Scientific publications can be rescinded for various reasons, which can be categorized by severity. Minor issues that are often corrected and then republished include administrative mistakes, reference errors, or copy-editing oversights. More serious grounds for retraction involve significant problems with the research itself, such as data fabrication, falsification of results, plagiarism, serious flaws in the research design, or unethical treatment of subjects.