Authors must accurately describe data provenance, collection or selection, exclusions, transformations, and analytical procedures. Report deviations from stated protocols where relevant and distinguish planned analyses from exploratory work.

Fabrication, misleading alteration, concealed selective reporting, and misrepresentation of evidence are unacceptable. Legitimate processing must be explained sufficiently to understand its effect on the findings.

• Quantitative research: Describe sampling, variables, instruments, missing-data handling, analytical choices, uncertainty, and relevant software.
• Qualitative research: Explain recruitment or material selection, context, researcher positioning where relevant, analytical development, and the relationship between evidence and interpretation.
• Computational or synthetic-data research: Identify data as simulated or synthetic, explain generation procedures, assumptions, parameters, software versions, and validation. Do not present simulated results as observed population findings.
• Humanities and interpretive research: Identify the corpus, sources, editions, selection rationale, and interpretive framework sufficiently for scholarly scrutiny.

Exact replication is not an appropriate universal expectation for every research tradition. Authors should provide the transparency needed to evaluate claims within the relevant approach.

Preserve research records, relevant code, and supporting documentation in accordance with institutional, funder, legal, consent, and disciplinary requirements. This policy does not impose a universal retention period or require retention contrary to valid participant-protection obligations.

Editors may request proportionate supporting evidence needed to assess reliability. Where direct sharing would breach consent, confidentiality, or other legitimate restrictions, explain those restrictions and discuss a lawful alternative for verification.

Restricted access is not, by itself, evidence of misconduct. An inability to substantiate central claims may nevertheless affect editorial assessment.