Journal of Statistics and Data Analytics
Similarity Check Policy
Similarity Check Summary: All manuscripts submitted to Journal of Statistics and Data Analytics (JSDA) are subject to originality and similarity screening as part of the editorial evaluation process. The journal uses similarity-detection procedures to identify possible plagiarism, inappropriate text reuse, duplicate publication, and excessive overlap with previously published material. The overall similarity level should normally not exceed [MAXIMUM SIMILARITY LIMIT], but editorial decisions are based on qualitative examination of the similarity report rather than on the percentage alone.
Similarity screening is conducted to protect the originality, credibility, and integrity of scholarly publications in JSDA. The screening process assists the Editorial Team in identifying:
| Initial Screening
Similarity may be checked before a manuscript is sent for external peer review. |
After Revision
A revised manuscript may be screened again when substantial changes have been made. |
Before Publication
The Editorial Team may conduct an additional check before final publication when necessary. |
Maximum Overall Similarity
[MAXIMUM SIMILARITY LIMIT]
Subject to qualitative editorial assessment
Important: The numerical similarity percentage is not interpreted as an automatic measure of plagiarism. A manuscript with a relatively low percentage may still contain serious plagiarism, while a manuscript with a higher percentage may contain legitimate matches arising from references, standard terminology, methodological descriptions, or other appropriately attributed material. Therefore, each similarity report is evaluated contextually.
The Editorial Team may consider the location, source, extent, and nature of matched material. Particular attention may be given to overlap in:
Because JSDA publishes research in statistics and data analytics, some manuscripts may naturally contain standard mathematical notation, statistical terminology, well-established formulas, algorithm names, or conventional methodological expressions. Such similarities are not automatically considered plagiarism.
However, original derivations, new statistical models, new algorithms, new theoretical results, proofs, interpretations, and methodological contributions must be presented as the authors' own work and appropriately distinguishable from previously published material.
Descriptions of widely used procedures such as regression, ANOVA, Bayesian estimation, Monte Carlo simulation, machine learning algorithms, time-series models, sampling techniques, or other standard statistical methods may naturally use common technical terminology. Nevertheless, authors should write methodological descriptions in their own scholarly language and appropriately cite the original, foundational, or relevant methodological sources.
Authors should not substantially reuse their own previously published text without appropriate citation or disclosure. Limited overlap may be understandable in descriptions of established methods or research settings, but extensive reuse of the Introduction, Results, Discussion, Conclusion, or other substantive content may constitute inappropriate text recycling.
When a submitted manuscript extends previous work by the same authors, the relationship between the new manuscript and previous publications must be transparent. Authors should clearly explain:
Authors must not submit a manuscript that is substantially identical to work already published or simultaneously submitted elsewhere. Fragmentation of one study into multiple manuscripts without sufficient independent scientific contribution may also be considered redundant publication.
Tables, figures, graphs, data visualizations, or other materials previously published elsewhere must not be reproduced without appropriate attribution and, where required, permission from the copyright holder. Changing colors, labels, orientation, or other superficial graphical features does not transform previously published material into original content.
Reuse of previously published datasets, computer code, algorithms, simulation procedures, or analytical workflows must be transparently documented and appropriately cited where applicable. Authors should distinguish between reused resources and original contributions developed specifically for the submitted manuscript.
References, bibliographic entries, article titles, institution names, software names, standard statistical terminology, and other unavoidable expressions may contribute to similarity scores. Such matches are assessed in context and are not automatically treated as evidence of plagiarism.
| Finding | Possible Editorial Action |
| Minor Legitimate Overlap | The manuscript may proceed if the overlap is properly attributed and does not affect originality. |
| Correctable Similarity | Authors may be asked to revise, paraphrase, cite, or clarify the relevant material. |
| Substantial Unattributed Overlap | The manuscript may be returned for substantial revision or declined. |
| Serious Plagiarism | The manuscript may be rejected and further action may be considered under the journal's Publication Ethics Policy. |
| Post-Publication Concern | The journal may investigate and issue a correction, expression of concern, or retraction where appropriate. |
A similarity percentage above the journal's stated threshold does not necessarily result in automatic rejection. Likewise, a similarity percentage below the threshold does not guarantee that a manuscript is free from plagiarism. The Editorial Team evaluates the actual matched content, its source, its scholarly relevance, and the adequacy of attribution before determining the appropriate editorial action.
Authors are responsible for ensuring that their manuscripts comply with standards of originality and proper attribution before submission. Authors should carefully check all quotations, paraphrases, citations, figures, tables, data, code, and other reused materials. Submission to JSDA constitutes an acknowledgment that the manuscript represents original scholarly work except where sources are appropriately identified.
The use of generative AI does not exempt authors from responsibility for originality. Authors must verify that AI-assisted text, code, references, data descriptions, or other content does not reproduce copyrighted or previously published material without appropriate attribution. AI tools must not be used to disguise plagiarism or improperly paraphrase another author's work.
Similarity screening forms part of the journal's editorial evaluation process. Manuscripts, similarity reports, editorial comments, and related communications should be handled with appropriate confidentiality by authorized editorial personnel.
Cases involving suspected plagiarism, duplicate publication, text recycling, citation manipulation, data misconduct, or other serious ethical concerns are handled in accordance with the JSDA Publication Ethics and relevant editorial policies.
Journal of Statistics and Data Analytics is committed to protecting the originality and integrity of the scholarly record. Similarity screening is used as an editorial tool to support responsible publication, but findings are always interpreted within their academic, methodological, and disciplinary context.
Journal of Statistics and Data Analytics
PT. Gammabeta Edu and Tech
Jl. Rijali RT 003 RW 001
Ambon 97123, Maluku, Indonesia
E-mail: jsda@gammabeta.id
Journal of Statistics and Data Analytics
Statistics • Data • Analytics
Published by PT. Gammabeta Edu and Tech
|
Editorial Office
PT. Gammabeta Edu and Tech
Jl. Rijali RT 003 RW 001 Ambon 97123, Maluku, Indonesia E-mail: jsda@gammabeta.id
|
Editorial Office Location
|
