The 27% Similarity Report and the Chair Who Is a Co-Author
A doctoral scholar submits a chapter. The institutional similarity report shows 27% overall: 11% from one journal article, 6% from the scholar’s own earlier conference paper, and 5% from a thesis held in another university’s repository, which she says she “only read for ideas”. The passages flagged from the journal article are reworded but keep the same sequence of arguments and the same three illustrative examples. The statistical analysis was performed by a colleague from the statistics department who is not listed as an author; the supervisor is listed as second author although he says he “only read the final draft”. The supervisor told the scholar that “everyone does this and the committee never checks”. The departmental research committee that will hear the case is chaired by the supervisor and meets next week; the scholar’s fellowship renewal is due in eight weeks.
1. Classify each issue as plagiarism warranting formal action, poor practice, or an authorship dispute. Justify each classification and rank them by severity.
Read the worked answer
Read the number, then read the text. Under the UGC 2018 plagiarism regulations the similarity bands run broadly 0–10%, 10–40%, 40–60% and above 60%, with graduated consequences, and the percentage is computed after excluding quotations, references, contents and similar matter. Twenty-seven per cent therefore falls in the second band, which typically calls for a revised submission within a defined period rather than the most severe penalty. Verify the current text and the institutional policy before acting, and check whether 27% is the pre- or post-exclusion figure - the difference changes the band.
But the number is not the finding. The flagged passages from the journal article retain the argument sequence and the same three examples: that is substantive plagiarism even where the string match was diluted by rewording, and rewording is exactly what a similarity score under-detects.
- Self-plagiarism / duplicate publication: reuse of her own conference paper. Ordinarily poor practice rather than misconduct if disclosed and cited; check the conference proceedings’ copyright terms.
- The other thesis (5%): overlap at that level is unremarkable, but unquoted reuse of its structure needs checking before it is dismissed.
- Authorship: the statistician performed the analysis and is omitted - that is a misconduct-level attribution failure. The supervisor’s gift authorship is a second, separate violation.
Ranking. The supervisor’s instruction and the gift authorship are the most serious, because they are systemic and they corrupt the process that is supposed to adjudicate the rest. The textual plagiarism is next. The self-reuse is the least serious.
2. The supervisor chairs the committee. Set out the procedure that is required before any finding is recorded.
Read the worked answer
Recusal is the first step, not a courtesy. The chair is a co-author of the chapter, the subject of a separate authorship allegation, and the source of the advice not to worry about checking. He cannot sit in judgement on a matter in which he is a party and a witness. The case should go to the institutional-level committee, with a departmental committee constituted without any co-author and without anyone who supervised or examined the scholar.
Then the elements of natural justice.
- Written notice of the specific allegations, itemised - the 11%, the 6%, the 5%, the argument-sequence similarity, the omitted statistician, the gift authorship - not a single lumped charge.
- Access to the similarity report, the comparison documents and the source passages, so she can answer the substance rather than the score.
- A hearing with the right to be accompanied, and a recorded opportunity to reply in writing.
- The statistician asked for her own account before any finding on authorship; she may have been content with an acknowledgement, which changes the finding.
Manage the pressure points. The eight-week fellowship deadline will push everyone towards a fast, procedurally defective outcome, which the scholar can then overturn on appeal; the timetable must be set by the committee and communicated, not allowed to compress the process. Minutes must record who decided what and on which document. The supervisor’s conduct should be referred separately so that it is not traded against the scholar’s penalty.