Skip to content

Video Replay Analysis

Principle · All principles
Evidence · 25 claims (15 for, 10 mixed) · 28 studies (10 qualitative, 4 quant-synthesis, 4 review, 3 causal, 3 theoretical, 2 associational, 2 design), q1–q4 · 4 of 28 report an effect size · 21 claims rest on one study

Conditional relationship

This page owns reviewing a recording of a performance to change the next one: a teacher or teacher candidate watching their own lesson, a trainee clinician reviewing a recorded consultation, a learner replaying their own presentation, interview or physical skill. For a learner who is building a complex-skill performed in real time (teaching, consulting, presenting, a motor or technical routine), usually an adult in the postsecondary or adult-workplace band, and whose starting response is that they cannot say accurately what they did at the moments that mattered, reviewing a recording against one stated focus, with pauses at chosen moments, an explanation of each moment (what happened, why, what else was possible), a debrief in which someone else asks before they tell, and one change carried into a recorded next attempt is expected to improve what the learner notices and can explain about their own performance (conceptual-understanding of its cues and decisions) and the next comparable performance (skill-performance, near-transfer). Rewatching without a focus or a follow-up attempt is expected to give little: a novice watching a complex scene sees less than an expert does, and a recording on its own changes nothing until something is tried differently.

No claim in the wiki tests this relationship as stated. No recorded study compares reviewing one's own recording with an alternative of matched time (a debrief from memory, an observer's written notes, practice without review), or compares focused with unfocused replay. What the wiki holds is: teachers' reports, in one qualitative study of web-mediated coaching, that watching their own lessons was among the most powerful parts of coaching; a review's second-hand report that, in a thirteen-school coaching study in England, positive outcomes were mediated by using video to analyse teacher–student interactions; one design-based study in which preschool teacher candidates' observation assignments scored higher after workshops on authentic classroom video than on a virtual simulation, with analytic gains levelling off after the third workshop; and the general findings on self-explanation prompts and contingent support that this page carries to replay. Two findings limit it: in a meta-analysis of coaching, programmes that supplied a video library were associated with smaller effects on instruction, and novice teachers watching classroom video monitored less and made contradictory observations where experts did not.

How the converted neighbours sit beside this one. Observation and Shadowing owns watching someone else's practice, live or recorded, as a newcomer; this page uses that page's model for the early sessions in which learners analyse another person's recording before their own, and does not repeat it. Reflection owns examining one's own performance against a criterion and committing to a change, with or without a recording; this page adds what the recording changes (exact moments instead of recall, pausing, a second viewing of the next attempt). Debriefing owns the facilitated conversation after a shared experience; the conversation after a replay is one such debrief. Self-monitoring owns checks of one's own work during a task; replay is a check made afterwards, on a record. The Professional Development pattern builds coached cycles for teachers in which the teacher films a lesson and, in late cycles, chooses the segment; this page is the principle behind that step.

Default design, while the relationship is untested

The earlier page's guidance (a clear analytic focus, pause-and-discuss, a feedback or debrief structure, psychological safety, others' video before one's own), kept as a concrete default a designer can act on and revise. Each step is a design proposal with its evidence status beside it.

  1. Set one focus before anything is recorded. Name one to three things to look for, tied to what the learner is trying to improve (wait time after questions, the opening two minutes of a consultation, eye contact during a talk), and how each will show on screen. Basis: learners notice gaps and adjust when they compare performance with a stated criterion, Self-regulated learning models treat self-monitoring as a core process through which learners notice gaps and adjust strategy [+M] (a synthesis and a theoretical account, not a replay study). Where the brief already has criteria, a rubric or an observation tool, the focus comes from it.
  2. Record a short, real performance and cut it to the moments that matter. Record the whole performance if possible but review two or three clips of 2–5 minutes each, chosen for the focus. Untested proposal; the duration is a proposal. Coded data from a recording can show a pattern the performer did not notice: a first-grade teacher found from a coded lesson video that she had called on fewer than half her students, EQUIP data reports surfaced participation inequities that teachers were unaware of and motivated instructional change [+W] (one case report).
  3. Start with someone else's recording. In the first one or two sessions, learners analyse a peer's or an anonymous practitioner's clip with the same focus, before their own. Untested in the wiki as a sequence; from the earlier page and its source (Nagro 2020, which has no claim page). Analysing authentic classroom video built teacher candidates' analysis and support scores more than a simulation did, in a fixed order, Authentic video outperformed virtual simulation on the higher-order analysis and support dimensions of observation assignments [+M]; how to run these sessions is modelled in Observation and Shadowing.
  4. Watch once through, then pause at moments tied to the focus. On the second viewing the learner stops at each moment, time-stamps it, and writes what happened (what was said or done, what the other people did) apart from what it means. Untested proposal. A novice sees less in a busy scene, so the focus and the time-stamps do the selecting: Expert teachers interpret complex classroom phenomena more accurately and fluidly than novices and postulants [~M] (one video study reported in a review).
  5. Explain each paused moment. For each moment: what was I trying to do, what happened, why do I think it happened, what else could I have done. Basis: Prompting learners to self-explain improves understanding and problem solving on immediate tests by a moderate average amount, with little evidence yet on delayed or classroom gains. [~S], carried from studying examples and texts to one's own performance, which none of its studies tested.
  6. Debrief with someone who asks before telling. A coach, mentor or peer opens with questions about the learner's own moments, listens, then adds what the learner did not see and one suggestion. Keep the learner talking for at least half the time (a proposal). Basis: support that responds to the learner's own response beat fixed or no support in one-to-one tutoring, Contingent scaffolding improves learning more than fixed or absent support. [~M] (children, long division; carried to coaching); in one practicum case a mentor's routine of open questions, listening and suggestions supported the mentee's reflection, A structured feedback-and-reflection cycle (pre-lesson plan review, immediate verbal comments, written feedback, take-up time, and open-ended questioning) supported the mentee's reflective development [+W], while in the same case the mentor out-talked the mentee in every recorded dialogue, Talk-time analysis showed the mentor spoke more than the mentee in video-recorded dialogues, with classroom management talk ranging from 3% to 39% of total talk time [~W]. Coaching in which video was used to analyse interactions is reported as the route to positive outcomes, Video-based analysis of teacher-student interactions mediates positive coaching outcomes, and challenging teachers supports reflection but is uncommon [+W] (second-hand, no effect size); the same report says challenging the teacher supported reflection but was uncommon, so the debrief should name a gap, not only praise.
  7. End with one change and record the next attempt. The learner names one change to try, tries it in the next comparable performance, records it, and compares the two clips at the same moments. Untested proposal. The account offered for why self-review motivates change is the gap a person sees between their aims and what the tape shows, Perceived discrepancy between actual teaching performance and goals motivates teachers to change their teaching [+W] (a 1970 programme's theoretical argument, untested); a recording without a next attempt leaves the gap unused.
  8. Keep it safe to be seen. The learner owns their recording: they decide who watches it, it is deleted when the cycle ends unless they choose to keep it, and everyone else in frame (pupils, patients, clients) has consented or is cropped out. Critique is about moments on screen, not the person. Untested; from the earlier page. Teachers in one coaching study described moving past initial self-consciousness to find watching their own lessons among the most useful parts of coaching, Video-based self-observation of teaching was among the most powerful parts of the coaching process for teachers [+W] (self-report).
  9. Plan the arc across sessions and hand over the choices. A default for a placement, term or coaching programme of five to eight cycles, one every one to two weeks (proposals): sessions 1–2, another person's clips, focus set by the coach, group or pair analysis, 30–45 minutes; sessions 3–4, the learner's own clips, focus agreed with the coach, coach chooses the moments; sessions 5 onward, the learner chooses the focus and the clip and leads the analysis, and the coach only asks and adds. Within a 40-minute session: 5 minutes to restate the focus, 10 minutes of viewing, 15 minutes of paused analysis, 10 minutes of debrief and one change. Move the learner on to choosing their own clip once, in two consecutive cycles, the moments they pick and explain match those a coach would pick (a proposed criterion). In the one sequential study, teacher candidates' analysis scores rose between the second and third authentic-video workshops and then levelled off, Analytical competence showed a leap-plateau trajectory across sequential authentic-video workshops, improving mainly between Workshops 2 and 3 [+M], while their suggestions for supporting children kept improving to the fourth, Supportive competence grew continuously through Workshop 4 while basic observation ability remained stable across sequential AV workshops [+M]: change the focus or the task after three or four sessions on the same one, rather than adding more of the same.

When a condition is not met: if no one can debrief, give the learner a written prompt card for steps 4, 5 and 7 and a worked example of a completed analysis; if recording is not allowed, use a short audio recording or an observer's time-stamped notes and say that the evidence for video does not transfer to them; if the learner already notices and explains their moments accurately, skip the early sessions on others' clips.

Fitting the design to a situation

The three facts that most change the decision:

  • How accurately the learner already sees their own performance. A novice needs a narrow focus, others' clips first and a coach choosing the moments; a practitioner who already names the right moments needs only the recording, their own focus and a check. If unstated, ask: after one viewing of a two-minute clip, which moments does the learner pick out, and would an experienced colleague pick the same ones?
  • Who will see the recording, and who is in it. Self-review, review with one trusted coach, and review in a group are different levels of exposure, and other people on screen need consent. If unstated, ask: who may watch this clip, where is it kept, and has everyone in it agreed?
  • What the next performance is, and when. Replay changes performance only through the next attempt. If unstated, ask: when does the learner next do this task, and can that attempt be recorded too?
If the brief says… Then change… Basis
Learners are novice (teacher candidates, first-year trainees) Run sessions 1–2 on others' clips with a coach-set focus of one or two items; time-stamp and describe before interpreting; the coach picks the moments in early own-clip sessions Expert teachers interpret complex classroom phenomena more accurately and fluidly than novices and postulants [~M]; Authentic video outperformed virtual simulation on the higher-order analysis and support dimensions of observation assignments [+M] (preschool teacher candidates watching others' video, not their own)
Learners are advanced (experienced practitioners) Skip others' clips; let the practitioner choose the focus and clip from the first session; the coach asks and adds only what was missed untested proposal; the coaching reports below are from in-service teachers (Video-based self-observation of teaching was among the most powerful parts of the coaching process for teachers [+W])
Learners are child (primary age) Use very short clips (under two minutes) of the children's own work or play; ask what they notice in their own words before any focus; an adult runs the device and keeps the recording Young children's video interpretation of their own mathematical play reveals both mathematical and social-affective sensemaking [+W] (four children, no comparison; shows what they notice, not a learning gain)
Learners have low confidence, performance anxiety or low literacy Start with others' clips; review the learner's own clip privately first, then with one trusted person; open with a moment that went well; allow spoken or pointed-to analysis instead of written notes untested; from the earlier page (self-video "can feel exposing"); teachers reported getting past initial self-consciousness (Video-based self-observation of teaching was among the most powerful parts of the coaching process for teachers [+W])
Adults in work settings (clinical, customer-facing, trainers) Record a real or simulated work task (a consultation, a call, a handover); take the focus from the workplace's own standard; keep reviews in protected time, 20–30 minutes, and give the recording's owner the right to delete it untested proposal; medical students valued recorded review of simulated consultations but asked for clear feedback criteria (Learners valued repeatable practice and recorded review in MedEasy while questioning missing actions and feedback criteria [+W], 12 students, interviews)
The goal is a complex-skill performance the brief wants kept whole (a full lesson, a whole consultation, a complete routine) Record and keep the whole performance; choose clips only for analysis, and end each cycle by reviewing one clip in its place in the whole (what came before and after it) untested proposal
The goal is understanding or noticing (conceptual-understanding), not changing one's own performance Use others' or shared clips only; pause and ask for explanations of each moment; no own recording is needed Prompting learners to self-explain improves understanding and problem solving on immediate tests by a moderate average amount, with little evidence yet on delayed or classroom gains. [~S] (carried from texts and examples); see Observation and Shadowing
Mixed modes: some learners remote, some in the room, or phone-only Share clips by a link with time-stamped comments; each learner reviews on their own device before a short joint debrief; on a phone, keep clips under two minutes and the focus to one item untested proposal
One coach or teacher for many learners, or scarce facilitator time (volunteers, a busy shift, one check-in) Learners review their own clip alone or in pairs with a prompt card (focus, time-stamps, three questions, one change); the coach reads only the one-change line and the chosen moment, for 5 minutes per learner, and sees one clip a cycle untested proposal; the contingent-support evidence is one-to-one (Contingent scaffolding improves learning more than fixed or absent support. [~M])
A single session only Record and review in the same session: one 2–3 minute performance, one focus, 5 minutes of paused review, one change, a second short attempt recorded and compared; do not claim lasting change untested proposal
A fixed short time budget (e.g. 20 minutes) 2 minutes focus, 4 minutes performance, 6 minutes paused review, 4 minutes second attempt, 4 minutes comparison and one change untested proposal
The brief names its own observation tool, rubric, checklist or debrief routine Keep it as the focus and the debrief; change only what this page's evidence says matters: pause at moments, ask the learner to explain each one before telling, end with one change and a recorded next attempt untested; coded data from a named tool did reveal patterns the teacher had not seen (EQUIP data reports surfaced participation inequities that teachers were unaware of and motivated instructional change [+W])
Private or sensitive material (a teacher's own lesson, patients or minors on screen, a learner's struggling attempt) The learner controls who sees the recording and the notes; review one-to-one or alone; crop or blur others, record only with consent, store locally and delete at the end of the cycle; no clip is shown to a group without the owner's agreement untested; from the earlier page (privacy and storage)
The performance is for a real audience (a public talk, a client, a performance) Record a rehearsal, not the event, for detailed review; after the event, review only with the performer's agreement and focus on one moment for next time untested proposal
High stakes (a licensing assessment, a graded placement) Keep formative replay apart from the assessed recording; the learner chooses what the assessor sees; judge readiness on an unreviewed performance untested proposal
The programme supplies a library of model videos and little else Use the library only for sessions 1–2; tie each library clip to the learner's own focus and move to the learner's own recordings with a debrief Pairing coaching with group trainings is associated with larger effects on instruction and achievement; instructional materials with larger, and video libraries with smaller, effects on instruction [~M] (between-study association in a coaching meta-analysis, not a causal contrast)
Recording cannot be made (policy, no device, no consent) Use an observer's time-stamped notes of what was said and done, or an audio recording, with the same focus, explanation and one change untested; the video findings above do not transfer to notes

Nine of the seventeen rows cite a claim, and none of those claims tests reviewing one's own recording against an alternative. Two carry a finding from watching others' video (teacher candidates) or from self-explanation of texts to this setting and say so; five rest on self-reports, four children or a single case; the coaching-library row rests on a between-study association; the one-coach row cites a claim only for its limit (one-to-one evidence). The other eight rows are untested proposals.

Observation, state and explanation

What can be observed is what the learner does with a recording under stated conditions: which moments they stop at, what they write at each (description, interpretation, alternative), how long they talk in the debrief, the one change they name, and what they do in the next recorded attempt at the same moment. "Sees their own practice accurately", "is defensive" and "has changed their practice" are inferred states. A fluent analysis of a clip shows what the learner can say about a recording; it does not show what they will do in the next performance.

Response Plausible explanations Observation that could change the interpretation
Comments only on appearance, voice or mannerisms Self-consciousness on first viewing; no focus given; the learner cannot yet see the decisions in the scene Give a clip of someone else with the same focus; if the comments move to decisions there, the issue is exposure, not seeing.
Picks moments a coach would not pick A different, legitimate aim; the focus was unclear; a novice's narrower view of the scene Ask what they were trying to do at their moment; show a coach's moment and ask what they see in it.
Names a change but does the same thing in the next recording The change was not practicable in the moment; the cue that should trigger it was not recognised; the change was not the learner's own Ask what happened just before the moment in the new clip; rehearse the change once on its cue before the next attempt.
Describes the clip accurately but cannot explain why things happened Description and interpretation are separate skills; little knowledge of the domain to explain with Prompt with two possible explanations and ask which the clip supports, and what would distinguish them.
Declines to record or to show a clip Fear of judgement; a real privacy or consent problem; no value seen in it Ask which of these it is; offer private self-review, audio only, or others' clips first.
Analysis improves over sessions, the performance does not Clip analysis is being practised, not the performance; too few attempts between reviews Count the attempts between reviews; check whether the next performance is recorded at the same moment.

These rows are proposals, untested with learners as diagnostics. A review is itself an intervention and changes what it measures, so record each one, and keep the explanations that remain open.

Evidence and qualifications

No claim in the wiki compares reviewing one's own recording with another way of reflecting on the same performance, or varies the focus, the number of reviews or who leads the debrief. The claims below test neighbouring parts of the relationship.

Learners, tasks and outcomes differ across these claims (in-service teachers' reports, teacher candidates' group assignments, children tutored in arithmetic, learners studying texts), so do not rank them by effect label. None measures a change in the learner's own later performance after reviewing their own recording.

The evidence by question

Further evidence, not yet read against this model

Claims this page cited before it was rewritten, or found in the search, that bear on it only at a distance.

Objective and learner-valued goal

The designer's objective is usually a change in a specific part of the next comparable performance (skill-performance), with the learner able to say why. Elicit separately what the learner wants from watching themselves: to stop a habit they have been told about, to see whether a new technique works, to prepare for an assessed observation, or to get through the requirement with as little exposure as possible. Agreement and divergence are both observations. A teacher candidate who values a good placement grade may pick flattering clips; a practitioner who values a particular class going better may choose a focus the designer did not plan, and that focus is usually the one that will be acted on.

For example, a new nurse's programme may set "closed-loop communication during handover" as the focus, while the nurse wants to stop being interrupted by a senior colleague. Review one handover clip for each: the designer's focus tests the programme's objective, the nurse's tests what they came for, and both are recorded so that one is not reported as the other.

What would revise this model?

The expectation should weaken if studies comparing review of one's own recording with a matched-time debrief from memory or an observer's notes, with comparable learners and an outcome measured on the next real performance, show no advantage; if focused replay does no better than unfocused rewatching; or if gains in clip analysis do not appear in later performance. If others' clips alone produce the same change as one's own, the self-review steps are a preference, not a requirement. If a debrief led by the learner does no better than one led by the coach, drop the "ask before telling" step. If the analytic plateau after three or four sessions does not recur in other populations, drop the proposal to change the focus at that point. Do not protect the model by calling every failure "the wrong focus" or "not enough psychological safety" after the fact.

Noticing in a recording, explaining it, the next performance and lasting change in practice are separate claims. The present evidence establishes no number of review cycles, no clip length and no criterion for handing the choice of clip to the learner.

Claims

  • Purposeful Reflection — replay provides concrete evidence for reflection instead of relying on memory alone.
  • Immediate Feedback — replay can make feedback more specific and discussable.
  • Self-monitoring — video helps learners compare what they intended with what they actually did.
  • Multimodal Instruction — replay adds a visual and auditory record that learners can revisit.
  • Observation and Shadowing — owns watching someone else's practice, live or recorded; the early sessions here on others' clips follow it.
  • Debriefing — owns the facilitated conversation after an experience; the conversation after a replay is one.

Examples

Key Sources

Page metadata
Field Value
type principle
id video-replay-analysis
title Video Replay Analysis
description For a learner building a real-time performance skill who cannot yet say accurately what they did at the moments that mattered, reviewing a recording against one stated focus, explaining paused moments, debriefing with someone who asks before telling and carrying one change into a recorded next attempt is expected to improve what they notice and their next performance, though no claim tests reviewing one's own recording against an alternative; the evidence is coaching reports, one study of teacher candidates analysing others' video, and findings carried from self-explanation and contingent support.
status review
canonical True
generated by claude/unspecified on 2026-10-15
trust unverified — no one has confirmed this page's content
sources 4 — baecher-2015, dovigo-2020, nagro-2020, tripp-2012

Frontmatter is the page's machine-readable record — see the Schema & Guide.