Recently, I have been editing a special issue of a journal in my field. The special issue focuses on the topics in which I am an expert, so I know virtually all the contributors (at least the group leaders) fairly well. When soliciting reviews, I tried to contact postdocs and junior faculty about as often as I did senior people. This being a special issue means I have had to deal with a large number of papers in a fairly short time span and on a relatively small number of topics, which has helped make some interesting patterns in the review process obvious.
My initial guess was that younger people, postdocs and junior faculty, would be less busy and would be more likely to accept to review, would accept/decline to review more promptly (this journal provides an accept/decline option), and would overall be submitting their reports faster than the ultrabusy senior folks. Here's what I found:
Bar none, the fastest review (and a very detailed, to-the-point one) was received from a well-known, well-respected, and presumably very busy senior person in the field. In fact, several of the very busy, well-known people, whom I was reluctant to even ask for review because they likely have multiple review requests from different journals on their desk at any point in time, were very prompt in accepting to review, and were either on time or just a tiny bit late with their reports. These reports were always professional, detailed, and very useful.
So we can conclude "When you want something done, get a busy person to do it," right? Hold on, not so fast. Unfortunately, among the people who took the longest to even respond that they would indeed review, senior faculty are also the most numerous. Clearly, when it comes to plotting the distribution of referee responsiveness, the senior-folks distribution is tail-heavy.
Overall, I am quite puzzled by how many people take forever to accept/decline to review. I have never understood that. You read the abstract, and then decide whether or not to review, and click on a link. How hard is that?
I don't mind people declining to review, but I really appreciate it if they do it promptly. Unfortunately, postdocs and junior faculty have actually been much less likely to accept a referral for papers clearly within their expertise than I had hoped. I know that junior people are not idle, quite the contrary. My guess is that they don't want to review because they feel there is no immediate benefit to their careers from review, but I think that is short-sighted. Becoming a well-regarded reviewer improves your standing in the field and it keeps you at the forefront of your specialty. (FSP wrote about it. ) Junior people should have a light review load, but I think it's important that they review. It's part of training and part of being a good member of the scientific community. Being a sloppy or disinterested reviewer, while presumably expecting others to pull their weight on your behalf, is certainly not going to enhance your standing in the editor's eyes.
There is one person who submitted two papers to the special issue. Yet, I have not been able to get a single report out of him after multiple reminders. First he took forever to accept, and now now he's again taking forever to submit that one review. I would think that it's only fair that, if you want others to review your papers (note the plural), you should pitch in yourself, shouldn't you?
Another interesting issue, which was also discussed by FSP here, is whether to indicate that the paper you are reviewing is missing citations to your own work. In the case of a special issue, we are dealing with a focused set of topics, so this issue came up several times. I have found that most people have no qualms about requiring that their papers be cited (and generally rightly so), even though when the list of papers to cite contains only those from a single group it becomes quite obvious who the reviewer is (perhaps up to the uncertainty coming from a multiperson author list).
We all want our papers reviewed promptly and thoroughly -- I cannot imagine anyone wanting their paper reviewed slowly or sloppily -- but many people provide exceedingly terse reviews or are so late that the referral has to be withdrawn. It's an interesting feature of the human psyche -- we fail to treat others the way we would like to be treated ourselves.