An average is useful shorthand, but a film can make it look like every specimen produced the same result. When variation matters to a supplier's test account, show enough of the spread to preserve the meaning of the summary. The buyer should decide this before a single oversized number becomes the visual centrepiece.
NIST's discussion of process variability distinguishes individual observations and their spread. This supports a basic editorial choice: a central value and the variation around it answer different questions. The production should use the client's approved summary rather than calculate or select one independently.
Ask what the average is summarising
Identify the specimen group, measured quantity and relevant conditions. A number drawn from one group should not appear over footage of an unrelated product range. If several batches or test conditions are combined, the specialist should decide whether that combination is meaningful for the public account.
The film need not display every laboratory record. A small set of identified points or an approved range graphic may be enough. The form should preserve what the evidence actually reports, not replace it with a familiar bell curve.
Avoid inventing a smooth distribution when the source only provides a few observations. A polished statistical shape can imply more knowledge than the available data supports.
Give graphic marks precise meanings
Error bars, shaded bands and ranges are not interchangeable. Ask the technical reviewer what a proposed band represents and how it should be named. If the answer is unclear, choose a simpler display rather than leaving the audience to guess.
Keep the scale consistent across comparisons. A narrow crop around one result and a wide scale around another can change the apparent variability. The design should not create a ranking that disappears when the figures share a common basis.
Do not call specimen spread measurement uncertainty unless that is the documented meaning. The two may appear in the same report, but they should not be merged into one vague “plus or minus” caption.
A hypothetical supplier graph
Imagine a manufacturer presenting results from a set of material specimens. The proposed film shows the average as a fixed value beside a row of identical product icons. Viewers interpret it as a guarantee for every delivered item.
The buyer revises the sequence to identify the tested group and show an approved representation of its variation. The average remains, but no longer substitutes for each observation. This fictional example concerns honest summarisation, not statistical certification.
During review, ask what the viewer believes an individual specimen could have produced. If the film has erased the distinction between a summary and an item, its visual grammar needs attention.
The commission should communicate the supplied evidence clearly. Sampling design, statistical inference and acceptance decisions remain with the client's specialists, and the film should not manufacture certainty by removing visible variation.
Discuss the film
Contact Live Images to discuss whether this approach suits your brief and what production scope would be appropriate.
Prepared with AI assistance for Live Images. Planning guidance is general; suitability and production scope must be agreed for your project.