In that framework, I have found the New Oxford Book of Romantic Period Verse to be far and away the most useful in its area.
First and foremost, the selection is nearly comprehensive. Practically every important poem of the period is included, and a number of interesting but much less well-known works too.
This is generally a good thing, of course, but especially so when it comes to Romantic poetry, as what makes this period so powerful and engaging is the fact that so many crucial verse works appeared during this time, so the more the students (and others) can juxtapose them, the better sense of both the individual poems AND the period as a whole they will receive.
This is reinforced by the method of presentation, which is fundamentally chronological -- i.e., the poems are arrayed by the year in which they first appeared, and only within that year by author, if the writer produced several during that time.
This is quite innovative and tremendously useful, especially for my purposes, but, I would argue, even in general, as a chronological approach necessarily gives the reader (student or not) a strong sense of the historical relationship among these works -- an approach that I, at least, think is far more important than many others apparently do today -- and is valid for understanding painting and other cultural phenomena as well.
Having praised the near-comprehensiveness, and innovative chronological mode of presentation, there are certain flaws here.
The main substantive one is the way-too-abstruse tone and content of the introduction.
Having made such an advance by presenting the poems in roughly chronological order, the editor should have continued with the instinct towards accessibility and understanding.
Why write a stuffy and not very interesting academically-oriented introduction, when he could have written instead a clear piece that would help the uninitiated understand the chief issues involved in Romantic poetry, while, at the same time, offering within that framework insights that would be intriguing for experts in the field?
This is not that hard to do, although most academics -- Simon Schama and Michael Wood being notable exceptions in the cultural history field -- seem to have a problem with this approach.
From a methodological point of view, finally, it seems a little strange not to have included certain very important works -- notably Wordsworth's Prelude, but also others -- simply because they were not published during the period in question (1785-1832, a choice of years that in and of itself seems quite appropriate).
When dealing with well-known works like the Prelude, or Keats' "'Lear'," or Blake's "4 Zoas" or Shelley's "Epipsychidion," it seems bizarre not to include them during the years they were written, with little asterisks indicating that they were not published, but written, at that time.
These issues aside, this is a wonderful collection that will be incredibly useful for anyone with any reason at all to be concerned with English Romantic poetry -- which, in my view, should be all of us.