Research methodology
Make every number traceable.
These are the publication standards for the Zvendor statistics collection.
Start with the original source
Prioritize national statistical agencies, government business registers, and primary research with a published methodology. Industry surveys may be included when their sample and limitations are explicit. A reputable publisher alone does not make two statistics comparable.
Keep definitions intact
Record each source’s definition of a small business and its population. Do not equate nonemployer firms, self-employment, sole proprietorships, and solopreneurs. Record the country, unit, denominator, and reference period for every observation.
Separate the dates
The reference period describes when the measurement applies. The source publication date, retrieval date, and last verification date are separate fields. A recent verification does not make an older measurement current.
Explain uncertainty
Carry through sample size, survey coverage, estimates, and source limitations. Do not extrapolate national results from unrelated samples, add overlapping populations, or silently compare incompatible definitions.
Review before publishing
Each record needs an original source link, a locator such as a table or page, and an editorial review. A record identifier stays stable for corrections; increment the revision and explain the change. New reference periods receive new identifiers.
Review schedule
Each published record has a next review date. Updates are editorial and manual at launch. No automated collection or scheduled verification is running yet.
Attribution and reuse
Cite the original publisher, reference period, and source link alongside Zvendor’s record identifier. Source material remains subject to its publisher’s terms. Inclusion does not grant a license to redistribute the underlying report.