Chapter 19 — Quiz

Twenty-five questions to check that the interview setup, the audio safety, the question craft, and the listening discipline are yours. Mix of multiple choice, true/false (with justification), short answer, and "what's wrong here?" perception questions. Answers and one-line explanations are in the Answer Key at the end — try each before you look.

Multiple choice

1. In a standard nonfiction interview, the subject should look:

  • A) Directly into the lens
  • B) Just off the lens, at an interviewer beside the camera
  • C) Down at their hands
  • D) At a mark on the wall behind the camera

2. The interviewer should sit as close to the lens as possible mainly because:

  • A) It's easier to hear the subject
  • B) It keeps the off-axis eyeline angle small, so the gaze reads as near-eye-contact rather than lost profile
  • C) It improves the audio
  • D) It lets you operate the camera at the same time

3. "Two-mic safety" means:

  • A) Recording with two identical lavs
  • B) Using a mic with two capsules
  • C) Recording the subject on two independent mics (e.g., lav + boom) on separate channels, so either can fail without losing the take
  • D) Having a second mic in your bag as a spare

4. For a warm documentary look, a good key-to-fill contrast ratio is around:

  • A) 1:1
  • B) 2:1 to 3:1
  • C) 8:1
  • D) 16:1

5. Which is an open question?

  • A) "Did the accident change you?"
  • B) "Were you scared?"
  • C) "Walk me through what happened that night."
  • D) "You loved the job, right?"

6. A subject answers "Twenty years" to "How long have you done this?" This is a problem because:

  • A) It's factually wrong
  • B) In the edit the question is cut out, so the fragment answers a question no one heard
  • C) Twenty years is too long to be believable
  • D) It's a leading answer

7. Dialogue peaks in an interview should sit around:

  • A) 0 dBFS
  • B) -12 dBFS, with headroom below 0
  • C) -40 dBFS
  • D) +6 dBFS

8. The main reason to key an interview from the side the subject is facing is that:

  • A) It uses less power
  • B) The light "leads" the eyeline and reads as motivated, dropping the shadow onto the camera-near side
  • C) It's the only correct way to light a face
  • D) It prevents lens flare

9. In a two-camera interview, both cameras should be:

  • A) On opposite sides of the subject's eyeline
  • B) On the same side of the line, offset by at least 30° and ideally a shot size
  • C) At exactly the same angle and size
  • D) One in front, one directly behind the subject

10. The single most common way to hide an internal cut in a single-camera interview is:

  • A) A dissolve
  • B) Laying B-roll over the cut while the interview audio continues underneath
  • C) A jump cut
  • D) Slowing the footage down

11. The best closing question in an interview is often:

  • A) "Are we done?"
  • B) "Can you summarize everything in one sentence?"
  • C) "Is there anything I didn't ask that you think I should have?"
  • D) "Do you have any questions for me?"

12. A "noddy" (a cutaway of the interviewer nodding, shot after the subject leaves) is:

  • A) Always dishonest and never used
  • B) A legitimate editing bridge, but dishonest if used to fabricate a reaction that implies something false
  • C) The main source of an interview's audio
  • D) A type of microphone

True / False — justify your answer in one sentence

13. A great color grade and sound mix in post can rescue an interview that was badly framed, badly lit, and recorded on one failed mic. True / False?

14. Keying an interview from the side opposite the one the subject faces is always wrong. True / False?

15. Holding two or three seconds of silence after an answer usually produces a truer, more valuable second answer. True / False?

16. In an interview, the interviewer should say "mm-hmm" and "right" during the subject's answers to show they're listening. True / False?

17. A window behind the subject is a good source of key light for an interview. True / False?

18. With a matched, synced, two-camera setup you can tighten an answer invisibly without any B-roll. True / False?

Short answer

19. Define the interview setup in one sentence, and name three of the elements it arranges.

20. What is the non-answer? Give two examples of the forms it takes.

21. Explain active listening and name the one technique from it that most reliably produces the best material.

22. Your subject gives a brilliant thought but tangled as a fragment. What exactly do you say and do to get a usable version — without faking anything?

What's wrong with this shot / this edit?

23. A described interview: the subject sits dead-center in the frame, framed from the waist up, staring straight into the lens with a flat, even light on both cheeks and a wall six inches behind their head. It's for a documentary about grief. Name three things working against the intended feeling and the fix for each.

24. A described audio setup: one lav, clipped over a wool sweater, going into the camera; levels set once at the start and never checked; the operator watches the meters but wears no headphones; no room tone recorded. List the risks and the fixes.

25. A described edit: a single-camera interview has been tightened by cutting three times within one continuous answer, on the same locked medium shot, with nothing laid over the cuts. Describe what the viewer sees, name the error, and give two ways it should have been prevented on set.

Answer Key

Reveal answers **1. B.** Off-axis at an interviewer beside the lens — the candid "conversation" register of documentary. (Into the lens = direct address; see §19.1.) **2. B.** Sitting close keeps the off-axis angle small so the subject's near eye reads full and the gaze feels almost like eye contact, not a lost three-quarter profile. **3. C.** Two independent mics on separate channels so a single failure never loses the take. (Not a spare in the bag — both run *simultaneously*.) **4. B.** 2:1 to 3:1 gives the warm, dimensional documentary look; ~1:1 is corporate-clean; 4:1+ is weighty/uneasy. **5. C.** "Walk me through what happened that night" — open and specific. A, B, and D are closed and/or leading. **6. B.** The question is cut in the edit, so "Twenty years" answers an unheard question; you need "I've been doing this for twenty years." **7. B.** ~-12 dBFS with headroom below 0 dBFS so a sudden laugh or emphasis doesn't clip (clipping is irreparable). **8. B.** The light leads the eyeline and reads as motivated (by the interviewer/window), with natural modeling as the shadow falls camera-near. **9. B.** Same side of the line, ≥30° apart plus a size step, so the A↔B cut reads as a shot change, not a jump. **10. B.** B-roll over the cut with the interview audio continuing underneath (an L-cut) — the single most common solution and the reason B-roll exists. **11. C.** "Anything I didn't ask?" hands the subject the floor and routinely produces the best answer of the day. **12. B.** A legitimate bridge, but dishonest if used to fabricate a reaction implying something false — an ethics line, not a technical one. **13. False.** Post can only arrange what was captured; a badly framed, badly lit take on a failed mic has no clean image or usable audio to rescue — *fix it in pre/production, not in post*. **14. False.** It's a legitimate, deliberate mood choice ("sinister"/weighty); it's just not the flattering default. The rule is *motivate the choice*, not "never do it." **15. True.** The first answer is usually the prepared one; the unrehearsed, truer line tends to arrive after the pause — if you don't fill it. **16. False.** Verbal encouragements land on top of the subject's audio and can make a perfect soundbite uncuttable; nod and show you're listening silently instead. **17. False.** A bright window behind the subject throws the face into silhouette (or blows out the window); put the brightest source *in front*, on the looking side. **18. True.** Cutting from A-cam to B-cam on the trim hides the seam, so no cutaway is needed — the payoff of shooting two matched angles. **19.** The interview setup is the repeatable physical arrangement of subject, camera, interviewer, light, and microphone — framing, eyeline, and interviewer position — that you build before the conversation starts. Three elements: e.g., framing/thirds, eye-height camera, off-axis eyeline (also: key light, mic placement, background separation). **20.** The non-answer is a reply that can't stand alone in the edit. Forms: a yes/no, a fragment, an echo of the question, an "it depends," or an evasion. **21.** Active listening is genuinely attending to the current answer (not rehearsing your next question) so you can follow the thread, ask the right follow-up, and use silence. The most reliable technique: holding the pause after an answer so the subject fills it with the truer line. **22.** Warmly say something like, "That was perfect — can you give it to me one more time as a single, complete sentence?" You're not faking; you're getting a clean, cuttable take of a true thing (and it lets the answer stand alone once the question is cut). **23.** (1) Dead-center framing feels like a mugshot — put them on a third with looking room. (2) Direct-to-lens address is the wrong register for observed grief — use an off-axis eyeline so we observe a conversation. (3) Flat even light + a wall pressed to the head is dimensionless and hostage-video-flat, wrong for grief — key from the looking side at a deeper ratio and pull them off the wall for soft, separated, moodier light. **24.** Risks: the lav will rustle against wool (clothing noise), single point of failure (no backup), an unmonitored take can be ruined without you knowing, unchecked levels can drift or clip, and no room tone means no way to patch edits. Fixes: dress/isolate the lav (and prefer a smoother spot), run a second independent mic (two-mic safety), monitor on headphones the whole take, set levels with headroom (~-12 dBFS) and watch them, and record 30 s of room tone. **25.** The viewer sees the subject's head visibly "jump" at each cut — a jump cut — because near-identical frames butt together on the same locked shot. Prevention on set: (1) shoot enough B-roll (Ch.20) to lay over each internal cut, or (2) shoot two cameras (or reframe between questions past 30°, or punch in from 4K) so there's a genuinely different angle/size to cut to.

Topics to review by question

  • Framing & eyeline (§19.1): 1, 2, 19, 23
  • Lighting (§19.2): 4, 8, 14, 17, 23
  • Audio & two-mic safety (§19.3): 3, 7, 24
  • Question design & the non-answer (§19.4): 5, 11, 20, 22
  • Listening & the full-sentence answer (§19.5): 6, 15, 16, 21, 22, 29-style thinking
  • One vs. two cameras / the edit (§19.6): 9, 10, 12, 18, 25
  • Synthesis / "fix it in pre": 13, 23, 24, 25