This protocol establishes a consistent, documented, and human-reviewed procedure for assessing text similarity in submitted manuscripts and helping editors identify passages that require further examination before peer review or publication.

1. Purpose
To establish a consistent, documented procedure based on editorial assessment for evaluating text similarity in submitted manuscripts and identifying passages that require further examination before peer review or publication.

2. Scope
This protocol applies to newly submitted manuscripts, revised manuscripts, and pre-publication checks. It is intended for manuscripts submitted to AKSIOMA: Jurnal Program Studi Pendidikan Matematika, subject to the coverage of the databases available in the similarity-checking software used.

3. Similarity-Checking Tools and Settings

• Use a similarity-checking tool approved by the journal, such as iThenticate, Turnitin, or an equivalent service.
• Apply consistent settings to all manuscripts. References and quotations are excluded in accordance with the journal’s policy.
• Record the original report, date of screening, name of the person conducting the check, manuscript version, and settings used.
• Editors must examine the sources and contexts of matched text rather than relying solely on the overall similarity percentage.

4. Similarity-Checking Flowchart

1. Manuscript received by the editor
                                 ↓
2. Run a similarity check (e.g., iThenticate, Turnitin)
                                 ↓
3. Review the report: overall score, main sources, and matched passages
                                 ↓
4. Manually assess the context of matched text (citations, methods, text recycling, etc
                                 ↓

DECISION: Are there indications of a potential violation?

NO — No indication identified	YES / UNCLEAR — Further examination required
Document the result and proceed to peer review.	Identify the sources and assess the seriousness; request clarification or revision and refer to the journal’s publication ethics policy.
Retain the report and screening notes in the editorial archive.	The editor decides whether to proceed, request revision, or reject the manuscript based on the evidence and applicable policy.

Note: Similarity scores are screening indicators and do not, by themselves, establish plagiarism.

5. Operational Procedure

1. Manuscript receipt. Record the manuscript ID, title, date received, and version of the document screened.
2. Software screening. Run the similarity check using settings consistent with AKSIOMA’s policy (references and quotations excluded). Do not alter settings merely to obtain a lower score.
3. Overall-score review. Record the overall score, number of matched sources, and sources contributing the largest matches.
4. Source-by-source and section-by-section review. Inspect matched passages and determine whether they involve properly cited quotations, common phrases, standard methodological descriptions, the authors’ previous publications, or unattributed text.
5. Editorial assessment. Determine whether the matches are reasonably explained or require clarification, revision, consultation with an ethics editor, or other action under journal policy.
6. Decision documentation. Confidentially retain the report, screening notes, correspondence with the authors, and rationale for the editorial decision.

6. AKSIOMA Similarity Thresholds

The following thresholds are specified in the supplied AKSIOMA protocol.

Indicator	Threshold	Follow-up action
Overall similarity	≤ 20%	Review the report and sources; do not make a decision based on the score alone.
Similarity from a single source	≤ 1–3%	Inspect the matched text, its location, and attribution.
Quotations and references	Excluded	Apply exclusions consistently and still examine the use of quotations.
Results and discussion	No substantial unattributed use	Conduct a detailed review and request clarification if necessary.

A score below the threshold may still involve a serious integrity concern, while a score above the threshold may result from legitimate matches.

7. Manual Evaluation Criteria

• Is the matched text adequately cited and attributed?
• Do the matches come from a single source or are they distributed across multiple sources?
• Is there substantial copying in the abstract, introduction, results, or discussion?
• Has text from the authors’ own previous publications been reused without adequate transparency?
• Are the matching methods or technical terms conventional and necessary in the field?
• Have another party’s tables, figures, data, or ideas been used without the required permission or attribution?

8. Possible Editorial Decisions

Finding	Possible action
Matches are explainable and adequately cited	Document the result and proceed to the next editorial stage.
Attribution or paraphrasing issues that can be corrected	Request specific revisions and screen the revised manuscript again.
Extensive matches or an unclear pattern	Request clarification and conduct further editorial examination.
Strong indication of an integrity breach	Follow the journal’s publication ethics policy, due-process procedures, and relevant guidelines.

9. Key Principles

• Similarity is not synonymous with plagiarism; reports must be interpreted in context.
• Do not reject a manuscript solely on the basis of one percentage without examining the matched passages and sources.
• Apply screening settings consistently and transparently.
• Protect the confidentiality of unpublished manuscripts.
• Follow the journal’s policies, publisher requirements, and applicable publication-ethics guidelines.

10. References and Guidance

• Understanding the Similarity Report. https://www.crossref.org/documentation/similarity-check/similarity-report-understand/
• iThenticate and Plagiarism. https://guides.ithenticate.com/hc/en-us/articles/38053179354125-iThenticate-and-Plagiarism
• Committee on Publication Ethics (COPE). Text Recycling Guidelines. https://members.publicationethics.org/sites/default/files/Text_A29298_COPE_Text_Recycling.pdf

 

SIMILARITY CHECKING PROTOCOL PDF download