There is a lot of discussion about optimizing and modifying scientific data, plagiarism and multiple publication of (parts of) the same data. This is very important and necessary for the quality of and esteem for science. But, what are the definitions of these disputable activities? What is acceptable and what is wrong?
I my self re-use definitions of certain scientific parameters in publications: a consequent application of the same parameter makes a clear and consistent definition necessary. This was, however, recently discussed as a possible violation of correct citation. The problem could be that other co-authors of previous publications are not credited when (all) those previous publications are not correctly citated.
Do we need clear definitions? What are acceptable forms of self-citation and of obvious use of basic principles? Do we have to cite Pearson all the time when we calculate a correlation coefficient?
That's a good point. I don't think we need to cite Pearson all the time, mainly because the coefficient already carries his name and his legacy. But when it comes to equations, definitions, metrics, and other concepts created by early, not so acclaimed authors, it is not appropriate, I suppose, to not give them credit and recurring instead to self-citation.
So, I'd say that while there is nothing wrong in principle with self-citation, lots of scientific malpractices may be disclosed when closely inspecting the motivations and contexts with which authors cite their own previous research.
In my case, I use self-citation to illustrate applications areas that were previously explored by the research groups I've been involved with, when investigating some method, technique or algorithm of interest. I think this motivation is fair and resasonable because I usually can't think of a better recent alternative to illustrate a point and because I tend to trust the scientific quality of the works I've authored more than third-part papers that I am not familiar with/or couldn't grasp all relevant methodological details. A few authors, nonetheless, will abuse self-citation and include their previous papers in the references section without a properly justified motivation.
While I don't think self-citation equals plagiarism in the sense you mention, I do believe that it is a good place to start if one wants to detect scientific misconduct in many levels. Your points are certainly valid, but this is nonetheless controversial, since motivations can be subjective and many authors who self-cite all the time may claim to be doing it in good faith (and they may be speaking the truth).
Anyway, I strongly believe that scientific reputation should not be based on citation metrics anymore. We need a new, more open and reliable model asap. Researchers need to make their data and protocols public and be evaluated by the quality of their data, methodology and analysis, not on the gibberish that some try to pass foward when overselling their findings. Something like a Github for science; some open-access, public space, at which researchers can deposit in an organized way the details of their hard work, including, but not limiting to, source codes, lab notes and diary entries, etc; something like an online portfolio that could be evaluated by the whole community.
No it is not plagiarism. It must be used as and when required in future.
- Badges
- Science topic
- Similar topics
- Philosophy
- Philosophy Of Science