|Part of my off-campus summer reading|
While I do not have PhD level courses in my program, and focus on research only, the topic of reading loads itself is interesting enough to discuss.
The question "How much should we read?" is always related to "Should students read deeper or broader?"
In my opinion, it really depends on which stage you are at in your reading, and for which purpose. In a way, I've dealth with the phases of reading in an earlier post, but let's just elaborate on the significance of reading in depth as well as reading in breadth.
1. In-depth reading
In-depth reading is, in my opinion, very similar to studying. If you have skimmed through a paper, and found a gem in there for your research, you need to sit down and work through the material until you fully understand it. In my case, that means a lot of sketching and seeing if I can derive the formulas myself. Reading like this (for technical work/derivations) can go as slow as only 2 pages an hour (on average, I'd say 10 pages if it's very well detailed in a thesis, and much less if you need to go and hunt in other references to figure out why the others took a certain step in their derivation).
2. Reading for a broader understanding
My EndNote PhDBibliography library is holding a little under 600 references at the moment - all of which I have read (and a few more piles in my office on shelves to work through). I have read about shear in reinforced concrete and prestressed concrete beams (with and without web reinforcement), punching in slabs, live loads on bridges and design of reinforced and prestressed concrete bridges. I wanted to broaden my understanding and background knowledge of these topics, so I read as much as I could lay my eyes on. Also, when I need to discuss a "new" topic (for example, sustainability aspects related to the possible outcomes of my research), I gather as many references as possible and get into sponge-mode. For that purpose, I spend about 20 to 30 minutes per paper, trying to understand the main ideas, the implications, and have a deeper look at the abstract, conclusions and figures - and see if it refers to other interesting work.