From: Introduction to the Responsible Conduct of Research.
by Nicholas H. Steneck

Researchers share the results of their works with colleagues and the public in a variety of ways. Early results are usually shared during laboratory meetings, in seminars, and at professional meetings. Final results are usually communicated to others through scholarly articles and books. Public communication takes place through press releases, public announcements, newspaper articles, and public testimony. All forms of publication should present:

• A full and fair description of the work undertaken.
• An accurate report of the results.
• An honest and open assessment of the findings.

Deception
Deceptive authorship practices are arguably the greatest daily source of corruption in the sciences. Some examples include:

• Authorship by authority. A department chair or division chief or laboratory director either requires or permits his or her name to be placed on documents emanating from the unit. The most common justifications for this deception is that the leader either paid for the work, wrote the grant, provided resources, or in some other way was necessary for the work. But these are inadequate justifications - it is a deceptive mistake to include someone who did not do any of the work directly related to the part of the project described in the article.
• Gift, courtesy or honorary authorship. Attributing authorship to especially prestigious or socially "useful" colleagues. The problem here is still that the beneficiaries of these courtesies did not do any of the work. In other contexts, a well-liked student or lab technician might be rewarded with authorship; but this, too, is inappropriate if the student or tech's contributions were not adequate to justify bona fide co-authorship.
• Political authorship. This is related both to authorship by authority and courtesy authorship. The idea behind it seems to be that certain (important) colleagues will be angry, hurt or disappointed if they are not included as co-authors - despite that they did not do any of the work.
• Ghost authorship. Awarding authorship of a paper to someone (often a noted scientist) who is either unrelated or only peripherally related with the project, for a fee, in order to lend more credibility to the work.

It should be clear that such deceptions are additional forms of corruption or pollution of the scientific corpus. If you did not do any of the work, you should not get any of the credit. RULE OF THUMB: "If you are willing to TAKE CREDIT, you also must be willing to TAKE RESPONSIBILITY."

Authorship Guidelines
The International Committee of Medical Journal Editors (ICMJE) has established criteria for authorship. Authorship credit should be based on:

Authors should meet conditions 1, 2, and 3. [Emphasis added.] Acquisition of funding, collection of data, or general supervision of the research group, alone, does not justify authorship. Although each of these items are essential, authors must contribute intellectually to the project in a way sufficiently like the items above.

Coauthors and Determining the Order of Authors
Collaborators should discuss authorship at the start of their collaboration. The publication guidelines for many journals state that:

• The order of authors must not change without permission of all living authors once the article has been submitted for publication.
• Although coauthors may remove their names after that time, no author's name is to be removed by others.
• No authors names may be added after submission.

Redundant Publication, Self-Plagiarism, and Fragmentation
According to the International Committee of Medical Journal Editors (ICMJE), "Redundant (or duplicate) publication is publication of a paper that overlaps substantially with one already published in print or electronic media. ' Redundant publication corrupts - and unnecessarily bloats - the scientific corpus by suggesting that a particular scholar is more prolific than s/he actually is; which is deceptive. Such cases, in which one copies one's own work and then passes it off as novel, are sometimes called "self plagiarism." There are some justifications for what the ICMJE calls "secondary publication," but they all require some sort of disclosure in print, permission of both editors, etc. It should be uncontroversial to point out that not only must such redundancy be disclosed in print - it should also be disclosed or labeled in one's CV. Another way to inflate a CV is to divide the results so as to generate several articles, or even as many as possible (fragmentation). As with the order of authors' names, the question of what constitutes an appropriate publication strategy should be discussed and debated by the research team early in the research process.