Skip to content

Use ordered categories without pretending equal distances

Define an ordinal research scale through observable distinctions and avoid arithmetic that treats unequal category steps as measured intervals. Keep unknowns separate.

Guides

Can qualitative obstacle severity be summarized as an average?

A review may classify public task accounts as minor interruption, workaround required or task blocked. These states have an order, but the distance between them is not necessarily equal. Treating them as one, two and three and averaging can imply a measurement precision the evidence does not provide about the actual burden experienced by authors.

Example research prompt
Classify supplied task accounts as interruption, workaround or blocked using explicit outcomes. Preserve unknowns and report category counts without averaging arbitrary severity numbers.
01 / Guides

Define observable levels

Use explicit task consequences rather than emotional wording to distinguish levels. Include a blocked task only when the author reports inability to complete it under the described conditions. Exclude unknown outcomes from the ordered scale; an account that stops before completion does not automatically belong at either the mild or severe end.

02 / Guides

Test boundary cases

Compare examples where a workaround is described but completion remains unclear, or where a task succeeds after a short interruption. Write rules for these cases and preserve unresolved assignments when necessary. Do not infer severity from the author’s tone or a post’s popularity, since both can differ independently from the actual task consequence.

03 / Guides

Summarize distributions

Show exact reviewed counts in each supported category and retain unknown outcomes separately. If comparison across groups is useful, present the category distribution with its eligibility rules rather than a fabricated mean severity. A more complex analysis would require additional assumptions that should not be smuggled into a simple qualitative review table.

04 / Guides

Deliver scale limitations

Provide the ordered definitions, example evidence and the reason interval arithmetic was not used. Explain that the scale describes reported task consequences within the corpus, not calibrated harm or customer value. If the owner needs a measured burden, recommend an appropriate study instead of converting category labels into an unsupported numerical impact score.

Keep the boundaries clear.

  • Ordered categories establish rank, not equal numerical distance.
  • Badges identify a research context; they do not promise access or complete evidence.

Start with a read-only check: ask your agent to report actual source coverage before collecting records or exporting an audience.

Primary references

Documentation reviewed 2026-11-13. Product capabilities, client availability and plans can change. Third-party marks identify their products, not an affiliation.

Know what social media shows about you.
Start with a question.

Get started
montage