With wit and practical wisdom, Christopher Howard draws on more than a decade of experience teaching research methods to transform a typically dreary subject and teach budding political scientists the critical skills they need to read published research more effectively and produce better research of their own. The first part of the book is devoted to asking three fundamental questions in political science: What happened? Why? Who cares? In the second section, Howard demonstrates how to answer these questions by choosing an appropriate research design, selecting cases, and working with numbers and written documents as evidence. Drawing on examples from American and comparative politics, international relations, and public policy, Thinking Like a Political Scientist highlights the most common challenges that political scientists routinely face, and each chapter concludes with exercises so that students can practice dealing with those challenges.
|Publisher:||University of Chicago Press|
|Series:||Chicago Guides to Writing, Editing, and Publishing|
|Sold by:||Barnes & Noble|
|File size:||711 KB|
About the Author
Read an Excerpt
Thinking Like a Political Scientist
A Practical Guide to Research Methods
By Christopher Howard
The University of Chicago PressCopyright © 2017 The University of Chicago
All rights reserved.
Imagine that someone starts talking to you, at length, about bananas. About how they cost more at Kroger than at Walmart; then again, the ones at Walmart sometimes have more brown spots on the outside, and sometimes those spots mean a bruise but, you know, not always. About how bananas are a great source of potassium, but lima beans and acorn squash are even better sources. This person asks you, "Isn't it odd that grocery stores will let their customers shuck fresh corn yet won't let them peel bananas (or grapefruit, for that matter) before they check out? Did you know that monkeys peel bananas from the bottom, while humans do it from the top?" If this person happens to be anyone that you truly care about, then you may nod your head and try to look interested. With just about anyone else, your eyes will start to glaze over, and the voice inside your head will be screaming, "WHY are you WASTING my TIME?! Why should I CARE?"
When we talk or write, it's easy to assume that our audience will be fascinated by what we say. That assumption is perfectly reasonable for those of us who are under the age of eight, or who are dealing with people who love us unconditionally. Otherwise, it makes more sense to assume that our audience is really busy and doesn't share our particular passion. To truly connect with people, we need to persuade them to stop, listen, and engage.
Granted, the example of bananas is vivid but trivial. Suppose instead that you are writing a college paper about democratization in Eastern Europe or land reform in Brazil. Or suppose that you work at a "real" job analyzing environmental regulations, military budgets, court cases, medical costs, or consumer behavior. No matter what the issue, almost everyone you hope to communicate with is feeling inundated, maybe even overwhelmed, with information. In any given day at the office, people (and that includes college instructors) deal with dozens and dozens of e-mails, phone calls, memos, reports, meetings, and conversations with coworkers. While some of these messages are expected, many are not. A significant number of these exchanges are distracting or pointless. In addition, apart from work, the average American received an estimated 74 gigabytes of data from television, radio, cell phones, personal computers, and the like in 2015 — every single day. That's the equivalent of nine DVDs of words, sounds, and images, every single day. The amount of data aimed at us is staggering, literally staggering, and it keeps growing.
To keep our heads from exploding, we've developed a number of coping strategies. Much of this information is simply ignored. Other times, we quickly skim a document, or read the first lines carefully and then skim. One recent study found that most people stop reading the average web article when they are halfway through — and halfway means only a few paragraphs. Many items are mentally filed under "If I Have Time Later ...," but that time rarely comes. Not when there are deadlines to meet, groceries to buy, clothes to wash, and ballgames to attend. Clearly, anyone trying to send a message, and have it heard, will find the odds stacked against them.
Many students, however, don't realize this. For years they have inhabited this strange world where their written work has been guaranteed one reader — their teacher. After all, it was the teacher who had created the assignment in the first place, and that teacher is paid to read and comment on whatever the student writes. Students write for their teacher. That mindset works fine if you plan to take orders from one person for the rest of your life. It doesn't work well anywhere else. Ideally, students will start to view research and writing less as an exercise in power (or, more accurately, powerlessness) and more as an opportunity for persuasion.
Professional scholars realize how difficult it is to capture a reader's attention. They understand that motivating their readers is important, which is why they usually answer the Who Cares / So What question early in their article, chapter, or book. Before they describe what happened and explain why, social scientists have to generate interest in their work. Otherwise, people will stop reading and find something better to do. Later, in the conclusion, these authors usually highlight the larger significance of their research findings. Generally speaking, answers to the Who Cares question appear at the beginning and the end of a project, while answers to what happened and why occupy the middle.
As careful readers, students need to recognize how authors frame the Who Cares question. Scholars employ a variety of techniques to establish the larger significance of their research, and I will discuss several of them in this chapter. Choosing among these techniques depends in part on the intended audience, which can range from a small group of academic specialists to the general public. Knowing these strategies will also help students when they conduct their own research. In particular, it will help them make that crucial leap from finding a topic that interests them to identifying a specific research question that will interest many others.
Joining a Conversation
My brother is a high school English teacher, and a few years ago he introduced me to a terrific book about writing called They Say, I Say. One of the authors' core insights is that all writers, no matter what their topic, are part of some larger conversation that extends across time. Writing is both a solitary activity and a collective enterprise. Except for personal diaries, we write to communicate with other people.
For many scholars, these conversations take place primarily with other academics. This is especially true of writing that takes the form of articles appearing in scholarly journals or books published by university presses. No one expects the average person to read the latest issue of the American Political Science Review or International Organization. Most of the books published by Oxford University Press and the University Press of Kansas never reach the shelves of our local bookstore. Yet professional scholars know that these kinds of journals and books are where they can go to join a good conversation. As a practical matter, many scholars also realize that contributing to these conversations is necessary if they want to earn tenure and promotion (and maybe even honor and respect) within their profession.
These scholarly conversations can be quite disjointed and hard to follow. They involve people who might never meet face to face. Participants often want to shift the conversation in new directions. And these conversations could persist, off and on, for years or even decades. It is therefore essential that scholars clearly indicate which conversation they are trying to join. They typically do this early in their written work, and they do it fairly quickly, because they realize that few readers will want to spend much time rehashing old arguments. This part of the project is called the literature review (better known as the lit review), which has nothing to do with reading Charles Dickens or Toni Morrison but everything to do with joining and contributing to a scholarly conversation. In the lit review, scholars summarize what is already known about some event or pattern in the political world. Most lit reviews also indicate what we don't know, and thus how additional research could enrich our collective understanding of that event or pattern. To return to the house metaphor introduced earlier, a lit review provides the foundation on which the rest of the scholarly edifice is built.
How scholars design their literature review tells us a lot about the intended audience for their work (Who) and the potential significance of their research (Cares). Readers should pay close attention to which literatures are being invoked. Someone analyzing the outcome of the 2008 US presidential election, for example, could start by engaging with past studies of the presidential campaigns in that year. After digging through articles and books, he or she might identify a group of analysts who believe that Barack Obama won because his campaign did a better job of mobilizing new voters, and another group believing that John McCain lost because he blundered in choosing Sarah Palin as his running mate. A broader literature review might encompass multiple US elections before 2008, with a focus on the role of economic conditions. These elections could be presidential, congressional, or both. Maybe the 2008 election needs to be examined against a larger pattern in which incumbent parties do poorly when the economy falters.
In principle, there's no reason why the 2008 election has to be analyzed strictly from the perspective of American politics. Perhaps the larger question is how members of marginalized groups manage to overcome historical obstacles and win national elections. Obama was, after all, the first African American to win the US presidency. From this angle, previous studies of female prime ministers (e.g., in Israel, England, India, Mozambique) could be relevant. The target audience would now include scholars of comparative politics, not just American politics. There are undoubtedly many other ways to use the 2008 election to further a scholarly conversation. Same case, different literatures. Nevertheless, it is hard to imagine any single book or article embracing all these literatures; a certain degree of analytic focus is important and usually expected.
Unless our collective knowledge of a topic is quite limited, it will be practically impossible for scholars to cite, much less summarize, every previous study in their lit review. (Nor should they try before verifying that their health insurer offers a full range of mental health benefits.) Scholars will need to generalize, they will need to condense, and they will need to discuss a number of studies while leaving some out. Most lit reviews are therefore organized around schools of thought rather than individual authors or studies. The two most common designs for a literature review represent some variant of the Conventional Wisdom and the Ongoing Debate.
With some questions, most political scientists agree on what happened or why. Among specialists in international relations, for instance, the conventional wisdom is that democracies rarely go to war with each other, hence the "democratic peace." Although democracies certainly go to war with nondemocracies (think World War II), and they regularly spy on other democracies, full-fledged war between democracies is rare. In comparative politics, the conventional wisdom is that electoral systems based on proportional representation give third parties more voice in government than do systems based on single-member districts. Both of these generalizations describe important patterns in politics. A review of either literature could cite multiple published studies and historical examples.
Conventional Wisdoms describing what happened in politics are probably more common than Conventional Wisdoms explaining why, but the latter do exist. Most specialists in American politics would agree on some of the reasons why public trust in government declined after the 1960s. According to the American National Election Studies, three-quarters of Americans in 1964 felt they could trust the national government most of the time or almost always. By 1980, that figure had dropped to one-quarter. Why did that happen? The conventional wisdom is that the Vietnam War and the Watergate scandal diminished trust in government. As a second example, experts generally agree that party identification plays a larger role than specific issues (e.g., abortion) or candidate traits (e.g., intelligence, honesty) in determining how people vote in US national elections. For most Americans, knowing that Obama was a Democrat and McCain a Republican provided important information in deciding whom to vote for in 2008.
On most questions, though, political scientists actively disagree. Sometimes these Ongoing Debates divide scholars into two main camps — not necessarily equal camps, but each side big enough to sustain a meaningful debate. Any question involving the impact of something — such as UN peacekeeping missions, school reforms, gender quotas in national legislatures, economic sanctions against "rogue" nations — can pit those who claim the effort was effective versus those who say it was ineffective. To determine impact, we must figure out what happened. Theoretical debates, centered on the Why question, are very common in political science. A literature review about inequality might feature two camps: those who argue that the roots are primarily economic, and those who say the roots are political. Similarly, studies of international migration might pit those who stress "push" factors versus those emphasizing "pull" factors as the main causes. In these examples, a literature review would summarize the basic logic and some of the key evidence for each side.
Such debates can easily expand to include more than two sides. A debate over the effectiveness of some agency or program might boil down to those who believe it is usually, sometimes, or seldom effective. A debate over the causes of inequality might feature economic, political, and cultural arguments. Introductory textbooks in international relations often frame the main theoretical debates as a contest among realism, liberalism, and constructivism. Three sides is by no means the upper limit. Over the years, scholars have offered many different reasons why the American welfare state seems to lag behind its European counterparts, including highly fragmented institutions; core values of individualism and limited government; the absence of strong left-wing political parties and labor unions; the power of business interests; and the historical importance of racial divisions.
One function of the literature review, then, is to signal which conversation the author hopes to join. Authors say, in effect, "I'm talking specifically to these political scientists who generally agree about X." Or, "I'd like to weigh in on this argument between these two groups of scholars over Y." But that's not enough. Authors also need to set up the Conventional Wisdom or the Ongoing Debate in such a way that they can contribute something new. This step is crucial. A professional literature review does more than simply show that the author has read the classic studies in her field. It does more than provide accurate summaries of those studies. A strong lit review helps readers recognize potential gaps and misunderstandings in our collective knowledge of some part of the political world. It makes readers feel that they want to learn more. Who + Cares. The rest of the paper, article, chapter, or book will start to fill in these gaps or correct these misunderstandings.
In more formal terms, a good lit review generates one or more hypotheses that the author plans to investigate more fully. A hypothesis is an educated guess, in this instance an educated guess about some feature of the political world, whether that's a single event such as the 2008 US election or a general pattern such as the democratic peace. A hypothesis is "educated" in the sense that it's informed by prior research or experience. It is a "guess" in the sense that it could turn out to be true or false, depending on the evidence. For instance, given what scholars have learned about US elections in the twentieth century, we would expect economic factors to have played a role in the 2008 election — but we won't know for sure until we research this case more thoroughly.
Anyone trying to contribute to a conversation has a few options, and the same is true for political scientists. Broadly speaking, research tries to extend, refine, or challenge what we (think we) know about the political world. This is admittedly a simple list of options, and the choices aren't mutually exclusive, but as an initial framework it will do the job. Imagine, for example, that you were conducting research in the year 2000 on the so-called resource curse in comparative politics. After reading numerous studies, you would probably notice a piece of conventional wisdom: countries that rely heavily on oil for revenue often suffer from poor economic growth. They also tend not to be strong, healthy democracies. An abundance of oil turns out to be a curse. So far, this is a descriptive hypothesis; it indicates a relationship without explaining how or why oil impedes development.
Excerpted from Thinking Like a Political Scientist by Christopher Howard. Copyright © 2017 The University of Chicago. Excerpted by permission of The University of Chicago Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of ContentsPreface for Students
Preface for Teachers
Part I. Asking Good Questions
1. Who Cares?
2. What Happened?
Part II. Generating Good Answers
4. Choosing a Research Design
5. Choosing Cases
6. Using Documents as Evidence
7. Using Numbers as Evidence