Writing a Literature Review: Advice for Political Science Senior Theses

No man is an island, and no thesis is wholly original. By writing a thesis, you are engaging in a conversation with those who have written on your topic and topics that you will claim (and establish!) are closely related to what you are studying. However, you are also trying to show that you are adding something to the debate. Consequently, your goal in the literature review section is to show that you

  • have identified, read, and analyzed relevant arguments
  • understand and respect the contributions of prior authors
  • know how they (or at least some subset of ‘they’) got it wrong

These goals conflict! “I respect you, but you are wrong” is a tough line to follow. I’ll discuss strategies below to address that.

Throughout, though, I want you to remember:

  • the literature review is a supporting actor. The star is your thesis (and its accompanying theory and research design).
  • you need to straddle humility and confidence in demonstrating your contribution.

This page supplements resources such as the UNC Writing Center’s guide to writing literature reviews and Dr. Raul Pacheco-Vega’s work (especially on using Excel to structure lit reviews!), which you should also read.

Cautionary Note

Literature reviews carry with them a dangerous seed. Reading and analyzing what other people have done feels like work. Thus, people often engage in it in the mistaken belief that it is work. But make no mistake: literature reviews do not constitute research–they form part of the pre-writing process. They are separate from the process of original thinking and testing that you want to carry out. They are, in other words, mere labor, not work. (I make a special exemption for the kinds of high-level literature reviews that get published in venues like the Annual Review of Political Science, which are actually theoretical essays masquerading as reviews of the earlier literature, but that distinction is irrelevant for you right now.) Incidentally, this means that most ‘research’ reports you have written until this point were literature reviews, not ‘research’ in the sense that we mean it here. Your goal is to produce original knowledge not to recapitulate what others have written.

You don’t want to get trapped in a cycle of reading an article, taking notes on it, following a citation trail to new articles, and then reading, note-taking, and following up on those articles. You need to read with a purpose–and write your assessments fairly and ruthlessly. Remember: every moment that you spend on a literature review that you don’t need to is a moment that you’re not spending on your own work! (This balance is difficult to strike, but so often the temptation for folks is to spend too much time on the comfortable business of repeating others’ findings rather than striking out to try to establish some new argument’s validity.)


Your goal is to show

  • what the major debates related to your topic have been
  • what different groups of scholars have shown (almost always, debates cluster around different groupings or ‘schools’ of scholars engaging in systematic debates, rather than a bunch of individuals spouting off)
  • what the key claims of each are

Among the most efficient ways of doing this happens to be the first two paragraphs of Liou & Musgrave (2014):

That’s it — two schools, key claims, and a bunch of citations that show who believes what. This is in some ways overly simple (we combined those who think a curse exists but is conditional and those who think that resources are a blessing)–but our point was to get to the identification argument in the second paragraph (shown) as fast as possible.

A more useful (and really excellent) guide comes from the first four pages of Haber & Menaldo (2011), which I reproduce only in part. Since much of their introduction covers the “literature” clearly and concisely, I focus here on their “Literature Review”, which they use to claim that everyone else has been using poor methods:

Note that Haber & Menaldo briskly address each article/book’s contribution, the advantages it has, and the flaws with their approach. Since their goal here is to show that their data and methodological approach is superior (the same as Liou & Musgrave, above), they focus their attention on methods. However, you may find that you are critiquing the data used by each author, the theoretical approaches that earlier authors have used, or the framing of the puzzle that has guided these debates.

Regardless, the key point is that you are not summarizing literature to summarize it. You are seeking to show what the contours of a debate is and emphasizing the elements of the debate you will be contributing to. If you think everyone’s conclusions rests on the inclusion of a case that should be excluded, then you would want to show why the decision to include it was made (perhaps in error, perhaps by inattention) and what effects that had; if you want to show that new methods would fix earlier analyses, you want to establish that other folks used different methods and that those methods drove their results; if you think that the question you want to resolve is a debate between two seemingly equal valid schools, you need to establish exactly how those schools differ and set up the research design that will let you see which (if any) of the existing explanations are correct.

Mechanics of Finding Articles and Books to Cite

There is a rough hierarchy of sources that you should be citing:

  1. Peer-reviewed books published by university presses
  2. Peer-reviewed articles published in scholarly journals
  3. Specialist publications and technical assessments published by government agencies and consultancies
  4. Non-specialist (non-peer reviewed) journal articles and books (e.g., a Foreign Affairs article or a book by a commercial press, like Simon & Schuster or Farrar Straus Giroux)
  5. Expert analyses on the Internet (e.g. The Monkey Cage)
  6. Newspaper articles
  7. Other sources

This is rough! But note that you want to spend more time with academic sources than non-academic ones. How do you find them? Ordinarily (something like 95% of the time!), the answer is “Google Scholar“. You can visit it directly, but more often the best route is to get there via the UMASS Library’s web page:

And start searching for relevant keywords there.

You should be creative in looking for keywords, by the way: not just “resource curse” but also “rentier state”. The Google folks have established a nice page of search tips that you should use.

But finding (good, peer-reviewed) articles is only the first step. The next step to begin looking at their bibliographies to see what they’ve cited–and then using Google Scholar to see who has cited the article you began with.

For instance, if we look for Michael Ross’s original 2001 World Politics article on the resource curse, we find a link to it:

Clicking on “Cited by 2863” (the number will change based on how many scholars have cited it) brings you to:

Each of these sources have engaged with Ross–and some of these articles are fabulously well-cited themselves. Going through this process can help you identify more recent works, including those that extend or revise the findings of the original article, and also locate new links between your thesis and other debates.

(You will spend a lot of time on this.)

Mechanics of Citing the Articles You Find

You should never be typing in bibliography information (at least, not for journal articles)! That’s the whole point of RefWorks! See here for advice about Google Scholar and RefWorks; I find that the automatically generated Google Scholar citations occasionally have errors or weird formatting quirks, however. Consequently, I try, whenever possible, to get my bibliography entries from the Web site of the journal publisher (or database, like JSTOR) directly. For Ross’s article, for instance, I click on the Full Text via UMLinks link and go to JSTOR, where I see

Clicking on the “Cite This Item” button yields

And the “Export to RefWorks” button will, in fact, export the citation to RefWorks.

Sometimes this is slightly hidden, especially for articles accessible only through journal Web sites; in that case, you have to look for a menu command that is something like “Export Citation”, “Reference”, or a similar name. For Cambridge University Press journals (as of 2017), for instance:

you would click on “Export Citation” and this would then bring up a pop-up:

Clicking on “Export” yields

And once again you would “Export to … RefWorks”.

Leave a Reply

Your email address will not be published. Required fields are marked *