ask a scatterbrain: how do you do a lit review

From a lurker: something we all have to do but are rarely taught about is constructing a literature review. Any strategies out there? Our reader provides some VERY HELPFUL strategies (after the break). But further suggestions from you would would be much appreciated! For me, one of the central questions here is not just how to do this, but how to decide what literature is the best literature to situate your project within. This isn’t always as obvious as it seems it should be. My advice: when doing a lit review for a project, it is often very helpful to know what your data say. I often write introductions last, after all the other work is done. So projects often start with lit reviews. But in my experience these are radically re-worked once you’ve gotten your hands dirty with the data. More practical suggestions after the break:

1. ISI Citation Databases/Social Science Citation Index
-If you aren’t already familiar with this then you need to be!  Extremely useful for finding key articles because of the ability to sort by number of citations.  This isn’t a sure-fire way to find the most important articles so don’t just use number of citations but it pretty much always leads you to some key articles and from there you can quickly get to others.

2. Google Scholar
-Also can be good for articles but I use it primarily for books.  Particularly nice for books because of the google books project that let’s you flip through some of the book to get a sense of whether or not it might be useful before you run to the library or purchase it.

3. Comp reading lists from our department and other top departments
-A good place to find an introduction to a general field.  It will give you a sense of some of the most important articles and the general topics for the field and you can dig deeper from there.  Some times reading lists will be fairly comprehensive, though they rarely have the latest “cutting-edge” research.

4. Syllabi from key people in the field or really respected institutions
-Works similarly to the comp reading list though it may be better in some instances.  Of course, if you’re looking for syllabi based on the key people in the field this requires you to know who these key people are in the first place.  If it’s a fresh syllabus then it will often have newer research.  Depending on the time put into the syllabus and the level of detail it may also give you a much better sense of how the field is loosely organized.

5. Annual Reviews
-Not surprisingly, these tend to have diminishing returns as they age.  If there’s one that’s recent and for your specific interest then these are often money.  If the article is less recent but not especially “old” it should still give you a nice framework upon which to build.  If it is older then it can still be useful though.  Some times the key features of a debate last a long time, often debates are cyclical, and, if nothing else, they can give you a bit of a history lesson to help you understand where the current literature is coming from.

15 thoughts on “ask a scatterbrain: how do you do a lit review”

  1. these tend to have diminishing returns as they age

    Don’t we all.

    A meta-suggestion from Murray Davis about the point of doing lit reviews in the first place: “Most social researchers do not clearly understand that the purpose of their review of the literature is to articulate the assumptions of their audience, and not merely to fulfil a rhetorical ritual. Nor do they clearly understand that the purpose of the rest of their research presentation is to increase our interest by refuting these assumptions, and not merely to “increase our knowledge” by confirming or ignoring them.”


  2. Reviewing the literature and writing the literature review are not the same thing. Shamus’s suggestions seem to be more about how to find what to read. Kieran’s comment is more relevant to the writing.
    The important thing to understand is that you don’t put everything you know or have read into a literature review. This is worth saying twice.

    Scholars can make two kinds of mistakes about reading. One is to think you have to read everything before you can write anything. This is the path to paralysis because, of course, you cannot read everything. You cannot even read everything that is potentially relevant to your project, no matter what your project is. You have to be selective and stop reading to write. The other mistake is to think that nobody is as smart as you, so why bother reading, and to view the necessity of a literature review as an empty ritual of deference forced upon you by egoistic elders. (Although this is primarily a disease of the young, some people in their 60s still haven’t figured out that there are other smart people in the world.) A variant on type 2 is to see yourself as part a partisan political/academic tendency and read only things by your friends or the people on your intellectual team.

    You take courses and do generalized reading so you’ll have a general broad background to draw from when thinking about problems. This is the kind of stuff you get from course syllabi and from a general habit of looking over the journals. Most of this isn’t going to make it into a lit review. (A lit review is not a prelim.) You do more specialized focused reading to find out about other people’s empirical or theoretical work that might help you do your own work better. Here you are doing a much more focused search and course syllabi are not likely to be of much help, except for telling you what “everyone” will have read in an area.

    Having wandered into a field I had no background in, I’ve been doing a LOT of reading in the past few years. I have found the appropriate Annual Review articles to be very helpful. I have also done keyword searches in appropriate library databases and Google Scholar searches. As I moved into a new area, I had to learn new databases to find literature that wasn’t in Sociological Abstracts. As I pursued a topic, I’d realize there was something else I needed to look up. A lot of the best references come from reading one good piece and then looking up the stuff that seems important that is mentioned there. It is important to actually look it up: I learned early in my career that people are OFTEN misquoted. Frequently the original source has a lot of good material not mentioned by the citing source, and also frequently the original source either does not say what the citing source says it said, or is actually methodologically incompetent and shouldn’t be cited unless it is also criticized. The hardest part is some system for organizing and keeping track of sources. Lots of time you read something that isn’t relevant to what you need then, but later you remember it when it is relevant, and it is nice not to have to spend two days tracking it down again.

    The writing is an entirely different matter. Ideally you avoid having a section called “literature review.” Instead, you are summarizing the key theoretical ideas that have been used to address your problem in the past, or that have not been used that you say should be used. And you are summarizing the past empirical literature: what has been found and what methods did they use and what the research any good or not? Where will your research fit in? Is it one more study that adds a bit more information to a particular well-defined vein of research? You’ll write the review one way. Are you trying to launch and justify a whole new line of research that nobody else has done before? You’ll write it another way. Are you writing a conceptual/theoretical piece that reviews old ways of thinking about a problem and offers some new twist on how to think? Yet another way of writing the review.

    As Kieran’s comment suggests, part of the writing is to think about the audience. What literature will THEY think is relevant to the problem? You do have to do something with that literature, either to integrate it or to explain why it only seems relevant but is not. The problem of gratuitous citations to major figures in the field appears here in the sub-problem of meeting audience expectations.


  3. Ideally you avoid having a section called “literature review.”

    I think this is a pretty easy ideal to reach. I don’t think one should ever have a section in a paper called “Literature Review”. Even more, I don’t think one should even refer to the “Background” section of a paper as the lit review as it encourages the wrong mindset.


  4. Shamus: Jeremy can speak for himself but I think he and Kieran and I are saying the same thing in different ways. “Reviewing the literature” as a reading task is one thing and it is important. But what you do for that and the notes you take are not the same as writing a paper. In writing the front end of a paper, you are explaining why your problem is important, what previous research/thinking has been done about that problem, and what your contribution will be. I think Jeremy’s point is that calling that section of the paper “literature review” or “background” gives you the wrong mindset, as it it makes you think of summarizing everything you know, instead of developing the argument about the problem at hand and what the contribution of your work is. Edit: But sometimes the first draft of the front end may be what is essentially a literature review, as you summarize what has been done on the way to rewriting it as an argument about what should be done next.


  5. Ideally you avoid having a section called “literature review.”

    Completely agreed! Thank you for mentioning it. Whenever I teach this point – which is pretty much in every class I teach – students will then point to lots and lots of examples of people having such a heading in their papers. I then have to explain that just because lots of people do it doesn’t make it a good thing.


  6. I totally agree with everyone – reading the literature & writing about it are entirely different. As I tell my students, the front of your paper should serve an instrumental purpose, to set up your argument and demonstrate your contribution. It is not some ritualistic performance, where you try to pay homage to everything that is possibly related to your paper that has come before.


  7. OW: thanks. I’m not trying to be cute. I just think that for many people hearing, “of course you’re not supposed to do a lit review section” will be more confusing then enlightening. Because much of getting feedback on papers consists of “you didn’t reference the right literature” or “your framing doesn’t fit your data” or “you’re not setting this up right.” Which usually means the student goes back and does more literature review.

    So I simply ask the dumb question perhaps others aren’t willing to ask, “if no literature review, then what?”

    Let me say that from my point of view, I fully agree with OW and Kieran. It’s about “developing the argument about the problem at hand and what the contribution of your work is” and not referencing everything that’s been said on the topic.


  8. (Incidentally, I didn’t mean my comment above to say that papers shouldn’t have a section titled “Background.” I use that myself. I just meant to say papers should have a section titled “Literature Review” or “Literature” or “Review of the Literature, which at least to my ear/mind if very different from “Background.”)


  9. I almost forgot about a great reference here: Howard S. Becker, “Terrorized by the Literature,” Chapter 8, Writing for Social Scientists, U Chicago, 1986. Becker’s entire book is worth a read.


  10. Great summary (olderwoman’s postings especially appreciated, thank you). For more detail, and many examples, may I reccommend:
    Hart, Chris (1998) Doing a Literature Review: Releasing the social science research imagination. London: SAGE.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.