Political Science Dissertation Examples

Chapter 3, "Research Design" (page 39), is the model for a research design chapter that argues rather than describes. Jo opens by setting out what existing scholarship claims: that national unity or civic culture underpins democratic institutions, that narratives are path-dependent. Then she states her disagreement and what follows from it.

The design exists to test a claim against rivals, and the reader can see which rivals. Compare it with the Dyzenhaus entry: both are chapter 3, but his is procedural (which case, why, how far it generalizes) and hers is argumentative (what the field believes, why it is wrong, what this study does instead). Different fields want different weightings of the two, and having both here is deliberate, so read whichever your supervisor keeps asking for. It is also a usable model for a paired comparison, South Korea and Taiwan.

The shortest political science entry and the most useful if you are proposing a mixed-methods design and have been asked to justify it. Scott uses archival data, closed-ended survey questions and semi-structured interviews, and the justification is one sentence doing real work: the research question "is so analytically complex in that no one method alone can sufficiently answer it", followed by which method answers which part.

Most mixed-methods justifications amount to "using both gives a fuller picture", which is not an argument; this one ties each method to a specific sub-question, which is. It also carries an explicit Hypotheses section ahead of "Data and Methods", so you can see hypotheses stated before the data is described, the order that makes a quantitative chapter readable, and the one students most often invert.

Read this for Appendices A and B (from page 222), eighty-odd pages of screen captures from Parler, thedonald.win and 8chan, and for the paragraph at the head of each explaining the citation system: every capture has a figure number, its source platform and a describing clause; footnotes in the body cite the appendix location; longer multi-page exchanges live in named supplemental files using the same convention. This is the problem nobody teaches and a great many students now have.

Your primary sources are posts on platforms that deplatform, delete and disappear. Parler was taken offline weeks after some of this material was captured, so a URL in a footnote is worthless. Travis's answer is to capture, number, describe and cross-reference, so the evidence travels with the dissertation and an examiner can check every claim. If your data is social media or anything else ephemeral, look at this before you design your citation scheme rather than after.

The best answer in this collection to "it's only one case, so what can you generalize?", the question that stops more qualitative students than any other. Read chapter 3 from page 44. Dyzenhaus does three things in order.

He shows how the literature has treated South Africa: excluded from Africa-wide studies, paired with Brazil or the US, occasionally a single case. Then he says which camp he is in. He states the characteristics that define his case rather than the country: historical rural land inequality, state-led land reform, reform through market means. Then he defines the population of cases his theory should travel to, in three widening circles: minority settler colonies, majority settler colonies making amends to a dispossessed indigenous minority, and any state addressing rural land inequality by market redistribution, with named examples including US reparations to Black farmers, the Scottish Highland clearances and India's colonial land legacies. That is scope conditions, written out.

The only work of pure political theory in the collection, and it is here because theory students have almost nothing to model themselves on. There is no methods chapter, no data section, no participants.

That is not an omission; it is what the genre is. What stands in for a methods chapter is an explicit statement of interpretive approach: Mares invokes Collingwood's principle that to understand a proposition you must recover the question it was intended to answer, and that therefore "history precedes logic". The dissertation then does exactly that, tracing how Francis Deng's anthropological work on the Dinka of Sudan fed into the doctrine of the Responsibility to Protect, and refusing the easier available story, that R2P simply grew out of 1970s human rights discourse, because the intellectual history does not support it. Read it if you are writing conceptual work and every example you have been handed is empirical.