Fall 2023 Aaron Sparks

**POL 2200 - Research Methods**

**Prompt for Research paper**

**The following is an outline to guide the development of your research paper. You do not need to follow this formula exactly, as long as you think about and have justification for doing it differently. Throughout the problem sets, we will be progressing on each section, so that by the end of the semester you will need to compile the various sections, expand where necessary, and dress it all-up into a complete and coherent paper. Always keep in mind the four causal hurdles and if your paper can pass them and why or why not. Transparency is better than finding the results you hoped for.**

**You are using the literature and your own analysis to write the paper. For the literature, expect to cite 10-15 peer-reviewed sources. The analysis includes the appropriate bivariate hypothesis test and multiple regression.**

**Use first person and active voice! You are telling a story. It is not a mystery novel, so don’t hide the results until the end.**

** **

**The paper (80%)**

**1. Abstract**

**a. An abstract should introduce the paper and clearly state the findings. Why should a reader want to continue reading? What puzzle does the paper solve? What is the theory/argument? What are the results? What do the results mean for political science research?**

**2. Introduction (Some lit review, and overview of methods and findings)**

**This section can be fairly short, more of an overview of the paper.**

**a. Introduce your dependent variable – without talking specifically about your data or how you operationalize the concept (2-3 paragraphs)**

** i. What is it?**

** ii. Why should we care about it?**

** iii. What do we know already about it?**

** iv. What is something interesting we might want to know more about? (This is the hook, the puzzle, the reason why you chose this theory, the motivation for your research etc…)**

**b. Introduce the Independent variable (2-3 paragraphs)**

** i. What do we know about it already?**

** ii. Why does it cause your dependent variable (credible causal mechanism)?**

** iii. How does this relationship solve the puzzle introduced above?**

**c. How do you test the relationship and what do you find (one paragraph)?**

** i. Data used**

** ii. Type of analysis conducted (eg: multiple regression)**

** iii. Interpretation of results**

**3. Theory (Literature review) (5-7 paragraphs)**

**In this section, make the argument, using the literature, for why and how X causes Y. The theory section expands on some of the points made in the introduction and goes into more depth. Keep your analysis at the conceptual (not operational) level.**

** i. What do we know about the causes of the Dependent variable?**

** ii. Why does the independent variable cause, the dependent variable?**

**1. Explain the causal mechanism – the story you tell for how X causes Y.**

** iii. What are other key factors that may influence Y?**

**1. What are the alternative explanations of Y?**

**2. How could they be accounted for?**

** iv. Think about how you can integrate a causal diagram ( X --> Y) into this section**

**4. Methods (6-8 paragraphs)**

**a. Data Source**

**b. Description of sample (or population if your data include the universe of cases)**

** i. Textual description and summary table (along with data source, one paragraph)**

**c. Operationalization of DV, IV, and controls (short paragraph for each, with clear subsection headings for DV, IV, and one for control variables)**

** i. Specifically, how are each measured (eg: verbatim survey question and response categories)**

** ii. Mean and standard deviation of all ordinal and continuous variables**

** iii. Mode and frequency for categorical variables**

** iv. Report each statistic M(SD) when you discuss each variable and include a summary table**

**d. Hypothesis**

** i. Ceteris paribus, an increase in X is associated with an increase/decrease in Y.**

** ii. You may have more than one hypothesis, if you have more than one operationalization of the DV.**

**5. Results**

**a. Talk about what type of statistical model you employ (either a correlation or T-test to test the bivariate relationship, and an OLS multiple regression model).**

**b. Include the figures and tables in the text.**

**c. Explain the statistical results – what does it mean if relationship is negative or positive, significant or not?**

** i. In multiple regression, this means explaining the betas for each variable in the model.**

**d. Can you reject the null hypothesis?**

**6. Discussion Conclusion**

**a. Explain in more depth and breadth the results of statistical tests. This means going beyond the “1 unit increase in x, is associated with a beta change in y). Explain what it means in the real world.**

**b. Draw conclusions about what these results mean for your theory**

** i. Why was it supported or not?**

**c. What are the limitations of the study?**

**d. What are the next steps going forward?**

** **

**Presentation (20%)**

**· 5 minute presentation during the last week of classes**

**· Motivate the research question**

**· Explain the theory**

**· Outline your methods**

**o Data source**

**o Analysis type**

**· Discuss results**

**o Include figures**

- Last Updated: Oct 3, 2023 11:12 AM
- URL: https://elon.libguides.com/POL2200
- Print Page