Consider the factors that make political science, and thereby comparative politics, a science.
Identify and describe the steps used in the scientific method.
Introduction
When many people consider the field of science, they think of laboratories filled with clinicians in white lab coats, chemical experiments with bubbling vials, and vast chalkboards of mathematical equations. These images represent the "hard sciences," such as chemistry, mathematics, and physics, and those researching in the hard sciences work to advance scientific understanding in the natural or physical sciences. Yet the "soft sciences," such as psychology, sociology, and political science, are also science. Those researching in the soft sciences work to advance scientific understanding of human behavior, institutions, society, government, decision-making, and power. Both hard and soft sciences provide significant contributions to the world of scientific inquiry and rely on the scientific method, yet they face different challenges. Highlighting a major challenge in the soft sciences, Physicist Heinz Pagels (1988) refers to the social sciences, those that scientifically study human society and relationships, as the "sciences of complexity." Numerous factors influence social phenomena and, in many cases, experimentation is not possible; it is inconceivable, for instance, to randomly assign a dictator to some countries and a democratic leader to others and then to measure how outcomes differ among the two sets of cases.
Despite this difficulty, science is simply the systematic and organized approach to any area of inquiry that utilizes scientific methods to acquire and build a body of knowledge, and political scientists use scientific methods to gain knowledge. As defined in Chapter 1, "comparative politics is a subfield of political science that seeks to advance understanding of political structures around the world in an organized, methodological, and clear way." To this end, political scientists rely on the rules of scientific inquiry to conduct their research. In the sections below, we introduce how the principles of scientific methods apply to the study of comparative politics.
Why is Political Science a Science?
While ancient political thinkers like Aristotle, Plato, Socrates, and Thucydides provided observations on their political worlds, they did not always operate under a strictly scientific scope. In some instances, these thinkers used a normative approach, focusing on how the political world should be. However, the modern conception of political science is one that, like the other social sciences, follows the scientific method. It also largely engages in empirical research, analysisbased on experiment, experience, or observation (as opposed to normative analysis).
Summarizing the features of empirical scientific research within the field, Gary King, Robert Keohane, and Sidney Verba wrote Designing Social Inquiry: Scientific Inference in Qualitative Research (1994). Although much of the book discusses political science in relation to qualitative research methods, which we will discuss later in this chapter, it also considers the general approach to scientific research in political science.
Figure \(\PageIndex{1}\): Book cover of Designing Social Inquiry: Scientific Inference in Qualitative Research, by Gary King, Robert O. Keohane, and Sidney Verba. (Source: King G., & Keohane, R.O., & Verba, S. (1994). Designing Social Inquiry: Scientific Inference in Qualitative Research. Princeton University Press.)
According to King, Keohane, and Verba (1994), scientific research has four main characteristics. First, one of the primary purposes of scientific research is to make descriptive or causal inferences. An inference is a process of drawing a conclusion about an unobserved phenomenon, based on observed (empirical) information. A descriptive inference summarizes observed data while a causal inference identifies the causes of observed effects. The accumulation of facts, by itself, does not make an effort scientific; this is true no matter how systematically a researcher collects the facts. In order for a study to be scientific, it requires the additional step of going beyond the immediately observable information in an effort to learn about something broader that is not directly observable. The process of making inferences can help us learn about the unobserved facts by describing them based on empirical information. For example, while we cannot directly observe the concept of "democracy," political scientists identify various characteristics of democratic states, which we can then use to describe and measure the concept.
Second, the procedures of scientific research must be public. Scientific research relies on "explicit, codified, and public methods" so that others can effectively assess the validity (accuracy) and reliability (consistency) of a study's results. As a condition for publication, publishers often require authors to share data files to ensure that other researchers can replicate the results, thereby allowing for ongoing assessment.
Third, because the process of making inferences is imperfect, the conclusions of scientific research are also uncertain. Researchers must therefore be aware of a reasonable estimate of uncertainty in their work to ensure that they can appropriately interpret, and qualify as necessary, their conclusions.
Fourth, scientific research depends on the method, not the content. This means that the manner of conduct, rather than the subject matter of study, is what determines whether a researcher's work is scientific. Scientific research must adhere to a set of rules related to inference, which relies on an understanding of the scientific method.
The Scientific Method
The scientific method follows a sequence of steps to standardize the research process: asking a research question, conducting background research, developing hypotheses, testing hypotheses, analyzing results, and reporting findings. Below, we discuss each step.
Step One: Asking a Research Question
The first step in the research process is to develop a clear, focused, and relevant research question. What characteristics define a good research question in political science?
First, the research question must be relevant to the real political world. This does not mean that a research question must address current political affairs; in fact, many political scientists study historical events and past political behaviors. However, the results of political science research are often relevant to the current political environment and may come with policy implications. Second, the question must be "falsifiable." Falsifiability, a term coined by Karl Popper, is the ability for a statement to be logically contradicted through empirical testing. Some research questions or claims are inherently non-falsifiable, meaning the question cannot be proven false. This occurs when questions are subjective (e.g., "Are oranges better than lemons?") and when technical limitations exist that make gathering evidence impossible (e.g., "Do angry ninja-robots live in Alpha Centauri?").
Suppose that a political scientist started with the following research question: "Is Argentina or Brazil better?" This question is subjective. What does "better" mean? This will vary according to an individual's taste. To refine this research question, the researcher will need to define "better." For example, the researcher may alter the question to the following: "Is Argentina or Brazil more economically productive?" This is a question on which a researcher can gather evidence. The researcher will describe metrics for what shows "economic productivity," then gather data and compare across the two countries.
Now consider a political scientist who asks: "Does investing heavily in a country's education system always mean that country will eventually become democratic?" This question has two technical limitation problems. First, the blanket statement that "investing heavily in a country's education system always leads to democracy" implies that a researcher will have to test every instance in which a country invests in education and democracy occurs (in addition to defining what "heavily" means). Second, and related, the word "eventually" implies that a country that invests heavily in education today may become democratic 500, 700, or 900 years from now. It is impossible to gather the evidence needed to test this research question.
Step Two: Conducting Background Research
The second step in the research process is to conduct background research. Before formulating hypotheses or gathering any data, researchers need to complete a "literature review" to situate themselves within the field. A literature review is a survey and synthesis of existing scholarly research that discusses prior research findings, best practices in methodology, existing theories, and remaining limitations related to the question of interest.
A literature review may take various forms, such as chronological, thematic, methods-focused, or theory-focused. Depending on the research question, a researcher may find that it makes the most sense to arrange the research in chronological order, beginning with the early research and culminating in the most recent research on a topic. Or, a researcher may find that previous scholars' works naturally fall into themes or categories (e.g., scholars A, B, and C argue that democracy leads to economic growth; scholars X, Y, and Z argue that economic growth leads to democracy). Or, a researcher may find that the most interesting part of previous research relates to the varying methods or theories used to explain the topic and therefore choose to organize the literature review accordingly.
No matter the form it takes, conducting background research to create a literature review is a key step in the research process. It allows a scholar to showcase, summarize, and synthesize previous research, and it also allows a researcher to demonstrate why their specific research question deserves attention.
Step Three: Developing Hypotheses
The third step in the research process is to develop a hypothesis, or a set of hypotheses. A hypothesis is a specific and testable prediction about the relationship between variables. A variable is any factor that can vary (i.e., take on different values). In a hypothesis, the researcher will identify a minimum of two types of variables: (1) an independent variable and (2) a dependent variable. In a cause-and-effect relationship, an independent variable (also called an "explanatory variable") is the presumed cause. A dependent variable (also called an "outcome variable") is the assumed effect. The independent variable leads to a change in the value of the dependent variable; in other words, the dependent variable's value depends on the changes in the independent variable. While a hypothesis is a testable prediction, a theory is a statement that explains how the world works based on experience and observation. An existing theory will often help a researcher build their hypotheses and, over time, repeated testing of hypotheses may lead to a new theory.
To develop, and later test (step four), a hypothesis, researchers may use either inductive or deductive reasoning. Inductive reasoning occurs when researchers make specific observations and attempt to form a hypothesis (and eventually a broader theory). Deductive reasoning occurs when researchers start with a broad theory and then test it with specific observations. While inductive reasoning uses a bottom-up approach (using specific observations to develop general theories), deductive reasoning uses a top-down approach (starting with general theories and testing with specific observations).
Steps Four and Five: Testing Hypotheses and Analyzing Results
The fourth and fifth steps in the research process work in tandem: testing hypotheses and analyzing results. The precise manner in which a researcher tests hypotheses depends on the research method that they employ (we will expand on types of research methods in the following section), but all testing requires a researcher to make observations of the relationship between the variables in the hypothesis.
When the testing is complete, researchers analyze their results and draw conclusions about the findings. Do the results provide evidence against the hypothesis? If not, this provides some evidence in support of the hypothesis and suggests the researcher may be on the right track. If so, the researcher will need to go back to the drawing board, develop a new hypothesis, and conduct testing again.
Step Six: Reporting Findings
The sixth step in the research process is to report findings. While not all research will result in academic publication, researchers can still share findings through research conferences, books, articles, or digital media. This information-sharing helps others to provide feedback and conduct their own research on the topic. It can also spawn new and interesting directions of research for the future.
Why is reporting findings so important? Consider the difference between two pandemics.
During the Spanish Flu, a pandemic in 1918, many countries of the world did not have freedom of press. In the midst of a hindered press and the lack of freedom of speech, doctors were unable to communicate their ideas and treatment plans for handling flu patients. Inundated with swarms of patients and flummoxed by the nature of a flu that was killing young, healthy adults, but largely sparing older individuals, doctors tried many treatment methods, yet were unable to broadly share their results.
In contrast, during COVID-19, many doctors worldwide worked on treatment plans, and were able to share their ideas on how to best treat patients. As one example, while there was a heavy reliance on ventilators initially, in time, doctors found that repositioning patients on their stomachs provided more time for patients to recover prior to resorting to a ventilator.
This contrast illustrates how the sharing of results can be critical to learning about a research question. If researchers are unable to share what they learn, it can stall the advancement of knowledge.
