Teznevise English

Research Integrity Rules

Core research-integrity rules for Teznevise academic writing and research support.

Research integrity rules define the minimum standards expected when planning, analysing, writing and presenting academic work. They apply whether a project is completed independently, with supervisory input, with statistical or editorial support, or with permitted use of digital tools. The core principle is simple: evidence, authorship and methods must remain honest, traceable and consistent with the rules of the relevant university, programme, journal or funder.

Use authentic evidence

Research data, participant information, observations, citations and results must not be fabricated, altered or selectively removed to create a preferred conclusion. If data are excluded, transformed or recoded, the reason should be methodologically defensible and documented. Where a result is inconvenient or non-significant, it should still be reported accurately if it is relevant to the research question.

Keep authorship and responsibility clear

The researcher remains responsible for work submitted under their name. External help should not replace the student’s or researcher’s authorship. Editing, tutoring, methodological guidance and technical support can be legitimate when they improve understanding or presentation without secretly producing assessed work on someone else’s behalf.

Where an institution, journal or assessment requires disclosure of external assistance, software or generative AI, that disclosure should be made. For the boundaries applied to Teznevise support, see the Academic Integrity and Fair Use Policy.

Choose methods from the research question

Analytical decisions should follow the study design, measurement level, assumptions and research question. A method should not be selected because it is more likely to produce statistical significance. Assumption checks, model changes, missing-data handling and sensitivity analyses should be reported when they materially affect interpretation.

Statistical significance is not the same as practical importance. Where appropriate, report effect sizes, uncertainty, confidence intervals and limitations rather than presenting a p-value as the entire result. Our statistical analysis support is intended to help researchers understand and report their own analyses responsibly.

Use sources accurately

Citations should point to sources that actually support the claim being made. Do not cite papers you have not checked, invent references, or rely on generated bibliographic details without verification. Direct quotations, paraphrases, data, figures and distinctive ideas should be attributed according to the required citation style.

Similarity software can help identify text that needs review, but a similarity percentage is not a target to manipulate. Replacing words mechanically or disguising copied structure does not resolve plagiarism. The aim is accurate attribution and genuine authorship.

Respect participants, confidentiality and permissions

Research involving people, confidential records, unpublished data or restricted materials may require ethics approval, informed consent, data minimisation or access controls. Support services cannot waive those obligations. Use only the data and documents you are authorised to access, and follow the rules set by the relevant ethics committee or institution.

Report limitations and uncertainty

Good research does not pretend that every design is stronger than it is. State limitations that affect generalisability, measurement, causal interpretation, missingness or potential bias. Distinguish what the evidence demonstrates from what is only plausible or exploratory.

Defence, viva and examination

Preparation support can help a researcher organise explanations, review likely questions and practise defending methodological choices. The researcher must still understand the project, attend the defence or examination where required, and answer for the work submitted under their name.

Using AI and software responsibly

Rules for generative AI, coding assistants, grammar tools and automated analysis vary by institution and assessment. Before using them, check the current policy that applies to the specific task. Where use is permitted, verify outputs, keep appropriate records and disclose use when required. AI-generated citations, calculations and factual claims should never be accepted without checking the underlying source or method.

What Teznevise will and will not support

Teznevise can support planning, research methods, analysis, editing, tutoring, formatting and revision within the limits of the user’s institutional rules. It will not fabricate data, create false evidence, conceal prohibited authorship, manipulate similarity systems or invent academic credentials. The How Academic Support Works page explains how requests are scoped.

A practical integrity checklist

  • Can every important factual claim be traced to evidence?
  • Are citations real, checked and relevant?
  • Were methods chosen for defensible reasons rather than a desired result?
  • Are exclusions, transformations and limitations documented?
  • Does the submitted work remain the researcher’s own responsibility?
  • Have institutional rules on external help and AI been checked?
  • Could the analysis and reporting choices be explained openly to a supervisor or examiner?

These research integrity rules are not a substitute for your institution’s formal policy. When requirements differ, the current university, programme, journal or funder rules take priority.

Need help applying this to your research?

Describe your programme, research stage and the decision you need to make.

Request guidance