Welcome | Arlington & Fairfax VA | Schar School of Policy ...



Political Science PhD Comprehensive ExaminationJanuary 2016American Politics In this exam, you must answer three questions, two from Part I and one from Part II. In each answer, you should develop a coherent argument in response to the question; your answer should critically engage and analyze, rather than simply review the literature. In selecting questions to answer, you should use this as an opportunity to demonstrate the breadth and depth of your understanding of multiple literatures, as well as of major theoretical issues, scholars, and methodological approaches. Part I: General American PoliticsAnswer TWO of the following four questions.In the United States, we expect presidents to provide leadership. What does “leadership” mean with respect to the presidency? Write an essay that considers the nature of presidential leadership in relation to the public, Congress and the bureaucracy. Consider the ways in which presidents lead, the meaning of leadership and the effectiveness of presidential leadership as well as the conditions under which presidential leadership is strengthened or weakened. Finally, are leadership styles time-bound, or are they primarily determined by opportunities and constraints?In a number of areas – like basic lawmaking, the initiation and conduct of military action, and agency appointments and judicial selection – more than one branch of the federal government is involved. How have political scientists sought to model and analyze areas of cross-branch interaction? Distinguishing by area where appropriate, identify (a) how the Constitution provides multiple branches of government with power/authority and (b) how political scientists have attempted to incorporate that power/authority into their models and analyses. Does one branch of government seem to have a power advantage – how and why? And is that power advantage absolute or conditional? Are there broad historical trends in the relative power of the three branches?Mayhew’s Congress: The Electoral Connection (1974) appears on virtually all graduate syllabi in American Government. Discuss and evaluate the principal components of his analysis as it pertains to legislative behavior and the organization of Congress as an institution. Which features of his argument have been confirmed by other scholars of TURN OVER→→legislative behavior and congressional organization? Which have not? How does Mayhew’s work relate to the most important recent Congressional development: the partisan polarization that has increasingly characterized Congress since at least 1994? Does his model help us understand this development (and if so how?), or does this development call for a reassessment of Mayhew’s model?Legal scholars maintain that the Supreme Court is a unique institution in American government because justices base their decisions on neutral principles of law. Many political scientists counter that justices are influenced not only by the law, but also by their own policy views, attitudes and personal background, and the behavior of their fellow justices. Evaluate the influence of these competing forces on Supreme Court decision-making. What theoretical and empirical evidence is there to support these different claims? What do you think best explains decision-making by Supreme Court justices?Part II: MethodsAnswer ONE of the following two questions.Select an issue in American politics that is, in your judgment, understudied or studded with contradictory findings. Describe the current state of knowledge, and explain why you believe that the issue/question deserves examination. Prepare a research design to address the question, providing solid theoretical and methodological foundations. Your essay must demonstrate an understanding of the importance of good research design, variable specification (operationalization), and the use of the appropriate analytical tools. You should also demonstrate a grasp of what data might reasonably be expected to be available on this topic, and how they might be deployed to answer your question.In a forthcoming paper, Alexander Bolton and Sharece Thrower argue that presidents should be most likely to use executive orders to circumvent Congress when there is divided government (“Legislative Capacity and Executive Unilateralism” American Journal of Political Science). Yet, they aver that this effect should be conditional on Congress’s (in)capacity to contest such unilateral presidential action. The authors operationalize congressional capacity in a number of ways, including measuring the aggregate legislative expenditures and committee staff. The unit of analysis is the year. The dependent variable for all analyses is the raw count of the number of non-ceremonial executive orders issued by the president in each year. Divided Government is an indicator variable describing when the party controlling the White House faces at least one house controlled by the opposition party. Legislative Expenditures and Committee Staff are both logged, but are not centered on their means. The authors present their results in the following table: Why do the authors use negative binomial regression rather than, say, ordinary least squares regression?Trend is a linear time trend variable. Is this a significant determinant of executive unilateralism? If so, what does this mean, substantively?What are the “constitutive terms” and the “interaction terms” across the two models (ignore the variables from “Inflation” on down)?What do the positive and statistically significant coefficients on Divided Government tell us? Should the authors interpret these results as support for their conditional hypothesis?Overall, do the results support the authors’ hypothesis regarding the conditional effect of divided government? Why or why not?TURN OVER→→Say that you download the replication data for this paper and want to assess whether Democratic presidents issue more executive orders than Republican ones. You create a variable for Democratic President, but your statistical software will not estimate an effect for this variable. Why?The authors chose to measure the dependent variable broadly by including all non-ceremonial executive orders. Other work in this field (e.g., William Howell’s Power without Persuasion book) argues that we should more narrowly consider significant executive orders. How might this coding decision have affected Bolton and Thrower’s findings? ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download