Plagiarism, Self-

Features of Self-Plagiarism

The definition of self-plagiarizing is quite broad, but it does have some specific features. The first feature concerns the person who is being plagiarized. Self-plagiarism clearly involves repeating one’s own work. In some respects, this can account for why self-plagiarism is controversial. With regard to traditional plagiarism, a researcher has victimized another scholar by not giving credit to his or her work. With regard to self-plagiarism, the same person is both the perpetrator and victim and by implication the author is aware of the plagiarism and has provided permission to do so. Hence, the person being plagiarized is not harmed.

The second element is focused on the content that can be self-plagiarized. Often plagiarizing is associated with lifting text without quotation marks or referencing the source and indeed, that is also part of self-plagiarizing. It is often referred to as “text recycling.” In some cases, authors engage in minimal paraphrasing so as to avoid repeating the exact words, but in other instances, authors “cut and paste” portions of their text from a prior work into a new paper.[Page 1259] In especially egregious cases, they publish the entirety of an article in more than one venue (e.g., an article is published in more than one journal, or as part of a conference proceedings and in a journal). However, self-plagiarizing can also occur when authors write about an idea, concept, or construct about which they have previously written. In this case, the text may not be exactly the same as in previous work, but the gist of the arguments and ideas are nearly the same. Finally, one can plagiarize oneself by reporting data that one has previously reported. This can take at least three forms. In some cases, individuals simultaneously submit the same paper to different outlets (e.g., multiple journals). In this case, the papers are identical and if published, no new information is provided. “Salami slicing” is a second form of this practice. In this case, authors submit papers that rely on the same data set, but report different analyses of it. Often they break a data set into the smallest publishable units (e.g., different measures or statistical analyses) and write different papers about them. Although each paper reports different analyses, each one makes little unique contribution. The third form is fragmented publication. In this case, a large data collection is conducted with a priori intention of treating them as unique projects. In such cases, a large number of measures are used and reported in separate papers. Unlike “salami slicing,” which often involves publishing tests involving a given relationship (e.g., using the same data set to test a direct relationship between X and Y in one paper and moderation in another), fragmented publication involves testing relationships between different variables (e.g., X and Y in one study and A and B in another).

...