If a professor is an editor of a journal, then should he ask his graduate students (PhD) to review articles? What do you think, good or bad on what points?
Different from "Is it common to review papers assigned to your supervisor?", where the professor is a reviewer and not the editor.
I think jakebeal's answer is too restrictive. When I was a senior graduate student, and my PhD supervisor was a journal editor, he would occasionally ask me to do a review for his journal where it was clearly related to my work. It was clear that by this time in my graduate program that he considered me an independent researcher and thus eligible to review for his journal. His ethics and judgement on such matters were beyond question.
Just as she does when she accepts an article for publication, an editor has to make a judgement call when soliciting reviewers. Some graduate students are clearly capable of making independent evaluations of the work of others and can do so without compromising the peer review process. We wouldn't give PhDs to students if they weren't ready for that responsibility (among other things), and most graduate students are ready for it well before they actually finish writing and do their defense. All such students are eligible in my mind to do peer review.
Edited to add: It's been awhile, but I'm pretty sure I was asked to do my first journal article peer review not by my PhD supervisor, but by the first journal I submitted an article to as first author (while I was a student). I don't think my full credentials were known to the editor of that journal, but even if they were, the editor clearly thought the first author of an article was a sufficient peer to ask me to do reviewing for their journal. It's very common in my experience to be immediately asked to do a review by a journal editor upon submission of an article.
I think that there is a significant difference between a professor reviewer asking a student to be a sub-reviewer (as in the linked question) and an editor directly asking their students to be reviewers.
An important goal of peer review is for the editor / program chair to obtain additional independent perspectives on a manuscript. If the editor requests a review from a student in their lab, whose review they will need to oversee and mentor, then they are losing the independent perspective and significantly compromising the peer review process. Thus, I think that it is inappropriate.
There is nothing wrong with asking your peers, including graduate students, to do a review for a journal, as long as it is done along the same guidelines as any other review for the journal. It would, for example, be poor if an advisor asked a graduate student to review a paper and providing hints of the expected outcome, "here's a joke of a paper for you to review". Then, of course, the editor would not be fit to be an editor.
Doing reviews is a job that must be learned by experience and since graduate school is preparing students to become independent researchers, having experience to review is a good thing. I do not think it is god to ask first year graduate students to do reviews but late-stage students should be able to manage. An advisor, relative to any unrelated editor, can of course best judge where the student expertise lies and hence provide papers that falls within the core expertise of the student. In the case of graduate student-advisor relationships, it will be key that the roles of advisor-student and editor-reviewer are kept separate.
It may serve an educational purpose, i.e. teach the student how the review process works. In that case it would be worth thinking about asking an extra reviewer in case the review is sub-standard.
In many cases I know (I am from sociology) PhD students develop before finishing their dissertation into full (albeit junior) collegues with an own perspective and specialization different from the advisor. If the paper that needs reviewing fits the specialization of the PhD student, then I see no problem asking them for a review.
Generally, I don't think this is appropriate. Here are a few reasons:
Ability to provide a completely independent perspective may be compromised: how does one ensure that one's research group is not vulnerable to groupthink? Even if one feels they are free from such biases, others outside the group may not be convinced--what would happen if they find out?
Possible conflict of interest: while an editor may choose not to 'hint' to the student on what an 'acceptable outcome' should be, it may be difficult to convince others that this is not the case. As another example, how about a scenario whereby a student wants to put in some objection to the article, but does not do so because of the editor? What if the editor deliberately chose the student because the editor knows that the student holds a certain set of viewpoints on that area of research?
Possible consequences if this becomes a widespread practice: supposing you just have one editor who thinks its perfectly fine to provide such 'hints' to his students? It may just take one person to break the system and undermine the reputation of a journal (that took a long time to build up).
Note: on the other hand, if the editor has already sent out the article to the reviewer(s), as per the journal policy, and you are providing an independent point of reference, then perhaps that is more acceptable.
I have one additional negative to nathaniel's list: student may feel compelled to perform a service (spend his valuable time) because of the control advisor has over student's Ph.D. Student may feel it hard to say no.
