Among comments to the previous post, Nick and Mizumi asked two questions that warrant a longer response.
Nick asked: I was not given a chance to review the paper post changes. Is this normal? As an author I just assumed that my responses always went back to the reviewer for final approval, but I guess not? If the flaws are major enough, do I request to see the authors' response prior to final acceptance?
I am speaking here from an author's perspective, and some of the responses may be field specific (I am in a hard STEM field). Those of you who are editors or simply in a different field, please chime in and correct/add.
If all referees requested minor revisions, it is likely that the editor will just look over the resubmission letter and make a decision without a second round of review. If the review featured a mix of minor and major revision requests, the resubmission can go back to all reviewers or it can only go back to those who requested major revisions (whereas they are sometimes asked to make sure the authors considered all the suggestions in all the reports). Indeed, when you request mandatory revisions, the resubmitted paper will most likely come back to you for a second round of review.
However, it is possble for you, as a reviewer, to request mandatory revisions and still not get the paper back. The scenarios in which this happens are the following:
(a) the other referee convincingly argues that the paper is beyond salvation, so the editor rejects it flat out and does not encourage resubmission;
(b) the other referee requires major revisions or recommends rejection, the editor does encourage resubmission, but the authors decide the reviews are so negative that it is not worth their time to try to fight it and never resubmit. They may go to another journal, but if they are smart they will take your review into account before submitting anywhere else.
(c) same as (b), but where the authors decide to wait a couple of months to resubmit. Due to the delay, the paper is considered a new submission and assigned a new set of reviewers;
(d) your review was so negative that the authors argued you were needlessly hostile and/or not objective, and they requested that the paper not be sent to you any more but that another referee be chosen. The editor was convinced and complied, so you as a referee are supplanted at resubmission. (FSP has a post that describes this scenario.)
I would say that (a), (b), and (c) are fairly common. I don't know if (d) has ever happened to me as a reviewer, but I have been on a team where we were the authors in a (d) scenario; once or twice we actually did get another reviewer.
As for final acceptance, the decision is ultimately in the hands of the editor. You as a reviewer are really advisory to the editor, and editors usually follow the reviewers' recommendations, but the final decision is the editors'. I am sure, though, that if a decision has been made on the paper you reviwed, you can shoot the editor an email and ask for a status update. I might have done that once or twice, and the response each time was that the authors never resubmitted.
@ Mizumi: Are there opinions on what is "fair" number of reviews to accept? Over the past few years as a PhD student and then postdoc I've maybe drawn 8-10 reviews per year. Does that mean 8-10 is a good number to take on, so at least I am not a net drain on reviewer resources?
Mizumi, how many papers to review per year is really a question of how much time you have and how much you like reviewing papers. It's an activity that is fun to do every so often (after all, it's the fun science you get to read and help improve), but it does become a chore if you are under constant siege. One paper a month, which is approximately what you have been doing, is likely feasible even for very busy people. Would you be able to do one review every two weeks? How about one review per week? I think more than one a week (~ 50 per year) is difficult to accommodate. Unfortunately, the higher you are on the totem pole, the more requests you get per unit time -- easily multiple requests per week. My goal is to hover around 1/week, but I will usually end up having 2 or 3 on my desk at any point in time, and another one or two which I have delegated to my students or postdocs. For instance, I review papers from multiple journals, maybe 8-10 journals fairly regularly. So even if each journal sends you a paper once a month, that's still plenty.
As with most things in life, start on the lighter side -- 1 or 2 papers per month are certainly a good service to your community, and the load will likely not smother you.
I am curious to hear how much of a review load other people have (please state career stage) and, from our readers who are also editors, how much refereeing load they think is appropriate.