Begin with an accountable standard
Integrity reaches far beyond avoiding plagiarism. It joins reliability, honesty, respect and accountability across the entire research process. Current university policies and disciplinary standards remain authoritative; these guides help turn them into concrete steps and records. Start with the audit for a new project or a broad pre-submission review.
Protect integrity across the workflow
Many problems arise at transitions rather than in one dramatic act: a result becomes a source, an analysis becomes a claim, or external feedback enters the submitted prose. These guides offer distinct records for those moments. They preserve when a decision was made, why it was defensible and which evidence supports it.
Data, documentation and complete results
Traceability needs an unchanged starting point, intelligible versions and a route to the reported finding. It does not require exposing protected data publicly. These method-aware guides combine evidence with privacy, access and local retention obligations. Balanced reporting gives uncertainty and disconfirming findings their proper place.
Make analytical grey areas visible
Exploration, cleaning and new hypotheses are not inherently improper. Quality fails when timing, alternatives or selection are concealed and the report invents a straight path. The following decision tools distinguish defensible change from misleading presentation. They cannot replace disciplinary methods advice, but they provide precise language for disclosure.
Correct errors and handle conflicts responsibly
A discovered error should be neither hidden nor instantly labelled misconduct. Preserve originals, describe reach and use the competent route. Competing interests require a factual declaration. Ombudspersons can advise confidentially before a disagreement becomes public or formal.
A practical route through the collection
For a new project, use the practice checklist, search log, documentation matrix and primary-data package. During analysis, add the audit trail, cleaning boundary and balanced result scheme. Before submission, reconcile the contribution matrix, integrity record and publication check. If a problem appears, follow the correction path, assessment tree and confidential ombuds route where necessary. The collection then functions as one process rather than twenty isolated articles.
Choose the guide that matches the problem
For preventive quality control, begin with the checklist or integrity record. If a stage can no longer be explained, open the relevant workflow for searching, analysing, writing or publishing. Questions about data belong with the documentation and primary-data guides; questions about analytical flexibility belong with the cleaning, HARKing and p-hacking decision tools. Once an error is known, use the correction route before reading broad definitions.
Keep three questions separate: What constitutes sound practice? What does this university require? What can this project actually evidence? The collection mainly answers the first and supplies tools for the third. Check the second in current original institutional documents. Record title, version and retrieval date rather than relying on a remembered summary.
Test the strength of the evidence trail
A strong record is timely, identifies one project state and explains both decision and reason. A retrospective account can still help, but should be labelled as reconstruction. Sample one central claim backwards from prose to analysis, recorded choice and original material. Repair a missing link or narrow the claim. Traceability never demands uncontrolled openness: personal data, confidential materials, credentials and third-party rights still require protection. Explain methods and limits without exposing material unlawfully.