Political Science Scope and Methods

Calendar and Assignments

WEEK #	TOPICS	ASSIGNMENTS
Section I: Method, Causal Inference, and Research Design
1	The Scientific Method as Applied to Political Outcomes In class: Introductions, review of the syllabus, discussion of what constitutes political science, discussion of ways to test a potential claim (from a prospective student thesis or on candidate appearance).	N/A
2	Theory and Method in Political Science In class: Research design and examples of good work, discussion of limits to scientific method, accumulation of knowledge, and “theory decay curves”.	One-page essay: Is a “science” of politics possible? Desirable? Desirable or not, what would such a “science” look like?   (Make sure to review the Hints for Writing Papers)
3	Experimental Work and Its Challenges In class: Student presentations, experimental social science and its discontents, discussion of readings.	Read one of the following articles and be prepared to comment in class on your selection in light of the readings listed in Session 3 of the table in the Readings section: Blair, Graeme, C. Christine Fair, Neil Malhotra, et al. “Poverty and Support for Militant Politics: Evidence from Pakistan.” American Journal of Political Science 57 (2013): 30–48. Habyarimana, James, Macartan Humphreys, Daniel N. Posner, et al. “Why Does Ethnic Diversity Undermine Public Goods Provision?: An Experimental Approach.” (PDF) June 2006. Your presentation should be 5 minutes long; it should summarize the article — its argument, its method, its empirical evidence — and offer your critique (which should be informed by the other readings). You will be strictly timed, and you should make use of visual aids; make sure to review the Hints for Making Presentations. Your slides should be submitted to the instructor by 4 PM the day before class.
4	Observational Data and Its Challenges In class: Discussion of measurement, discussion of causal heterogeneity in Fearon & Laitin article, discussion of causal inference and Brady’s critique of Lott, discussion of mechanisms and examples of misinterpretation of data.	One-page: Where did Lott go wrong? What specific things besides those mentioned by Brady probably threw off his analysis? What more could he have done to get the story right?
5	Cases Selection and Its Challenges In class: Discussion of Pape paper. Discussion of Axelrod. Discussion of proposed research designs.	Approximately one page: Pick between 1–3 possible thesis topics, posed as researchable questions, and a very basic research design for each, to be discussed in class. Be sure to email the instructor the proposed topics by 4 PM the day before class, along with as much of the proposed research design as you wish to submit. You should at least be clear what the dependent variable is (that is, the thing you want to explain).  Pass human subjects training at Committee on the Use of Humans as Experimental Subjects (COUHES) and submit a pdf of your certificate
6	Evaluating Individual Studies In class: Presentation of peer-reviews. Subjection of the readings to an exegesis normally reserved for the denser passages of King Lear. Discussion of assignment for Week 7.	Write two peer reviews, one on each of the articles listed in Week 6 of the table in the Readings section. Your peer reviews should each be about two pages in length. As background, a referee’s report normally contains the following elements: An overall assessment  A brief summary of the authors’ claims and evidence, to prove that the reviewer understands them  A mention of the article’s strong points A critique of its weak points In practice, as you may discover in the future, (3) is sometimes omitted. The overall assessment in a review typically takes one of four forms: reject, revise and resubmit with no guarantee of ultimate acceptance, a favorable revise and resubmit, or publish as is. (In practice, very few articles receive a “publish as is”.) Assessments are normally conservative, in order to encourage authors to submit their best work the first time around. Thus, an article that has significant problems would normally receive a rejection, even if these problems could potentially be fixed. Come to class prepared to present the results of your peer-reviews.
7	Evaluating a (Small) Body of Work: The Effects of Candidate Appearance In class: Discussion of candidate appearance studies (first hour). Introduction to Qualtrics (second hour). In preparation for your assignments over the following two weeks and in Week 12, you will gain familiarity with the survey software Qualtrics. As examples, we will preview the candidate appearance and immigration experiments. Challenge: How to Rate 608 Candidates (PDF)	Three pages: Assume that the articles you read this week, plus the Todorov article from Week 6, constitute a “literature”. (They are actually only a subset of one part of the literature on candidate appearance – the part on facial appearance – but pretend otherwise.) Now write a three-page review of this literature, based on what findings and flaws there might be in the individual articles, what tensions there might be among the articles, and what gaps there might be between the articles. What do scholars know about the effects of facial appearance on election outcomes? What do they not know? What studies are needed to resolve the debates? This review should not be a separate critique of each article but rather a holistic assessment of what the literature tells us.
Section II: Collecting Data
8	Conducting an Experiment (or Generating Quantitative Observational Data), Part I In class: Discussion of ethics in experimental contexts. Introduction to Qualtrics (continued).	Work on project — continue coding surveys in Qualtrics. Run test surveys with each other, look for errors, timing, etc.
9	Conducting an Experiment (or Generating Quantitative Observational Data), Part II In class: Work on project.	Work on project (continued).
10	Conducting Field Research, Part I In class: Discussion of field work challenges and stories from the field, discussion of how “scientific” field work can be, ethics in participant-observation.	Two pages: Imagine that you were going to do field work on a topic of your choice. If you do not have a thesis topic in mind, consider (a) a study intended to assess racial disparities on the use of force by police, per the Fryer paper from Week 6, (b) a study intended to assess how politicians’ physical appearance might matter, per the research in Week 7, or (c) a study of Cuban legislative elections, per the project for Weeks 8–9. Write a full COUHES proposal for your project and send it to the instructor; do not send it to COUHES.
11	Conducting Field Research, Part II: Interviews In class: Interview role-playing, ethics in field work.	Rewrite your two-page paper from Week 10, turning it into a six-page document that includes an interview guide.
12	Student Presentations of Experimental Findings	Working with the TAs, prepare and present the results of your group project in a 10-minute presentation. You will be strictly timed, and you should make use of visual aids. Your PowerPoint slides should be submitted to the Instructor by 4 PM the day before class. Be sure to review the Hints for Making Presentations.
13	Using Archives, Libraries, and the Like In class: Introduction to library resources and “treasure hunt” by Jennifer Greenleaf, Social Science and Management Librarian, MIT Libraries, discussion of electronic “archives” on police shootings, discussion of student data requests, discussion of content analysis.	Make a Freedom of Information request for a topic of your choice, for data that you have concluded are not already publicly available. Pick a coroner’s office or a police department and find out what information they collect on fatal shootings and what is reported to the federal government.
14	Picking Research Questions In class: Discussion of tradeoffs in choice of topic, informal student presentations of their questions. Is a science of politics possible (redux)?	Pick between 1–3 possible thesis topics (presented as researchable questions) to discuss in class. Make sure to email these to the class ahead of time. They can be quite rough, but there should be an identifiable dependent variable and one or more identifiable independent variables.