The Nature of Political Inquiry
 

As we begin our discussion of political science research methods is important to keep in mind that we are in fact studying social science and the scientific method.  We might ask, ourselves in what way is political science a science?  And, if my assertion that political science is a science is correct.  We must differentiate between politics and political science.

Politics can be understood as a competition between citizens over policy preference.  In my public policy classes I define public policy as a public response to perceived public problems.  Politics is the process we use to select those public responses.  There is nothing scientific about it.  It's all interest-based.  Political science, on the other hand, is the social scientific study of politics and public policy.   

What is scientific about political science?  The section below is an excerpt from an excellent handout created by the University of North Carolina.

 

 

What is scientific about political science?
Prepared by The Writing Center, University of North Carolina at Chapel Hill

Although political scientists are prone to debate and disagreement, the majority view the discipline as a genuine science. As a result, political scientists generally strive to emulate the objectivity as well as the conceptual and methodological rigor typically associated with the so-called "hard" sciences (e.g., biology, chemistry, and physics). They see themselves engaged in revealing the relationships underlying political events and conditions. And from these revelations they attempt to construct general principles about the way the world of politics works. Given these aims, it is important for political scientists' writing to be conceptually precise, free from bias, and well-substantiated by empirical evidence. They want to build and refine ever more precise and persuasive theories. Knowing that political scientists value objectivity may help you in making decisions about how to write your paper and what to put in it. Political theory is an important exception to this empirical approach. You can learn more about writing for political theory classes in the section "Writing in Political Theory" below.

Since theory-building serves as the cornerstone of the discipline, it may be useful to see how it works. You too may be wrestling with theories or proposing your own as you write your paper. Consider how political scientists have arrived at the theories you are reading and discussing in your course. Most political scientists adhere to a simple model of scientific inquiry when building theories. The key to building precise and persuasive theories is to develop and test hypotheses. Hypotheses are statements that researchers construct for the purpose of testing whether or not a certain relationship exists between two phenomena. To see how political scientists use hypotheses, and to imagine how you might use a hypothesis to develop a thesis for your paper, consider the following example. Suppose that we want to know if presidential elections are affected by economic conditions. We could formulate this question into the following hypothesis: "When the national unemployment rate is greater than 7 percent at the time of the election, presidential incumbents are not reelected."

In the research model designed to test this hypothesis, the dependent variable, or the phenomenon that is affected by other variables, would be the reelection of incumbent presidents; the independent variable, or the phenomenon that may have some effect on the dependent variable, would be the national unemployment rate. You could test the relationship between the independent and dependent variables by collecting data on unemployment rates and the reelection of incumbent presidents and comparing the two sets of information. If you found that in every instance that the national unemployment rate was greater than 7 percent at the time of a presidential election the incumbent lost, you would have significant support for our hypothesis.

However, research in political science seldom yields immediately conclusive results. In this case, for example, although in most recent presidential elections our hypothesis holds true, President Franklin Roosevelt was reelected in 1936 despite the fact that the national unemployment rate was 17%. To explain this important exception and to make certain that other factors besides high unemployment rates were not primarily responsible for the defeat of incumbent presidents in other election years, you would need to do further research. So you can see how political scientists use the scientific method to build ever more precise and persuasive theories and how you might begin to think about and analyze the topics that interest you as your write your paper.

Since political scientists construct and assess theories in accordance with the principles of the scientific method, writing in the field conveys the rigor, objectivity, and logical consistency that characterize this method. Thus, in contrast to scholars in such fields as literature, art history or classics, political scientists avoid the use of impressionistic or metaphorical language, or language which appeals primarily to our senses, emotions, or moral beliefs. In other words, rather than persuade you with the elegance of their prose or the moral virtue of their beliefs, political scientists persuade through their command of the facts and their ability to relate those facts to theories that can withstand the test of empirical investigation. In writing of this sort, clarity and concision are at a premium. To achieve such clarity and concision, political scientists precisely define any terms or concepts that are important to the arguments that they make. This precision often requires that they "operationalize" key terms or concepts, which simply means that they define them so that they can be measured or tested through scientific investigation.

Fortunately, you will generally not be expected to devise or operationalize key concepts entirely on your own. In most cases, your professor or the authors of assigned readings will already have defined and/or operationalized concepts that are important to your research. And in the event that someone hasn't already come up with precisely the definition you need, other political scientists will in all likelihood have written enough on the topic that you're investigating to give you some clear guidance on how to proceed. For this reason, it is always a good idea to explore what research has already been done on your topic before you begin to construct your own argument. (See our handout on making an academic argument.)

To give you an example of the kind of "rigor" and "objectivity" political scientists aim for in their writing, let's examine how someone might operationalize a term. Reading through this example should clarify the level of analysis and precision that you will be expected to employ in your writing. Here's how you might define key concepts in a way that allows us to measure them.

We are all familiar with the term "democracy." If you were asked to define the term, you might make a statement like the following: "Democracy is government by the people." You would, of course, be correct--democracy is government by the people. But, in order to evaluate whether or not a particular government is fully democratic or is more or less democratic when compared with other governments, we would need to have more precise criteria with which to measure or assess democracy. Most political scientists agree that these criteria should include the following rights and freedoms for citizens:

1.     Freedom to form and join organizations

2.     Freedom of expression

3.     Right to vote

4.     Eligibility for public office

5.     Right of political leaders to compete for support

6.     Right of political leaders to compete for votes

7.     Alternative sources of information

8.     Free and fair elections

9.     Institutions for making government policies depend on votes and other expressions of preference

By adopting these nine criteria, we now have a definition that will allow us to measure democracy. Thus, if you want to determine whether or not Brazil is more democratic than Sweden, you can evaluate each country in terms of the degree to which they fulfill the above criteria.   

(source:  http://www.unc.edu/depts/wcweb/handouts/polisci.html)

 

 

So, if political science is in fact science, what should we be concerned with?  As the excerpt above suggests, science is concerned with making empirical observations.  Empirical observations are based on making observations based on our empirical senses -- what can we see, what can we hear, what can we feel, what can we taste, what can we smell.   Empirical observations and bring us empirical knowledge -- which is distinct from normative knowledge. 

Normative knowledge is based on what we believe -- influenced by personal experience, by personal values.  Empirical observation allows us to document what is.  Normative knowledge allows us to assess whether what is is good or bad.  For example, political science frequently studies elections.  We can observe empirically how many people vote.  But whether 50% turnout is good or bad is based on normative judgment.

To be fair, we should note that political science includes normative philosophy.  Our subfield of political theory is concerned with normative questions -- like justice.  At the same time, the “science” of political research is empirically driven. 

 

Epistemology

But, what is knowledge, and how to we build it?  Epistemology is the study of the nature of knowledge.  Our interest in Epistemology revolves around the question, how do we know what we know? 

The Greek tradition argues that knowledge is found at the intersection of truth and belief (as the Venn diagram illustrates).  I suppose one could argue that “truths” are simply components of empirical knowledge (e.g., what we can observe as empirically true), and “beliefs” are simply components of normative knowledge (e.g., what we believe to be true).

Take a look at this videoclip.  As you see, we can argue over what is and what isn't until the end of time.  Is there a greater “truth”?  It all depends on whether or not we can agree on how we measure truth.  What we know as science today emerged from the Aristotelian desire to demonstrate truth using nature.  More to the point, science emerged as scholars sought to demonstrate that their theories were correct using in empirical observation.

As the section below illustrates, “science” emerged as a consensus protocol – as an agreement on the process that “scientists” would use to measure truth.



Kuhn’s Paradigm

In The Structure of Scientific Revolutions Thomas Kuhn argued that science is itself defined by the questions scientists ask.  These questions are invariably constrained by the beliefs scientists carry with them as a function of their training and accreditation.  These beliefs form the theories, which scientists and then seek to validate through empirical observation.  The bubble within which scientific inquiry takes place can be described as a paradigm.  A paradigm is a philosophical or ideological framework that defines our basic assumptions.  Consequently within any scientific paradigm we are able to validate many of our basic theories, but, according to Kuhn, we are unable to make any significant scientific discoveries. 

 

Kuhn argues that as we move through our scientific inquiry, validating some theories and invalidating others, we will eventually come to a point where our basic assumptions are no longer able to be sustained.  For example, if we are living within an “earth is flat” paradigm we will ask questions about the flat earth.  How flat is in the earth?   Why is the earth flat?   What happens when we come to the end of the earth – do we just fall off?  Perhaps we develop a theory stating that when we get to the edge of the earth the ocean drops in a huge waterfall.  As we seek to test this theory, we may find that we never in fact common to the edge of the earth.  If we never get to the edge of the Earth perhaps the earth is not flat.  This recognition shatters the earth is flat paradigm. 

 

Hegel argued that knowledge is the consequence conflict -- a dialectic process.  The first stage of the dialectic is the thesis (in Kuhn, the paradigm).  The thesis will eventually bring an opposing reaction – the antithesis (the paradigm will eventually be invalidated).  The tension between the thesis and antithesis à ß is eventually result through a synthesis.  But the synthesis in the dialectic is not an integration of the thesis and antithesis -- it is an entirely new understanding.  This is, in Kuhn, a scientific revolution.  Here is an excellent synopsis of Kuhn at http://www.des.emory.edu/mfp/kuhnsyn.html.

 

The Scientific Method

The term science is itself confusing.  When we talk about “science” we often mean the body of information that science has collected over time.  For example, in a geology class we may talk about earthquakes and plate tectonics.  Geologic discovery is indeed an important part of science.  But, scientific discovery is only possible as a consequence of using a scientific process.  I prefer to think of science as a collection of procedural tools we use to make empirical observations.  These processes can be thought of as the scientific method.  Science is process. 

We can understand the scientific method as a set of protocols that have developed over time to allow us to make unbiased observations of physical, biological, or social phenomenon.  Political science is a subfield of social science, just as physics is part of physical science, and biology a part of life science.  Though the unit of analysis – and the specific data collection tools used – may differ across these different sciences, the scientific method remains the same. 

The scientific method is a stepwise process which includes three major components:  science is based on making empirical observations; science is transparent; science is replicable.  When we say science is based on making empirical observations what we mean is that science is concerned with making observations of actual phenomenon, and documenting those observations.  Transparency simply means that the process we use, and the tools we use, must be transparent to our peers in the scientific community.  In other words, the scientific community at large must be able to see and understand what we actually did to make these observations.  And finally, our observations must be replicable.  Replicability is the cornerstone to good science.  If we make an observation and other scientists aren't able to replicate that observation is, it is assumed that our observation was either (a) a mistake on our part (we didn’t really see what we thought we did); or (b) an anomaly (an irregularity that we can't expect to see again).

The general process of scientific inquiry, consistent with the three principles above, include the following steps:

  1. articulating a clear and focused research question (RQ);
  2. researching the background of the RQ through an extensive literature review;
  3. conceptualizing a theory that explains a plausible answer to our research question – and constructing appropriate hypotheses to test the theory;
  4. test the hypotheses through empirical observation of the relationship between variables contained within the hypotheses;
  5. analyze the results of these tests and draw an appropriate conclusion;
  6. Report results;

 

Remember, science is a process of explaining the world using evidence.  If we can’t evidence to support our theory it is likely our theory is wrong.   The difference between political science and a cocktail party is that political science (like any science) uses evidence to support its theories. 

 

 

Identifying and Measuring Hypotheses

 

As the discussion above point out, scientific research is necessarily a stepwise process.  We are, by definition, at step one (developing the research question).  As you think about your research question, you will want to think about how you might answer the question.  To do that effectively we will conduct a thorough review of the scientific literature (step two), and using the knowledge of the literature review, we will break down our RQ into specific hypotheses (step three).  Hypotheses are simply an explanation of the relationship between two or more variables.  If my RQ is "To what extent does studying help students graduate?" I might conceptualize several hypotheses which can be tested to help me answer the RQ.  Hypothesis One (H1) might be:  When students study more they learn more.  Hypothesis Two (H2) might be:  When students learn more in earlier classes they do better in later classes.   The variables contained in H1 include: the dependent variable -- the time students put into studying in a particular class; and the independent variable -- the amount students learn.  The dependent variable causes change in the independent variable.  Increased study causes variation in the amount students learn.  H2 includes the variables amount learned (dependent) and how well students do in later classes (independent).  By measuring these variables we can test the hypotheses -- can the hypotheses be supported by the data? (step four).  And, by testing the hypotheses we can better determine the answer to the RQ (step five).  Don't worry about steps two through five at this point, as we will be revisiting them as we move forward.  At this point, think about your RQ.

 

 

 

How to Create a Useful Research Question

All research begins with a well articulated research question.  Think of your research question as your scalpel.  If you were surgeon your scalpel would be your most important tool.  It must be sharp, clean, and well focused.  If you try to operate with a rusty scalpel the outcome for your patient would not be very good.  Similarly, as a social scientist, if you use a rusty research question (ambiguous, unfocused, unmeasurable), the outcome for your paper will not be very good.  Trochim discusses three basic types of questions that research projects can address:

 

Descriptive:  When a study is designed primarily to describe what is going on or what exists. Public opinion polls that seek only to describe the proportion of people who hold various opinions are primarily descriptive in nature. For instance, if we want to know what percent of the population would vote for a Democratic or a Republican in the next presidential election, we are simply interested in describing something.

 

Relational: When a study is designed to look at the relationships between two or more variables. A public opinion poll that compares what proportion of males and females say they would vote for a Democratic or a Republican candidate in the next presidential election is essentially studying the relationship between gender and voting preference.

 

Causal:  When a study is designed to determine whether one or more variables (e.g., a program or treatment variable) causes or affects one or more outcome variables. If we did a public opinion poll to try to determine whether a recent political advertising campaign changed voter preferences, we would essentially be studying whether the campaign (cause) changed the proportion of voters who would vote Democratic or Republican (effect).

 

Your research question should be constructed in a way that allows you to measure something of importance.  It should be narrow and to the point.  For example, To what extent do Cars in the Los Angeles basin contribute to airborne pollution?   This question is specific – it identifies units of analysis (cars) as well as anticipates a meaningful hypothesis (cars à smog), and places the question in a geographically specific location make a measurement easier.  We could be even more specific: To what extent have Cars in the Los Angeles basin contributed to airborne pollution over the last ten years? 

As you think about articulating your specific research question start with a topic that you find interesting -- and mayor of the topic into a focused and specific question.  If I'm interested generally in the presidential election, I might ask myself “what about the election” is interesting?  If voter turnout is something I find interesting, I might ask myself “what influences voters to turn out or stay home on election day?” 

 

Keep in mind that a good research question is one which is both measurable (you can measure specific variables which will allow you to answer the question) and important.  When we say important we mean that a research question has high explanatory power.  We can't answer all questions – we sent five-time.  So, we try to articulate research questions that provide the most explanation on a given topic. 

Think about a topic that interests you, narrow that topic, and write down a draft Research Question.  Leave it for a few hours, then return and revise.  Leave it again, and revise.  The key here is in thinking broadly, then focusing, and then revising several times.  Later, as you begin researching your topic you will have the opportunity to refine your research question.