Skip to content

Distinguish stratified selection from quota balancing

Use known frame categories to design selection within strata while keeping sampling probabilities and qualitative review allocations conceptually separate.

Guides

When does balancing categories become a stratified sample?

A research owner may ask for examples from several explicitly stated device operating systems. Choosing convenient cases until each category looks balanced is a quota approach; drawing cases from defined lists within categories is a different design. Clarify the purpose and available frame before using stratified language or making quantitative claims about the resulting selected discussions.

Example research prompt
Plan selection within explicitly stated operating-system strata in the supplied discussion frame. Distinguish purposive contrast from probability estimation and preserve unknown membership.
01 / Guides

Define exhaustive frame strata

Assign eligible records to categories using explicit evidence and a rule for unknown values. Include an unknown stratum when needed. Exclude guessed operating systems based on device appearance or author affiliation. A useful partition must make membership clear enough that the same record is not quietly eligible in several strata under inconsistent rules.

02 / Guides

Choose the selection purpose

Decide whether the task seeks qualitative contrast or frame-level quantitative estimation. The latter requires an appropriate selection design and known inputs, not merely equal counts per category. Do not promise population weighting when frame sizes or inclusion probabilities are unknown, and do not describe a deliberately balanced reading allocation as though it reproduces the natural category mix.

03 / Guides

Preserve selection details

If a probability design is actually used, retain the eligible list sizes, selected identifiers and relevant design information. Otherwise, describe the selection as purposive within categories. Keep unavailable records visible under the documented handling rule rather than replacing them opportunistically with cases that support the emerging conclusion or make every category appear equally complete.

04 / Guides

Deliver design-aware summaries

Present stratum-specific observations and explain whether any combined estimate is justified. If the work is qualitative, focus on mechanism differences rather than unsupported weighted rates. A cleanly separated set of categories can improve research organization without establishing representative inference, so state the design actually performed rather than borrowing the authority of a more demanding statistical label.

Keep the boundaries clear.

  • Balanced category counts do not by themselves establish a stratified probability design.
  • Illustrative methods and platform badges do not constitute live verification or complete coverage.

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