This seems like a reasonable question for the main site.

I agree that it's a reasonable question for the main site.

I just wanted to point out that there was an article Topical Bias in Generalist Mathematics Journals by Joseph F. Grcar in the december 2010 issue of the Notices of the AMS. According to the text, the statistics are based on 854,547 entries from the 2000-2009 period of the Zentralblatt database. I am aware that this is not quite what you're looking for and unfortunately the article remains silent on exactly how the data was gathered, but it might be a starting point.

The question is now here by the way.

I really like the side question of ---- how could we convince the AMS to open up the mathscinet database?

Scott and Mariano, I have successfully asked narrowly focussed questions of the AMS Math Reviews editors, long in the past. As I suspect Mariano has done, I think a particular question ("could i please have statistics on this topic in this format") is likely to be well received, a question to open up the database less so. Those involved are justifiably proud of what they are doing. This includes many whose training has less to do with mathematics and more to do with libraries, archives, writing and publishing.

I have a related question:

Where can I locate reputable lists, by MSC or MO major tags, of what the community considers major journals in those areas?

My need is too local, and the question too broad, to post on MO, so I don't think I can post there. However, the need is real, so any advice on where to seek the answer is welcome.

The Canadian federal science funding agency (NSERC) is going through a 'reallocation' exercise, in which it determines how much money to spend on what discipline. It has decided to rely on a report being written now, which aims to 'compare' between disciplines. The comparison is based on a rather crude metric derived from bibliometric data called the 'Average Research Impact Factor (ARIF)' with 1 being 'average', and 'Relative ARIF', being ARIF - 1. Someone is using Thompson Reuters to generate a list of 'top N journals in all math+stats' by citation measures, which are then incorporated into the report. Unsurprisingly, journals on econometrics makes this cut when N=10 or 20, whilst other major journals do not.

Canadian Mathematics+Statistics is the only discipline which apparently has negative R-ARIF. The funding for basic research in Mathematics and Statistics has been cut over the last exercise, and will most likely be cut again. There's outrage and squabbling over how grants are reviewed, http://nghoussoub.com/ but the source problem- underfunding- remains relatively unaddressed.

We (mathematicians in Canada) need to coherently and rather quickly point out the methodological fallacies in how these metric are computed, but more importantly, suggest other measures. Unfortunately, these measures have to be simple enough for non-scientists to compute. I was looking, therefore, for lists of 'top 10 journals' by major MSC category, where the ranking was decided by mathematicians (and not simply by IF). Handing over rank-ordered lists by subdiscipline where the ordering is 'agreed' upon by mathematicians is a lesser evil than having such a list thrust upon us. It currently is.

Unfortunately, I haven't been able to find such lists. There are lots of lists by lots of people, but I'm not sure which are considered authoritative, say, in algebraic geometry. I looked at Scott's question on MO. That got lots of interesting answers, but none that quite address what I'm looking for.

Any pointers welcome.

an_mo_user, that's a great start, thanks!