See a study in practice
One turn, from recording to reported figure.
This is the whole discipline of DebateLens in six moves. The turn below is an illustrative example written for this page, not a real researcher’s private study; the structure, fields, review states and calculation rules are exactly those the workspace uses.
- 01
The recording
A televised debate is added to a study as video, audio, or an existing transcript. The file stays private to the account, and every later record points back to a timecode in it.
Computed from your material
- 02
The attributable turn
The recording is segmented into turns and each turn is attributed to a speaker. Turn 47 — Speaker A, 00:41:12–00:41:38: “I want to come back to what my opponent just said about the deficit, because the figure he used is not the figure in the budget.”
Draft until you confirm the attribution
- 03
The observable evidence
Evidence is what a second researcher could check in the recording: the speaker turns to face the opponent, holds eye-to-eye gaze for 19 of the turn's 26 seconds, and speech remains fluent throughout, with one phrase pause before “not the figure”.
- 04
The coded observation
Domain II, Pragmatics — breach of the maxim of quantity attributed to the opponent, with the interpretation recorded in its own field, separate from the evidence above. Confidence: 0.72. Review state: draft.
Excluded from all figures while it is a draft
- 05
The reviewed calculation
Once accepted or qualified, the record joins the counted corpus: it adds one link to the repetition matrix, 19 seconds to Speaker A's eye-to-eye gaze total, and one entry to the maxim tally. Every denominator is shown beside every proportion.
Counts only reviewed records
- 06
The report and its caveat
The report states its scope: how many turns were coded, how much of the recording carried gaze data, which records were excluded, and where the evidence is thin. The comparative interpretation reports which participant's discourse is denser in the framework evidence counted for this corpus. It is not a statement of who won the debate, a ranking of candidates, or a prediction of an election.
What is automatic, and what needs your review
- Transcription and segmentation — draft material
- Visible immediately so you can correct it. Not yet a research finding.
- Automatic coding — draft observations
- Clearly labelled as drafts and excluded from every computed figure until reviewed.
- Your review — approved or qualified
- Accepted records enter the matrices, gaze proportions and tallies. Figures recalculate as you review.
- Report — traceable output
- States its coverage, its caveats and what it excluded, and links each figure back to the records behind it.