Publication Policy

The institute’s own research programme is open access. The report, the technical method annex and aggregate data are released together; publications may be shared and reproduced freely with attribution to IRISP®. The CC BY 4.0 licence applies only to aggregate tables; other parts of a publication and individual-level records do not fall under that licence.

Every study is accompanied by aggregate tables, the questionnaire, the data dictionary and analysis code. Each shared package carries a version number and terms of use; no empty data area is published while no data exist.

For microdata, the access level is stated with the reason after re-identification risk is assessed: open, on application or not shared. Open microdata are coarsened; access on application is for researchers under a use agreement; where risk is high, microdata are not shared. Personal data are never placed in open access at any level.

The finding of a commissioned study remains with the commissioning institution; it is not published under the institute’s name and is not disclosed without that institution’s permission. If the commissioning institution makes the finding public, the technical method annex and the identity of the commissioning institution become accessible as well. The same measurement framework and survey ethics and anonymisation protocol apply; a study requiring a departure from the published protocol, or a result fixed in advance, is not undertaken. Commissioned studies appear in the annual register by number and type; the commissioning institution's name is given only with its consent.

Each research question measures only one concept: perception, attitude, trust, experience or factual estimate; the report states which concept is at issue. Concepts are not forced onto the same true–false scale. A perception gap is calculated only for factual-estimate and perceived-majority questions; no gap is set for attitude or trust. The source, date and scope of the comparison indicator appear in every table.

No figure is published without a technical method annex and a measure of uncertainty. The annex covers target population and scope; sampling frame, method and size; groups not reached; mode and dates of data collection and fieldwork organisation; full questionnaire and translation protocol; pilot study and changes; weighting variables and their source; quality controls; uncertainty assumptions; limitations and cases where comparison is not appropriate; and, if artificial intelligence was used, its task, limits and human oversight.

The colophon of every publication states one of three review labels: internal review, external review or unreviewed. Reviewers are chosen by the scientific board; a member with a tie that could affect a review recuses, and the recusal is recorded. Since the scientific board has not yet been appointed, documents published so far count as unreviewed. Authorship requires a substantive intellectual contribution and responsibility for the published text; contribution roles are stated in the colophon, and ghost or guest authorship is not accepted.

Every publication declares its funding source and any potential conflict of interest. The funding source may not affect question selection, the measurement framework, interpretation of data or the publication decision; a person with a relevant tie recuses from the related publication decision. Research ethics are binding through IRISP/2026/002: informed consent is obtained, records are de-identified at collection, the key is destroyed within ninety days after the close of fieldwork, and withdrawing consent is possible only while the key exists.

Every publication carries a record number in the form IRISP/YYYY/NNN. The title may be translated; the record number does not change and identifies the same publication across language versions. The main version is English, and the three language versions carry the same content. The texts follow the institutional style of each language, and no version omits a finding, condition or responsibility record. A translated title, short title or page placement does not replace the number.

When a material error is found, a dated correction note and a new version number enter the colophon. The correction explains the nature of the error, the affected section and the effect of the change on the result; spelling and formatting changes are marked this way only when the finding is unaffected. The previous version remains accessible, and the new version does not remove the earlier text without explanation.

If an error invalidates a finding and cannot be resolved by a correction note, the publication is retracted. Retraction does not erase the record: title, record number, version history and reason remain accessible. The reason states why the error leaves the publication unable to support its result. A retracted publication is not presented as a new finding, is not removed from the public text without explanation, and its relation to any later corrective publication is shown on the record.

Any person may submit a complaint or appeal concerning an error, missing disclosure, conflict of interest or alleged publication-ethics breach at info@irisp.org. The report is matched to the relevant publication record, classified as technical, editorial, ethical or data-protection related, and referred to the responsible body for review. The applicant receives a written response; the outcome is recorded as correction, version change, retraction or reasoned rejection.

Citation is given in this form: IRISP (2026). The IRISP® Measurement Framework. Methods note, IRISP/2026/005.