As an academic, how do you read?

March 3, 2009 1:49 PM

If you're a grad student or academic, how do you read new articles and books? Do you take notes/marginalia? Do you read some parts systematically and skim/ignore the rest? Read each piece multiple times? Do you find you learn more from reading than from talking with others or attending presentations? Please indicate what field you're in.
posted by shivohum to education (22 comments total)

I'm a grad student, in Critical Studies in Film, which is formally similar to a number of other Humanities fields. I read everything for classes through unless I'm badly behind in the class. I have other reading to do for my own research and for a Masters Exam to do besides that. I use highlighters, and don't usually miss the understanding or reminder that an explicit marginal note creates. Selective, somewhat coded highlighting is augmented by post-it flags for really important chapters or ideas that will be integral to some paper or or idea I'll explore further. I keep everything, filed and shelved.

Oh, I have recently been assigned some stuff I'm planning not even to skim - fiction related to genres I'm studying. It's just totally last priority.

I learn plenty from presentations and classes, and take notes in both, though I seem to be one of the more consistent note-takers during presentations that I've noticed. So far, I haven't had a major "learning" experience from one on one or group conversations, but it definitely supports my work and my sense of vitality and investment. I like to review my notes periodically to remember what I've noticed. For that matter, I jot ideas into my iphone at events, screenings, wherever I get an idea that relates to my studies.
posted by Ambrosia Voyeur at 1:56 PM on March 3 [1 favorite has favorites]

I'm a postdoc in genomics. I always have more papers to read than I have time for, so I try to focus on moving on as quickly as possible. Every paper has one of 12 different priority codes (3A-5D) and I will frequently downgrade an article after reading a couple of sections if I think there is more to be learned from it but other stuff is more important right now. I underline things as I go through, but usually take notes only as followup items for myself. I usually have several stacks of papers ready to go, one in my jacket, one on my fan binder I take with me on the bus, one in my desk at home, one in my desk at work, one in my portfolio I take to seminars.

Where I learn the best depends entirely on the quality of the presenter—I'll learn more from a great talk than a bad paper and I'll learn more from a great paper than a bad talk. On the whole, I find learning in multiple modes useful.
posted by grouse at 2:05 PM on March 3

I'm in Public Policy/Public Administration and I find that a mix of reading styles is really called for especially if you have a heavy reading load. For most grad students and professors it seems being able to skim and pull out the core concepts is extremely crucial. Some people basically read the abstracts and conclusions and only dig into the meat if the material seems particularly relevant to their research interests. Personally I try to read everything and only start skimming/skipping if the article/book seems extremely repetitive or overly technical.

I typically don't take a bunch of notes but I think most grad students should keep a master bibliography of articles that they have read, including citations and the core concepts expressed in that article/book. It can really help you with comps and writing a dissertation down the road.

Except for a handful of postmodern texts, I rarely have to re-read articles but yeah there are always going to be authors who seem to go out of their way to make their work as close to undecipherable.

I learn pretty well from texts but I can really get my brain cranking on all cylinders when I discuss things with other people. Sometimes just discussing a difficult concept with another person can give you insight into a problem that is otherwise eluding you.
posted by vuron at 2:07 PM on March 3 [1 favorite has favorites]

Never read anything once. Always read the beginning and end, then skim the entire thing, then read closely. Finally, scan for load-bearing sentences and make a note of them. After that you can do more if you have time.
posted by billtron at 2:10 PM on March 3

Oh, I'm in ethnomusicology
posted by billtron at 2:11 PM on March 3

Economics - read abstract & conclusion, if that's promising skim rest, if still promising read fully, throughout - underline key points and note key words in the margin.
posted by koahiatamadl at 2:23 PM on March 3

Neuroscience. I (unfortunately) tend to read most when I'm actively writing something. I have a hard time making myself read otherwise. Weekly "journal club" meetings help with that a lot; if you have a chance to attend something like that, do it.

I read abstract, intro, results, discussion, in that order, then I dissect the methods if anything seems screwy. Other times I mine the intro and discussion for further readings, seeing who they reference for main points if I think the idea is either helpful to my work or a potential hindrance.

Used to scribble main points onto a Post-It and stick that on the first page of the paper (I hate highlighting / marginalia) and have tried making a "master list" summarizing main points, but these days I just drop the entire abstract into EndNote and use the "Search Library" feature to remind me which references hit the right keywords when I'm writing. All papers are sorted into folders (10 per folder, indexed by EndNote ref # which I write in the top corner of the paper) If it's really important, I keep a printed copy out on my desk. If it isn't something I can print, say a book chapter, I try to photocopy the important parts if I don't actually have the book on hand.
posted by caution live frogs at 2:24 PM on March 3

Library Science for Masters, Public History for Phd.

I combine a number of methods. If I have the time, I read the book or article straight through and make notes of important or interesting thoughts in a notebook, almost like citations but much briefer. This helps me know what I'm going to refer back to when I write a book review or discuss the book or article in some way. I also like to discuss the book with fellow students and see if they've picked up stuff I've missed. Finally, I like to do a little digging about the author and see what his/her other publications look like, what institution he/she is associated with, and (tackily enough) what the "Rate my Professor" rating is.

If I'm slightly pressed for time, I'll read the introduction and the major few paragraphs in each chapter and the conclusion. I try to supplement that reading by reading a few reviews on the book from journals.

If I've completely blown all my time and totally forgotten to read something, I'll just read a review. If it's an article or there's no review available, I'll try to talk it over with some of my friends and see what they got out of it. Then I'll just keep my mouth shut during discussions. Failing that, I try to associate the information I have on the subject with what's being discussed.
posted by teleri025 at 2:29 PM on March 3

My field is finance, specifically Quantitative Finance (aka Mathematical Finance).

I teach part time at three Universities and am completing an Executive MBA; currently writing my dissertation and tying lose ends up on case studies.

For me reading papers is almost a constant. I read papers mostly as part of my own research.

I take a first pass online using Athens or another aggregator of academic articles and if the paper passes this filter I download and print.

It gets categorised according to high level use (e.g., my research, case study, perhaps a paper of interest for one of the finance classes I teach) and then is further classified. Some papers will fall into more than one category, and for those I kill another tree. Papers that form part of my research are categorised according to their use in my paper according to a simple A / B / C ... etc system.

Papers tend to travel with me, and I make copious use of two different coloured pens, as well as a yellow highlighter. The highlighter isolates text I'll need to reference in a paper I'm writing and cite, while the pens are used for black (meta - i.e., category, etc) and red (questions that arise from reading). These notes go on the front (black) and back (red) pages.

As I work everyday I copy / paste referenced text with full citations into a word document that I build and maintain on the go. This document becomes the nucleus of any papers I publish, push over to my dissertation advisor, or possibly reference in class.
posted by Mutant at 2:38 PM on March 3

Grad student in political science here. I tend to get a lot more out of readings when I'm reading them for something like a paper rather than just for seminar. Perhaps because then I'm both getting out the main points and explicitly thinking about what's useful or relevant to my work, rather than just reading for content. I generally get the most out of reading, followed by seminar, and take quite comprehensive, detailed notes.

I've recently switched from a paper-based, marginalia-writing system to one that's entirely electronic. So I have electronic copies of journal articles, scans of book chapters were possible, etc. As I read, I have a text file open to take notes in, with page numbers and quotes and so on.

The goal with switching to this electronic system is the ability to access everything later - that article whose name I can't quite remember, making sure that I don't lose some insight illegibly scribbled on the back of a printed-out paper, and so forth. To that end, I keep a 'library' folder on my desktop, with everything named "Lastname_Year_Title." I also use Punakea to tag those files, like "Globalization" or "French Revolution." I've also been using Zotero for bibliography management and other metadata, though it's a pain to make sure it's current. Also, it's pretty much useless to me as a way to actually manage the files, as it puts them in an obscure place which you apparently can't change, and isn't reliable at grabbing files from online journal databases, and manually adding them is too much of a bother.

So far I'm really happy with this system, except for 3 main problems: it takes me a LOT longer to read on a screen than on paper, so I've been doing a bit more skimming than thorough reading. Also, all my pdf's are images, which means that they're non-content-searchable. I'm looking at getting some kind of optical character recognition going to fix that. Finally, in taking the fuller, computer-based notes, I've noticed more of a tendency to just 'write things down,' without analyzing it, a problem I haven't had with the more limited marginalia I'd used earlier.
posted by foodmapper at 2:39 PM on March 3

PhD student in English. I read a lot (obviously), but I tend to be a bit choosier about which article or books I'll read (I'll vet the author a little more, be sure it's someone reliable). I spend extra time with indexes, glossaries and first and concluding paragraphs than I used to. I also read book reviews sometimes to see how a given work has been received (yeah, sometimes I'll do this to compensate for a superficial reading of the work).

I usually get more content-wise by reading than through conferences, but the live setting of conferences tends to spark fresh ideas and help me decide what's relevant in the field.

My marginal notes have evolved over the years. I tend to write simple keywords in the margin so that when I need to refresh myself on the text quickly, I can simply read the keywords and reconstruct the argument. I also have a technique where (always in pencil) I underline key points and then draw arrows through the text to the following key word in the argument. I find that mapping this way turns large blocks of intimidating text into handy mind maps.

The best advice I can offer about reading articles, which I don't always follow, is to jot a one-sentence summary of the writer's overall project at the top of the article. Nothing fancy. It's a little work, but the mental engagement it requires will really help you remember the reading.

In terms of what I read, I tend to follow my fancy more than when I was a spring chicken grad schooler. Some texts just click, others don't. Keep track (through searchable notes on your computer, etc.) of those you find stimulating. Worry less about the ho-hum ones.

Also, I keep a bibliography by topic; whenever I come across a text on a given topic, I'll type in the name of the work + author. I started keeping this when I realized I was often asking professors for book recommendations on a topic and realized that I'd have to do this someday for students too and I didn't want to come up empty-handed.
posted by cymru_j at 2:43 PM on March 3

useful previous theads:
How do you take notes as an academic?
Again, how do you take notes as an academic?

there are a lot of other useful threads on more-specific issues (eg good programs to help with notetaking) under the "notes" tag
posted by LobsterMitten at 2:46 PM on March 3 [3 favorites has favorites]

Depends why I'm reading the article. (physics postdoc, theory.)

I read some things just to keep up with their subarea; I won't be doing research or presenting on these papers, so sometimes I'll just read the abstract; maybe add intro and conclusions or a skim over figures if I get interested; I'll scribble in a few marginalia but they're mostly useful for humor value if I end up reading the paper again.

I read some things because I'm going to have to present on them (journal club, or background for a talk). These papers I read once straight through (not too heavily), then follow through and make sure I know what the variables in each equation mean; I will check the algebra for a few results to make sure I follow the definitions (usually in the margins so I have the check available for future use).

I read some things because I am doing/going to do research on that exact subarea. These papers end up with marginalia, scribbles, coffee stains, I-dropped-it-on-the-snowy-street stains, etc. Also multiple copies. I like using multiple colored pens so I can easily highlight things in different ways.

At this point, I am all paper-based for reading purposes; I scribble enough formulae that it's just easier that way. But maybe someday I'll get a tablet...

Good talks can be useful, but I feel talks are mostly advertisements.. i.e. "My paper is nifty! you should give it more than a cursory glance!" while also (hopefully) conveying a rough idea of the content. But I'm just not going to get a strong detailed understanding from a talk.
posted by nat at 2:47 PM on March 3

Before entering my doctoral studies I used to be much less critical in my reading. To improve my critical thinking skills I created a template of categorized topics to pay attention to in every sociological theory or methods book I read. My questions include things like "What is the author's worldview?" "What is his/her theory of being- does s/he take a relative approach? A critical or materialistic approach?" And so on. When I'm done with the book, I can go to any section and see evidence that the author believes such and such about, say, the causes of social change, or about structure vs. agency issues. Every entry I make into my document refers to a page number in the book so that I can easily go back and read the idea or direct quote in context.

Since getting into the habit of using this template I have found myself much less reticent to contribute to class discussions (I'm usually very shy). When I write papers I can refer to particular categories for applicable quote. The only time I really write in a book rather than use my template is if I'm reading for my own enjoyment or preparing for a very cursory class discussion.

The upside of being anal enough to create an in-depth template is that years later those notes make it easy for me to write a summary. In fact, I created an annotated bibliography of quite a few books using my note system. When I taught, the categorized notes made choosing books for certain topics very simple.
posted by Piscean at 3:56 PM on March 3

PhD in history.

I have quite a rigid procedure that I usually follow:

1. Take a quick look at the title, contents page, index, preface, and introduction to see whether it's worth bothering with at all


2. Search for the author on Google, look for book reviews online


3. Use the index and contents page to identify the most interesting sections, and skim read these


4. Write down the page ranges of sections that do turn out to be useful. Jot down some keywords describing the content. I have a special plain text file on my computer and a section in my notebook to do this: both are called 'Misc Notes'.


5. If the piece has a basic level of worth, add the bibliographic information to my Zotero database


6. Review 'Misc Notes' and return to the text and make detailed notes. Begin with the key sections identified in Point 4, then read more broadly if necessary. Take as long as is required. Re-read multiple times if necessary.


I also use my digital camera to photograph key chapters and/or the page ranges identified at Point 4. This makes reviewing a lot easier. If there are only one or two interesting points in the whole book/article, then I'll probably just jot them down in my 'Misc Notes' file immediately: this saves time and paper.

posted by mattn at 3:57 PM on March 3 [1 favorite has favorites]

Two other threads with specifics of the kind you're after:
How do you read and process a large volume of academic reading?
How do you notate texts as you read them?
posted by LobsterMitten at 4:01 PM on March 3 [3 favorites has favorites]

PhD, Cog Systems Eng/Human factors
We do mission critical systems, and thus tend to be a cautious group with rather sensitive BS detectors for straw man arguments... tend to use this general approach when scanning cognitive psych as well as cs/engineering research:

methods
results
intro
discussion
posted by oldefortran at 4:06 PM on March 3

Political Science, here, as well. When I was studying for my comprehensive exams, I started a system of notes that I still do now if I consider the book/article to be something I want to remember. I have about ten binders of summaries that are now really helpful (esp. for teaching). If I want to understand a book or an article (sometimes even just a paragraph), I have to read it more than once. Sure, time consuming - but as an old professor of mine once told me, someone may have put years into writing that sentence - who are you to think you can get it in 5 minutes? For notes, the minimum that I include is a thesis (usually a sentence summary), main argument (a bit longer), and then key points (could be empirical data, a summary of a chapter, a definition, etc - highly depends on the work). Sometimes if the book/article is really a classic or essential to my field, I'll go more in depth, trying to relate it to other books/arguments or talk about its contribution. Usually for these "great" books, I'll look for reviews of these books in the major journals to skim through, or if the work has been around enough - I'll see who is citing it and why. I usually spend alot of time on the citations of the original work, actually. Sometimes when I sit down to read one thing - I end up with a list of 30 new things to read.
posted by quodlibet at 4:57 PM on March 3

PhD in Cognitive Psych
When I pick up a new issue of a journal, I'll skim the titles, read each of the abstracts and generally make a go-no go decision on spending any additional time on an article (if I'm on the fence, I'll skim the figures to see if there's anything interesting).

For articles that I read, I'll typically just read straight through (skipping parts of the intro if I know the topic very well). If I find it worth referring back to, I'll add it to my Zotero database with a note describing the interesting points.

When I was in school, the most effective thing was to write up an outline of the article. I only did this when I was part of a study group, and typically only for the classics in my field. We'd then present our summaries to the rest of the group. Although it was time consuming, it was remarkably effective.

In general, I am more likely to get something out of an article if I talk about it with someone -- even if I'm just telling them about it.
posted by i love cheese at 5:10 PM on March 3

Poli Sci here.

For articles, I always try and make sure I get the "puzzle" first. It takes a little bit of practice to get it, but it's usually in the first couple of paragraphs. Then I skip to the conclusion. With that done, I find it a lot easier to skim, since I have a sense of the context.

I'm also have my own library of papers that I think are/will be important/useful. I have a colored sticky note system, and I write the author's name and date on the sticky. Then I put it in a filing cabinet and add it to Zotero. I've killed a lot of trees, but nothing beats paper for ease of reading. Just make sure you have a double-sided printer and you buy recycled paper.
posted by awenner at 6:48 PM on March 3

Ph.D. in Film/American Studies. When I was reading for my orals -- a book a day -- I kept a blank index card as a bookmark and took notes on that. That way I could take my reading anywhere and I forced myself to edit my notetaking drastically. My feeling was that, given the volume I was reading, there was no way I was going to remember more than a notecard's worth.

When I finished the book, I typed my notes into OmniOutliner (recommended) and compared my take on the book with book reviews in journals.

Now that I can read at a slightly more leisurely pace, I use a bookstand at my desk and take notes directly in OmniOutliner as I read. I keep bibliographic info in BibDesk and link my OmniOutliner notes directly to the reference.

I try to make a point of writing down references to books that sound interesting. Then I create bibliographic records for those works in BibDesk, along with related keywords and a note to remind myself where I heard about the book.

Whatever you do, I really recommend keeping an active bibliography -- whether in EndNote or BibDesk or RefWorks or Zotero or whatever -- of titles you read for class or on your own. If you're anything like me, you'll forget a lot of titles and authors, and you'll smack yourself for letting that knowledge escape.
posted by miriam at 6:48 AM on March 4

Biochemistry. It depends what I'm reading: if it's something not particularly related to my research (generally papers for journal club, random papers in my subfield, utterly unrelated bio and chem papers that I find interesting), I'll often read just the abstract, intro, and conclusion. If it's related to my research (either focusing on the same system or using techniques I need), I'll read it much more carefully, with a lot of attention to methods, error bars in figures, etc. I take notes on the paper itself, and will often revise those notes as I read the paper over again. If the paper's particularly important, I may actually save the annotated physical copy in a binder.

While I sometimes store the annotated paper, I always store a pdf version using Papers which is an incredibly awesome program. iTunes for scientific papers, more or less, allowing you to group papers by subject, sort them by title, journal, or authors, tag them with subjects, generate citations in a variety of formats, etc. (I'm not a shill; it's just been so much better using papers than trying to manually sort and rename all those sdarticle.pdf files, generate bibliographies, etc. It's also great for saving space - I don't necessarily need paper copies of every paper I've read, but it's nice to be able to easily find and refer back to them when necessary.
posted by ubersturm at 2:42 PM on March 4