assumption. Can we spend two minutes testing it?”

- How to spot it: The disagreement has not gone away, it has only gone underground.

## 5. The numbers, and how this differs from Oxbridge

Start with the fixed information. The A-level minimum standard is AAA, with an A in Maths, an A in Chemistry or Physics and a third useful subject, and the typical offer is stated as A\*AA; the IB minimum is 38 points with 6 at Higher Level in Maths and in Physics or Chemistry, 39 typical; the UCAS codes are JFM2 and I50; the equal-consideration deadline is 18.00 UK time on 13 January 2027; and the course sits inside the ATAS scheme under code CAH10-03-02.

Two readings matter. The deadline is in mid-January, so **the department reads you while your A levels are nowhere near finished**, which is why questions 1 and 8 deserve the earliest work. ATAS and the course code are administration and change nothing about what you are asked. And Imperial’s event is a **full day on site**, given as 10.30 to 15.30, while Oxford and Cambridge run half-hour subject conversations.

- **A-level minimum:** AAA with an A in Maths and an A in Chemistry or Physics, plus a third subject; typical offer A*AA
- **IB:** Minimum 38 points with 6 at Higher Level, 39 typical
- **UCAS codes:** Course JFM2, institution I50
- **Deadline:** 13 January 2027 at 18.00 UK time
- **ATAS:** Subject code CAH10-03-02
- **Admissions test:** The department uses no test

|  | Dimension | Imperial Materials | Oxford | Cambridge |
| --- | --- | --- | --- | --- |
| When | Applicant days on Wednesdays, November to March | December 2026, online | Main period 7 to 18 December |
| Live components | A 30 minute one-to-one, a 45 minute group task and a 45 minute student Q&A | Academic discussions of your subject; many applicants have more than one | Usually one or two, 35 minutes to an hour in total |
| Admissions test | No test used by this department | Depends on the course; see the subject’s own page | Depends on the subject; some add an assessment around the interview |

> Oxbridge dates come from the two universities’ published interview pages and the Imperial dates from the Materials course page. All three institutions assess independently; this table compares published formats only, not difficulty or chances.

## 6. How to use these ten problems, and the official checklists

1. **Try it alone, timed** (5 to 8 min)
Write every route you can see, including the failed ones, and mark the step where you stopped.

2. **Say it aloud in English**
Record three minutes in English and ask one thing on playback: could anyone follow this without the paper?

3. **Have someone push back** (The step that matters)
Ask someone outside your subject to use two lines only: why that assumption, and what if you heat it.

4. **Change a condition and redo it**
Swap single loading for cyclic and room temperature for three hundred degrees; if nothing moves, you missed the dependency.

### Imperial’s own tips, converted into actions

Imperial’s tips are reread your statement, since follow-ups come from what you wrote; take a school mock interview; practise body language and concise answers; arrive early; prepare questions; keep your invitation; no dress code. Our order is statement audit first, speaking practice second, logistics last.

### Turn your personal statement into a map of follow-ups

Copy out every “I am interested in X” and attach three things: a detail only you would know, one record of changing your mind, and one question you cannot answer. Cut the sentence that cannot produce the third, because the interviewer drills three levels into whatever you genuinely know.

### Online or in person: the email decides

The course page describes a full day while the central page allows virtual or in person, so go by the invitation you received. If it is online we add three things: camera at eye height, paper on the desk rather than a phone, and an early login with a backup connection, plus one rehearsal of writing where a camera can see it.

## 10 Materials Science interview problems

Every problem below is original practice written for this site, not a past paper. Read the question, try it aloud, then compare your reasoning with the approach.

### Question 1

**Prompt:** Why materials rather than mechanical or chemical engineering? And what in this course specifically feels essential to you?

**What this tests：** The department describes the thirty minutes as a conversation whose job is to judge whether you suit the course, so the opening question is not small talk. The lecturer is listening for one checkable signal: can you state the difference between these three routes, or only that you like science? The course requires Maths alongside Chemistry or Physics, which is the crossover it is aiming at, and you need to fit that shape out loud.

**How to work through it：**
1. One sentence for the entrance: the experiment or the broken object that started it. Skip the background and name the scene.
2. Narrow the interest into one testable question and say which half of it you do not know. For instance “why does the same aluminium alloy carry less load the more times it is loaded”, then admit that you can describe it qualitatively but cannot yet predict it.
3. Aim at something you can check: the course asks for Maths plus Chemistry or Physics, which places materials at the meeting point of three subjects. Say which of the three excites you most, with an example.
4. Close with the thing you hope to be asked about today. That turns “I like materials” into “I know which kind of question I want to be pushed on”.

**The likely follow-up:** If you had applied to mechanical engineering instead, what would you be missing? Name one concrete thing.

### Question 2

**Prompt:** Why is a phone screen made of glass? Name one property you accept losing because of that choice, and then tell me what you would change.

**What this tests：** This tests how a materials selection is actually made: list the properties that fight each other, then admit what you gave up. Saying “glass is hard and scratches less” just recites common sense; the interviewer wants to see you put transparency, hardness, fracture toughness, cost and repairability on one table and weigh them.

**How to work through it：**
1. Start with what the object meets daily: grit in a pocket (hardness and scratching), impact on the floor (shock and crack growth), skin contact (surface energy and fingerprints), and the need to transmit light.
2. Name the conflict: a material that is both transparent and hard lives in the glass and ceramic family, whose stiffness comes from strong ionic or covalent bonding, and the price is that dislocations cannot move, so it fails with little plastic deformation.
3. Show the working aloud: a polymer wins on transmission and toughness but loses badly on hardness, and a metal is tough but opaque. So the screen sacrifices crack tolerance to keep hardness, then buys that back with chemical strengthening to raise the crack threshold.
4. Make the change conditional: “if this were a site device I would put crack tolerance ahead of scratch resistance, take a thicker polymer or an ultra-thin glass and polymer laminate, and accept that it marks more easily”.

**The likely follow-up:** If I halve the glass thickness, which failure worries you first, and why?

### Question 3

**Prompt:** In winter, pouring boiling water into a cold ceramic mug can crack it, while a metal kettle will not. Why?

**What this tests：** This tests the chain from structure to properties: expansion, the temperature gradient conduction creates, and the fact that ceramics are weak in tension with almost no plasticity. The interviewer is listening for where the stress comes from and why the material cannot shed it by deforming, not for the word “brittle”.

**How to work through it：**
1. Start with the inside heating first: the inner wall wants to expand, the cold outer wall refuses, so the inside goes into compression and the outside into tension. Saying which surface is loaded which way shows the thinking better than any formula.
2. Then why ceramics cannot take it: directional bonding keeps dislocations immobile, so the tensile surface cannot yield and relax the stress, and a brittle material is hypersensitive to crack tips, so an existing micro-crack simply runs.
3. Finish with the metal as the control: it conducts heat fast so the gradient is small, and once it yields it redistributes stress plastically, so the same water is nothing to it.
4. Convert the answer into a design judgement: thermal shock resistance wants low expansion, high conduction and high toughness together, which is why lab glassware is borosilicate and why thick ceramic mugs lean on porosity and wall thickness instead.

**The likely follow-up:** Why does the same water not crack a polypropylene cup? Polymers fail in a different way under sustained load: describe it.

### Question 4

**Prompt:** Estimate how much copper is used in the electrical wiring of an ordinary three-bedroom house. Narrate every move as you make it.

**What this tests：** This tests estimation habits: do you decompose the object, do you keep units, do you convert volume to mass through a density, and above all do you read your assumptions out loud. The interviewer is not checking accuracy, they are checking whether you can sanity-check yourself.

**How to work through it：**
1. Decompose before numbers: copper sits mainly in the fixed wiring (socket and lighting circuits), the consumer unit and inside appliances. Pick the biggest single block and start there, not the hardest.
2. State the assumptions in a form you can calculate: “assume fifteen metres per socket circuit, three cores, 2.5 square millimetres, eight of them; lighting at 1.5 square millimetres, four circuits”. Say plainly that these come from you, not from a lookup.
3. Say the units as you go: 2.5 square millimetres is 0.025 square centimetres, forty-five metres is 4500 centimetres, so about 110 cubic centimetres; copper is roughly 9 grams per cubic centimetre, so about a kilogram a circuit and a fixed-wiring total in the tens of kilograms.
4. Check yourself and turn it into a materials question: name the assumption you trust least (number of circuits, or length), then answer the follow-up before it is asked: “aluminium would weigh roughly a third at the same volume, but needs a larger cross-section for the same current, and the terminations become a different problem”.

**The likely follow-up:** Does your estimate change if the house was built thirty years ago? Which term moves?

### Question 5

**Prompt:** Here are four materials as Young’s modulus in GPa, tensile strength in MPa and density in g per cubic centimetre: A is 3, 60, 1.2; B is 70, 300, 2.7; C is 200, 400, 7.8; D is 130, 1500, 1.6. Which would you choose for a bicycle frame?

**What this tests：** This tests your first move with data: do you ask which quantity the task is minimising. On a frame, weight sets the cost of accelerating and climbing, so the numbers to compute are stiffness per unit mass and strength per unit mass, not simply the largest modulus. The interviewer is also listening for whether you do the division out loud and whether you name the columns missing from the table.

**How to work through it：**
1. Write the objective as two or three ranked lines: it must not flex under load (stiffness), it must survive a crash (strength and toughness), and it must be light (density). Say which one you put first, because every later step hangs on that.
2. Do the specific stiffness aloud: A about 2.5, B about 26, C about 26, D about 81. State what that shows: B and C tie, so steel and aluminium are near-equivalent per unit weight, and D sits in another band.
3. Then the specific strength: B about 110, C about 50, D about 900, A about 50. Offer a provisional conclusion with its condition attached: on these three columns alone, D leads.
4. Close with what the table does not contain: no direction for D (a unidirectional composite is nothing like the same across the fibres), no fracture toughness, no fatigue, no jointing or cost, no crash behaviour. Say those out loud, then state the conditions under which you take D and the ones under which you fall back to B.

**The likely follow-up:** There is no fracture toughness column. If this were a real frame selection, which column would you demand and why that one?

### Question 6

**Prompt:** Carbon fibre is light and strong, and people often call it the green choice. If a sports component or a turbine blade is carbon-fibre composite, what are your real options when it retires?

**What this tests：** This tests life-cycle reasoning: can you split “green” into three separate accounts, being energy, performance and feasibility, and can you admit that the value of recycling depends on how long the part served. Shouting “it is recyclable” dodges the question; the interviewer wants you to say plainly what downcycling means.

**How to work through it：**
1. Start with where the value sits: carbon fibre buys stiffness and strength per unit weight. Any recovery route that shortens the fibres or damages the interface takes the material out of its original duty.
2. Make the difficulty concrete: fibre and resin are one object, so separating them needs pyrolysis, solvents or aggressive mechanics, all of which consume energy and turn long fibre into short fibre. That is why most recovered material goes to secondary duty rather than back into a structural part, which is downcycling.
3. Put both sides in one unit: how many years the part serves and how much fuel or generation it saves in that time, against what recovery hands back. A twenty-year component and a three-year one can get opposite verdicts from the same fibre.
4. Finish with something a designer can do: ask whether a single material or a dissolvable matrix could be used, whether a bolted joint could replace the adhesive, so disassembly is designed in rather than improvised at end of life.

**The likely follow-up:** If the customer only wants two years of life, does your answer change, and does aluminium beat carbon fibre?

### Question 7

**Prompt:** In the 45 minute group task, your four-person group has been working for twenty minutes. One member has not spoken at all and another insists your direction is wrong. What do you do?

**What this tests：** This maps directly onto the element the course page describes: an engineering problem-solving challenge used to see your team working skills. So the answer cannot be “I am a good communicator”; it has to be executable behaviour: the specific question that gets the quiet person talking, and the method that converts a disagreement into a testable assumption.

**How to work through it：**
1. Stop calculating and restate the brief to the group in one sentence: what we are being asked, what we must hand over, how much time is left. This re-aligns everyone and gives the quiet member a zero-cost way in.
2. Ask the silent member something small and specific rather than “what do you think”: “this step assumes repeated loading, is that right? Two seconds is enough”. Lowering the bar to one word works better than inviting an opinion.
3. For the disagreement, write each option on one line with the assumption it rests on. Do not argue who is right; compare which assumption is cheaper to test. Test that one immediately, and if neither is testable, allocate the remaining time explicitly to one.
4. At the end, spend thirty seconds making the decision the group’s rather than yours: “we took A because B needed the data we did not have time for”. That is the sentence an interviewer most wants to hear, and the one they write down.

**The likely follow-up:** If someone is still not engaging in the last ten minutes, do you sacrifice progress or coherence of the result? Explain your trade-off.

### Question 8

**Prompt:** You mention a project you did. Take me through the worst thing that went wrong in it: what did you decide at the time, and what have you changed since?

**What this tests：** Imperial’s first published tip is to reread your personal statement, with follow-ups built from what you wrote. In this question the lecturer is verifying two things at once: whether you actually did it, through the dated and quantified details, and whether you can admit the limits of your own method. A project with nothing wrong in it tends to be pushed until it breaks.

**How to work through it：**
1. Tell the whole timeline in sixty seconds using three sentences: what I first believed, when the numbers started disagreeing, when I admitted it. Do not turn a small failure into a thirty-minute epic.
2. Give one detail only someone who did it would have: the specimen number, the order of magnitude you measured, the temperature or cycle count where it failed. It is the cheapest proof of authenticity there is.
3. State the trade-off you made and what it cost: “I chose a faster loading rate to hit a deadline and paid for not checking the contact”. Owning a cost scores better than claiming there was none.
4. Close with the one habit you kept afterwards and the one question still open. The second matters most: it shows the project is still growing and hands the lecturer somewhere to follow.

**The likely follow-up:** If you did it again, which unanswered question would you tackle first, and why that one?

### Question 9

**Prompt:** Metals and semiconductors both conduct, but heat them up and their resistance moves in opposite directions. Explain why using the simplest picture you can.

**What this tests：** The course wants Chemistry or Physics alongside Maths, so boundary questions like this appear. It tests whether one model can explain two families: in one, heating changes how often carriers are scattered; in the other, it changes how many carriers there are. The interviewer is also listening for whether you say honestly which part you have not been taught.

**How to work through it：**
1. Finish the metal first: plenty of free electrons, and heating makes the positive ions vibrate harder, so electrons are scattered more often, the mean free path shortens and resistance rises. No formula needed; the picture has to be clear.
2. Then the semiconductor: a band gap separates the valence band from the conduction band, so few carriers are available at room temperature. Heating gives electrons energy, the number that crosses rises steeply, and that increase outruns the drop in mobility, so resistance falls.
3. Say which effect wins, because that is the heart of the question: the two families differ in what heating changes, carrier number versus scattering rate, so the signs must differ. Summarise it yourself in one line rather than letting the interviewer do it.
4. Admit the boundary and hand the question back: “doping shifts where the turning point is, and I have only seen that described qualitatively.” That invites a hint, which is worth far more than guessing.

**The likely follow-up:** Aluminium oxide barely conducts: explain with the same band picture, then say whether an ionic crystal has another way to carry current.

### Question 10

**Prompt:** Before the day ends you have 45 minutes with current students and you can ask us things too. What would you ask?

**What this tests：** The day is built to be two-way: the department says the one-to-one is as much for you to judge Imperial as for them to judge you, and thinking of questions to ask is one of the published tips. So this is not a courtesy at the end. It shows how far your thinking has gone: asking something the course page already answers is a different level of preparation from asking how a student actually changed over four years.

**How to work through it：**
1. Open by naming what you will not ask and why: entry requirements and course structure are on the course page and you have read them. That sentence on its own is evidence of preparation.
2. Ask a “how” question rather than a “what”. For example, when a first-year lab result disagrees with the prediction, what is the tutor looking for; or how groups normally split the work.
3. Ask about a tension you genuinely have, and state your own side of it: whether you would rather meet failure analysis early or build the continuum-mechanics mathematics first, and where in the course that choice appears.
4. Keep a closing move for yourself: after the answer, say how it changes your preparation or your decision. That turns the session into a conversation instead of a checklist read-out.

## Questions
**Does Materials Science and Engineering need an admissions test?**

No. The course page states that “this department does not use a test as part of its selection process”, and Materials appears as no-test on the department table; ESAT belongs to the departments listed in our [ESAT guide](/exams/esat).
**Do I have to stay the whole applicant day, and what happens in it?**

Yes. Applicant days run on Wednesdays between November and March from 10.30 to 15.30 and you stay all day: two opening talks, a 30 minute one-to-one, a 45 minute group task and a 45 minute student Q&A. The last two belong to the assessment, not to the waiting time.
**If I am not invited to an applicant day, is that the end of it?**

Not necessarily. Imperial says most departments interview applicants who demonstrate potential, and the course page keeps its wording conditional. No invitation or offer rate is published, so figures of that kind are not official information.
**I know nothing about the front edge of materials. Should I do some reading?**

Yes, but not to look knowledgeable. We push one piece of reading until you can name a question you cannot answer, because an interviewer follows your reading three levels down and the third level is usually what you make of it.
**What is ATAS, and must I sort it out before the interview?**

The course page gives ATAS subject code CAH10-03-02. ATAS is a UK screening exercise applying to some courses and nationalities; it is not extra subject preparation and it does not change what you are asked. Handle it after the invitation using official guidance.

## Official sources
- [Imperial College London: Materials Science and Engineering MEng course page](https://www.imperial.ac.uk/study/courses/undergraduate/materials-science-engineering-meng/)
- [Imperial College London: undergraduate interviews (usually November to February)](https://www.imperial.ac.uk/study/apply/undergraduate/process/interviews/)
- [Imperial College London: which departments use an admissions test](https://www.imperial.ac.uk/study/apply/undergraduate/process/admissions-tests/)
- [University of Oxford: a guide to undergraduate interviews](https://www.ox.ac.uk/admissions/undergraduate/applying/guide-for-applicants/interviews)
- [University of Cambridge: interviews (main period 7 to 18 December 2026)](https://www.undergraduate.study.cam.ac.uk/apply/after/cambridge-interviews)

## Related pages
- [All subject interview questions](https://www.frontiervue.com/interviews/questions)
- [Imperial College London interviews](https://www.frontiervue.com/interviews/imperial-college-london)
- [The Materials applicant day, in full](https://www.frontiervue.com/interviews/imperial-materials-science)
- [Engineering interview questions](https://www.frontiervue.com/interviews/questions/engineering)
- [Physics interview questions](https://www.frontiervue.com/interviews/questions/physics)
- [Chemistry interview questions](https://www.frontiervue.com/interviews/questions/chemistry)
- [Biomedical engineering interview questions](https://www.frontiervue.com/interviews/questions/biomedical-engineering)
- [Thinking aloud in an interview](https://www.frontiervue.com/interviews/guides/interview-thinking-aloud)
