For decades, there has been a raging debate among scholars regarding the differences between and advantages of qualitative and quantitative methods. In fact, this has probably been one of the largest and longest methodological debates in all of social science research. Perhaps it can be briefly summarized by the following two famous and opposing quotations: Donald Campbell says, “All research ultimately has a qualitative grounding”; and Fred Kerlinger says, “There’s no such thing as qualitative data. Everything is either 1 or 0” (in Miles & Huberman, 1994, p. 40). Although it is not necessarily critical to determine which—if either—of these approaches can be described as the better one, it is imperative to have a thorough understanding of these methods in order to be able to conduct sound political science research. After all, for a study to be of value to scholars and other individuals interested in the topic, it is necessary for one to choose the correct research approach, ask suitable questions, use appropriate research methods and statistical analyses, correctly deduce or induce inferences, and have suitable general goals driving the research.
The questions under consideration and the answers obtained by any particular study will depend on whether the study uses quantitative or qualitative approaches. The purpose of this article is to differentiate between these two types of research. First, the literature available on this topic is briefly summarized, focusing specifically on how qualitative and quantitative research is defined, as well as the different assumptions on which these types of research are based. Next, a summary of the similarities and differences in each stage of the research process is provided. Then, the different methods that these two types of approaches use are discussed. Next, since this is a book examining political science in the 21st century, current and future research directions are examined. In particular, the use of what are called mixed methods approaches is discussed. The article ends with a brief summary and conclusion of the information that has been presented. Finally, suggested books and articles for further reading are provided, including some material for individuals interested in conducting advanced statistical studies, which are beyond the scope of this article.
- Quantitative and Qualitative Research
- Definition of Quantitative Research
- Assumptions of Quantitative Research
- Definition of Qualitative Research
- Assumptions of Qualitative Research
- Comparing and Contrasting Quantitative and Qualitative Research Methods
- The Research Question
- Research Design
- Data Collection
- Data Analysis
- Reporting of Results
- Quantitative Methods in Political Science
- Limitations of Quantitative Methods
- Qualitative Methods
- Limitations of Qualitative Methods
- Future Directions
Quantitative and Qualitative Research
The following section introduces the definitions and assumptions of quantitative and qualitative research. First, however, it is worth briefly discussing two types of political analysis in order to understand the origins of quantitative and qualitative methods. Political scientists distinguish between empirical analysis—obtaining and dealing with knowledge and information—and normative analysis— determining how to use that knowledge. Normative analysis relies on the development of subjective goals and values to apply what has been learned to reality. Empirical analysis, however, focuses on using common terms to explain and describe political reality and can be either quantitative or qualitative in nature. If something is empirical, it is verifiable through observations or experiments. Empirical analysis is the focus of this article.
Definition of Quantitative Research
As a first step, it is necessary to define these two methods of research and examine their goals. Quantitative research can be defined as a process of inquiry examining an identified problem that is based on testing a theory measured by numbers and analyzed with statistical techniques. Thus, quantitative research involves the analysis of numerical data. A more technical definition is provided by Brady and Collier (2004), who define mainstream quantitative methods as “an approach to methodology strongly oriented toward regression analysis, econometric refinements on regression, and the search for statistical alternatives to regression models in contexts where specific regression assumptions are not met” (p. 294). The econometric refinements and statistical alternatives referred to by the authors are beyond the scope of this article but include logit and probit models, time-series analysis, and a variety of techniques to circumvent problems that can occur in regression analysis, such as heteroskedasticity and autocorrelation. Essentially, quantitative methods have played a major role in improving on commonly used research tools within the structure of regression models that are frequently used in the field of political science.
Assumptions of Quantitative Research
The goal of quantitative research is to examine particular instances or aspects of phenomena to determine if predictive generalizations of a theory hold true or to test causal hypotheses. As a result, there are several key assumptions underlying quantitative research methods, which are briefly outlined here. These include the following:
- Reality can be studied objectively.
- Research must remain independent of the researcher through the use of experiments, questionnaires, machines, or inventories.
- Research is value free, and the researcher does not become a part of or interfere with the research.
- Theories and hypotheses are tested in a cause effect order with research based primarily on deductive forms of logic identified a priori by the researcher.
- The purpose of research is to develop generalizations that contribute to theory and allow the researcher to predict, explain, and understand a particular phenomenon.
Definition of Qualitative Research
Qualitative research can be defined as a process of inquiry that builds a complex and holistic picture of a particular phenomenon of interest by using a natural setting. Thus, qualitative research involves the analysis of words, pictures, videos, or objects in the context in which they occur.
Assumptions of Qualitative Research
The goal of qualitative research is to understand social issues from multiple perspectives to have a comprehensive understanding of a particular event, person, or group. As with quantitative research, there are several key assumptions underlying qualitative research methods:
- Reality is socially constructed, and there are multiple realities.
- The researcher interacts and often works closely with the individuals or groups under study and serves as the primary instrument for data collection and analysis.
- The research is value laden, and the researchers become a part of the research, attempting to understand the lives and experiences of the people they study.
- Research is context bound and based on inductive forms of logic that emerge as a study progresses.
- The purpose of research is to find theories that help explain a particular phenomenon.
Comparing and Contrasting Quantitative and Qualitative Research Methods
The following section examines how quantitative and qualitative methods are similar to and different from each other throughout the research process, beginning with the creation of a research question and up to the reporting of the results. Although examining quantitative and qualitative methods as two separate categories is necessary for the sake of clarification throughout this section, it is important to realize that these two methods are not mutually exclusive, a topic that will be discussed in more detail shortly. As Manheim, Rich, Willnat, and Brians (2007) note, when examining the differences between quantitative and qualitative methods, “The distinctions discussed are generally more matters of degree than absolutes. The two types of methods often require only different forms of work, but are working toward similar objectives” (p. 323). This is important to keep in mind while reading this article.
The Research Question
The first step in conducting sound political science research is selecting a research question. An appropriate research question should fulfill either a scientific need or a societal need by helping to provide an answer to an important problem. Both quantitative and qualitative forms of research begin by creating a research question that is intended to produce knowledge of the empirical world. In terms of the research questions, the main difference between quantitative and qualitative methods typically exists in the type of questions that are being posed.
A theory is a potential explanation for events and is composed of a set of logically related propositions and assumptions. Theorizing is the actual process of stating these conceptual explanations for events that take place in the real world by proclaiming relationships among the concepts. Theories are created to help people understand phenomena. There are several characteristics that make a theory particularly useful in explaining observations. Theories should be (a) testable, (b) logically sound, (c) communicable, (d) general, and (e) parsimonious.
Theorizing is a critical phase of the research process for quantitative and qualitative researchers. However, quantitative researchers are more likely than qualitative researchers to focus on testing performed theories. Quantitative researchers base their studies on a theory that relates to their subject in an attempt to develop generalizations that contribute to theory. Thus, in quantitative research, theorizing occurs prior to the collection of data. Qualitative researchers, on the other hand, are more likely than quantitative researchers to elaborate on theories while making observations of a particular phenomenon. Many qualitative researchers argue that, as a result of this, their theories are far more grounded in reality than are those of quantitative researchers. However, quantitative researchers argue that the formulation of theory during the observation-making process can easily lead to the creation of a theory designed around those specific observations. As a result, these theories would be polluted and not testable. Furthermore, if a theory is based on observation of one particular group, the usefulness of the theory is quite limited.
Simply defined, a research design is the plan of a study. It organizes observations in a manner that establishes a logical basis for causal inference. Essentially, the research design can be viewed as the blueprint for a study. There are three main types of research designs in political science: exploratory, descriptive, and explanatory. Exploratory research attempts to discover which factors should be included when theorizing about and researching a particular subject. Descriptive research attempts to measure some aspect of reality for its own sake and not for the purpose of developing or testing some theory. Explanatory research uses observations of reality to test hypotheses and help develop an understanding of patterns of behavior in the context of a specific theory.
Regardless of the purpose of a study, every research design should have the same basic elements, which are outlined by Manheim et al. (2007): (a) a statement outlining the purpose of the research; (b) a review of the theory and any hypotheses that are going to be tested, if applicable; (c) a statement explaining the variables that will be used; (d) an explanation of the operationalization and measurement of the variables; (e) a statement of how observations will be organized, as well as conducted; and (f) a discussion of how the data that are collected will be analyzed.
Although both quantitative and qualitative researchers produce research designs for their studies, quantitative researchers are much more likely than their counterparts to base their designs on the logic of experiments. For instance, quantitative researchers often emphasize control groups, pretests, and other elements that provide them with the opportunity to hold some factor(s) constant in their attempt to make causal inferences. Qualitative research designs, on the other hand, typically focus more on who or what is being observed, where the observation will take place, how observations will be conducted, and how the data will be recorded. For qualitative researchers, more emphasis is placed on viewing people and events as they naturally occur, while for quantitative researchers there is a greater focus on establishing cause-and-effect relationships.
A sample is a small group of cases drawn from and used to represent a larger population under consideration. A representative sample is a sample in which each major attribute of the larger population occurs in approximately the same proportion or frequency as in the larger population. “In other words, a truly representative sample is a microcosm—a smaller, but accurate model—of the larger population from which it is taken” (Manheim et al., 2007, p. 119). When a sample is representative, the conclusions drawn from it are generalizable to the entire population.
In quantitative studies, sampling is based on the logic of probability to produce statistical representativeness. Additionally, in quantitative research, sampling is done before the data are collected. Qualitative researchers, on the other hand, usually create their sample once their study is already in progress. After observing, learning about, and gaining understanding from an initial case, qualitative researchers are then able to determine what they will observe next. Additionally, whereas generalizability is a chief concern for quantitative researchers, this is not the case for qualitative researchers, who are far more concerned with finding the specific information that they are looking for from their sample. Since this method is very time-consuming, qualitative findings are often based on fewer cases than quantitative findings.
Data are observations or information about reality that represent attributes of variables and result from the research process. Although data collection is an integral part of both types of research methods, data are composed of words in qualitative research and numbers in quantitative research, which results in a data collection process that differs significantly for quantitative and qualitative research. Furthermore, the data collection process is different: Although quantitative researchers have the ability to administer a previously prepared questionnaire or watch an experiment unfold behind blind glass, qualitative researchers are engaged—sometimes for long periods of time—with the people or groups under study.
As can likely be seen by now, quantitative researchers frequently have a detailed plan of action that is thought out prior to the beginning of a study’s taking place. Qualitative researchers, on the other hand, tend to take a more fluid approach to their studies. This holds true for the analysis of data, as well. Whereas in quantitative studies, the data analysis methods are planned out in advance and then occur after the data are collected, data analysis typically takes place at the same time as data collection in qualitative studies. To make appropriate future observations, analyses must often begin after studying one to several initial cases. As a result, quantitative researchers are not usually afforded the opportunity to modify their methods of data collection during a project, while qualitative researchers can do so at any point in a project after conducting the initial data analysis.
Additionally, although qualitative data are more subjective and sometimes difficult to interpret, quantitative data are easily coded into numerical formats. As a result, it is much easier to enter quantitative data into computer programs, such as Excel and SPSS, than it is to enter qualitative data. Furthermore, there are a number of programs that analyze the statistical data, such as SPSS and Stata. Although programs do exist for the interpretation of qualitative data, they are not used nearly as extensively as those used for quantitative data analysis.
Finally, whereas quantitative researchers have a variety of means to test the statistical significance and validity of the data that they are analyzing, this is not the case for qualitative researchers. Instead, qualitative researchers must do their best to present a clear, accurate, and convincing analysis of their data. As a result, a topic of much debate between quantitative and qualitative researchers is the validity and reliability of findings produced in studies. Validity is the extent to which measures correspond to the concepts they are intended to reflect. Reliability is the consistency with which a measuring instrument allows assignment of values to cases when repeated over time. Although a measure can be reliable without being valid, it cannot be valid without being reliable.
Additionally, since one of the main points of conducting quantitative research is to study causal relationships, part of the process involves manipulating various factors that could potentially influence a phenomenon of interest while at the same time controlling for other variables that could affect the outcome. For instance, if a researcher were examining if gender played a role in whether a person received a job, it would be important to control for other variables, such as education or previous work experience, since these factors may also determine why an individual would receive an employment offer. In quantitative analysis, empirical relationships and associations are typically examined by using general linear models, nonlinear models, or factor analysis to understand important information about the relationship between variables, such as the direction of a relationship. However, despite the results that may be produced by these models, it is important to note that a major tenet of quantitative research is that correlation does not imply causation. In other words, a spurious relationship is always a possible result of the data analysis.
Reporting of Results
When presenting the results of a study, qualitative researchers often have an arduous task in front of them. Since their reports typically rely on the interpretation of observations, it is necessary for them to be very careful in the selection of what stories, quotations, pictures, and so on, they will share in order to avoid bias. The reports produced by quantitative researchers tend to be more straightforward since they rely mostly on the interpretation of statistics. But here, too, it is important to make sure that bias was avoided in the sample and that appropriate data analysis methods were used in order to avoid bias in quantitative analysis.
To sum up, there are a lot of similarities among quantitative and qualitative research methods. Irrespective of which method is used, it is still necessary to create an appropriate research question, understand the theory behind what will be observed, create a research design, collect and analyze data, and create a report of the results. However, there are several key differences between quantitative and qualitative research methods. These methods differ in (a) the types of questions that they pose, (b) their analytical objectives, (c) the amount of flexibility allowed in the research design, (d) the data collection instruments that are used, and (e) the type of data that are ultimately produced. According to Mack, Woodsong, MacQueen, Guest, and Namey (2005), the fifth difference is the biggest. The authors argue that quantitative methods are generally inflexible since categories are typically closed-ended or fixed, while qualitative methods are more flexible, with a large amount of spontaneity and adaptation occurring during interaction with other people, especially in the form of open-ended questions.
To decide which research approach should be used, several things should be taken into account, including the problem of interest, the resources available, the skills and training of the researcher(s), and the audience for the research. Since there are considerable differences in the assumptions that underlie these two research approaches, as well as the collection and analysis of data, these considerations are important. The following sections provide a more detailed examination of the various types of quantitative and qualitative research methods, as well as the limitations of these methods in general.
Quantitative Methods in Political Science
Quantitative methods are essentially a variety of research techniques that are used to gather quantitative data. There are a variety of different types of quantitative methods, which are briefly outlined in this section: experiments, quasi experiments, content analysis, and surveys. First, in experiments, participants are randomly assigned to experimental conditions, as well as experimental controls. The individuals who are assigned to experimental controls are testing the independent variable. The difference between experiments and quasi experiments is the way that subjects are selected. In quasi experiments, participants are assigned to experimental conditions in a nonrandom fashion.
Next, content analysis is a systematic means of counting and assessing information in order to interpret it. For instance, scholars may count the number of times that personal characteristics, such as dress or hairstyle, are mentioned in newspaper articles to determine whether media coverage of male and female candidates differs. Finally, surveys are used to estimate the characteristics of a population based on responses to questionnaires and interviews from a sample of the population. Surveys provide five types of information: (1) facts, (2) opinions, (3) perceptions, (4) attitudes, and (5) behavioral reports. Essentially, questionnaires and surveys can serve as a means for helping scholars understand why people feel or act the way that they do, as well as measure their attitudes and assess their behaviors.
Limitations of Quantitative Methods
There are three key criticisms of quantitative research that are discussed here. First, since quantitative research methods were adopted from the physical sciences, critics argue that all cases are treated as though they are alike. Complex concepts are turned into numbers, and their unique elements are dissipated as a result. Furthermore, people can easily attribute different meanings to something even when they are experiencing the same phenomena. Second, and related to the first criticism, some people argue that quantitative methods are inherently biased. Since they are adopted from the physical sciences, critics argue that quantitative methods fail to take into account the unique cultural roots and other critical aspects of marginalized groups of people. Thus, according to critics, when it comes to populations that have been politically excluded, the usage of quantitative methods may not be appropriate, according to critics. Third, critics argue that quantitative research methods result in taking individuals out of their natural settings to examine very limited aspects of what a person thinks or believes. To these critics, context is very important, and by taking actions out of context, it is impossible to understand the true meaning of events or responses.
Qualitative Methods in Political Science
Just as quantitative research methods have a variety of research techniques that are used to gather data, there are also a variety of qualitative methods. This section focuses on several of these: ethnographic studies, phenomenological studies, case studies, focus groups, and intense interviews. First, in ethnographic studies, researchers examine cultural groups in their natural setting. Examples of cultural groups can include students in a dormitory, women in a crisis center, or people from a village in Asia. This type of study can provide rich, detailed information about the individuals in various groups, since it involves first-hand observation.
Second, in phenomenological studies, a small group of people is studied intensively over a long period to understand the life experience of the individuals being studied. Phenomenological studies can involve direct or indirect observation. Additionally, depending on the study, the individuals being observed may or may not know the purpose of the study or what exactly is being observed. Sometimes the researcher relies on building a trusting relationship with the subjects so the subjects act as naturally as possible even though they are being observed. As a result of this closeness, the researcher can often tell when a person is modifying his or her behavior. However, it is not always possible to establish this kind of relationship. As a result, some researchers conceal the purpose of their studies from those being observed to avoid the modifying of behavior by the subject. This process of behavior modification by the respondent is called reactivity and can greatly affect the results of a study.
Third, in a case study, a case is studied by a researcher, and detailed information about the entity or phenomenon is recorded. Sometimes information that is found in a case study can lend itself to the content analytical techniques discussed in the previous quantitative research section. Other times, newspapers, books, interviews, or other sources may be used. In content analysis, researchers are looking for specific words, phrases, or general ideas that are relevant to their study. The researchers will then count the instances of these items to learn more about a particular subject. For instance, some political scientists are interested in learning about gender bias in the media. By examining how often a female versus a male candidate is mentioned in an article or the type of coverage the candidate receives, these scholars are able to draw conclusions about gender bias in the media.
Finally, there are two other ways to collect and analyze qualitative data that are of relevance in this section—focus groups and intense interviewing. Focus groups are in-depth studies composed of small groups of people who have guided discussions. For instance, a focus group may be shown a political advertisement that a political campaign hopes to air on television. After watching the advertisement, members of the group are asked questions, and a discussion is prompted in which they can discuss their feelings about the ad, such as what they liked and did not like, as well as whether they were swayed by the ad and found it to be credible. These responses allow the advertisement’s producers to make changes that make the ad more effective.
Intense interviews are similar to survey questionnaires in that the interviewer generally has some thoughts in mind about what the respondent will be asked. However, although survey questions are planned out in their entirety in advance, this is often not the case in intense interviews where the interviewee has the ability to ask follow-up questions or a variety of other questions related to an answer provided by the respondent. Additionally, whereas survey questionnaire responses tend to be closed-ended (a particular response can be chosen from those available), intense interview responses are typically open-ended (no response categories) and can be very detailed. Thus, researchers have more flexibility when conducting an intense interview than they would if they were administering a questionnaire; however, their results are typically not quantifiable.
Limitations of Qualitative Methods
Just as quantitative methods have their detractors, so too do qualitative methods. Some of the biggest criticisms of qualitative methods are outlined in this section. First, some critics argue that qualitative methods focus too much on particular individuals, sometimes at the expense of seeing the bigger picture, and they fail to make their results generalizable to a larger population. Second, critics note that the quality of the results and analysis that are produced are highly dependent on the skill of the researcher. It is necessary for the researcher to have remained unbiased and provide a clear assessment of the subjects under study, or the results are essentially meaningless. Third, it is very time-consuming to conduct qualitative research studies. The amount of time spent conducting interviews and making observations is just the beginning. After these take place, the researchers still have to figure out a way to analyze the vast amounts of information that they have collected to produce results.
As can be seen from the information provided throughout this article, there has been a raging decades-long debate as to whether qualitative or quantitative research is better. Many scholars focus on qualitative versus quantitative techniques, automatically framing these methods and approaches in opposition to each other. Although it may appear that qualitative and quantitative data exist in opposition to each other, this is not necessarily the case. As King, Keohane, and Verba (1994) argue, “The two traditions appear quite different; indeed they sometimes seem to be at war. Our view is that these differences are mainly ones of style and specific technique. The same underlying logic provides the framework for each research approach” (p. 3). As a result, research does not typically fit into one particular category or another.
Additionally, King et al. (1994) note that we live in a world that changes rapidly, and to fully understand the changes that occur around us, it is necessary to be able to take into account information that can be quantified, as well as information that cannot. Furthermore, since social science requires comparison, it is important to examine both quantitative differences (such as which phenomena are more or less alike in degree) and qualitative differences (such as which phenomena are more or less alike in kind).
In recent years, scholars have been focusing a lot more on triangulation. Triangulation is essentially the idea that more than one research technique can be used to examine a research question to further verify the findings. Triangulation can help improve confidence about the results produced from a study. Quantitative and qualitative research can frequently be integrated, creating mixed-methods research that can depict a clearer picture of a social science phenomenon than one single method on its own.
Another way that quantitative and qualitative methods can exist together is by coding qualitative data into quantitative data. Just about any type of qualitative data can be assigned meaningful numerical values that can be manipulated to help condense the information and gain a different and more generalizable understanding of the data. One frequently used example is open-ended questions. Although more detailed insight is gained from an open-ended question than a categorical question, open-ended questions can typically be broken down into simple numerical categories allowing for a quantitative analysis of the data.
The Research Network on Gender Politics and the State (RNGS) serves as another good example. The researchers in RNGS had been conducting a crossnational, longitudinal, qualitative research project that explored changes in public policy processes dating back to the 1960s. Starting in 2000, however, the researchers began to code their vast qualitative data into a large quantitative data file. By using quantitative coding, additional useful information may be garnered, and a new form of data analysis is possible. As can be seen here, sometimes the line between quantitative and qualitative analysis may not be so clear after all.
On the other hand, quantitative data is inherently based on qualitative judgment because it is impossible to interpret numbers without understanding the assumptions underlying the numbers. When a person provides a numerical response to a survey question, for instance, many assumptions and judgments are present. For instance, if a person, when asked, “How satisfied are you with your life?” responds, “Very satisfied” (denoted by a value of 1), a variety of other questions could be asked. What does satisfaction mean to this respondent? Was he or she thinking only of the economic climate? Job? Family? Relationships? How does he or she define satisfaction, and how does this differ from how the next person defines satisfaction? Did the respondent even pay attention to or think about the question, or was he or she just offering quick responses? When and in what context was this question presented? The list goes on. As can be seen from this brief example, what appeared to be a simple numerical piece of information actually involved numerous judgments about the meaning of each response.
Quantitative and qualitative analysis are two general approaches to the analysis of data. Both seek to explain trends but have different means of doing this. Additionally, quantitative and qualitative research methods are each based on a basic set of assumptions. Both forms of research carefully follow each step in the research process, from formulating a research question to reporting the results of the data analysis. However, the order and ways in which this process is completed differ between quantitative and qualitative methods because of the different goals that researchers using these methods have for their studies. Essentially, though, at some level, quantitative and qualitative data are inseparable and do not exist in complete opposition to each other. Thus, it is almost self-defeating to claim that one method is better than the other. There are times when one is more appropriate to use in a given situation than another, but often, they can both be used together, whether at the same time or in different stages. As research progresses through the 21st century, it is highly probable that more scholars will use mixed-methods approaches.
- Achen, C. H. (1982). Interpreting and using regression. Beverly Hills, CA: Sage.
- Adcock, R., & Collier, D. (2001). Measurement validity: A shared standard for qualitative and quantitative research. American Political Science Review, 95, 529-546.
- Agresti, A., & Finlay, B. (1997). Statistical methods for the social sciences (3rd ed.). Englewood Cliffs, NJ: Prentice Hall.
- Berg, B. L. (2003). Qualitative research methods for the social sciences (5th ed.). Boston: Allyn & Bacon.
- Box Steffensmeier, J.M., Brady, H. E., & Collier, D. (Eds.). (2008). The Oxford handbook of political methodology. New York: Oxford University Press.
- Brady, H., & Collier, D. (Eds.). (2004). Rethinking social inquiry: Diverse tools, shared standards. Lanham, MD: Rowman & Littlefield.
- Creswell, J. W. (2009). Research design: Qualitative and quantitative approaches (3rd ed.). Thousand Oaks, CA: Sage.
- Denzin, N. K., & Lincoln, Y. (2005). The SAGE handbook of qualitative research. Thousand Oaks, CA: Sage.
- Fink, A. (2005). How to conduct surveys: A step by step guide (3rd ed.). Thousand Oaks, CA: Sage.
- Frankfort Nachmias, C., & Nachmias, D. (1999). Research methods in the social sciences (6th ed.). New York: St. Martin’s.
- Frey, J. H., & Oishi, S. M. (2004). How to conduct interviews by telephone and in person. Thousand Oaks, CA: Sage.
- Gerring, J. (2001). Social science methodology: A critical frame work. Cambridge, UK: Cambridge University Press.
- Goodin, R. E., & Klingemann, H. D. (Eds.). (1996). A new hand book of political science. New York: Oxford University Press.
- Greene, J. C., Caracelli, V. J., & Graham, W. F. (1989). Towards a conceptual framework for mixed method evaluation designs. Education Evaluation and Policy Analysis, 11, 255-274.
- Gujarati, D. N. (2002). Basic econometrics (4th ed.). New York: McGraw Hill.
- Hewson, C., Yule, P., Laurent, D., & Vogel, C. (2002). Internet research methods: A practical guide for the social and behavioural sciences. Thousand Oaks, CA: Sage.
- Hoover, K., & Donovan, T. (2004). The elements of social scientific thinking (8th ed.). Belmont, CA: Wadsworth.
- Isaak, A. C. (1975). Scope and methods of political science: An introduction to the methodology of political inquiry. Homewood, IL: Dorsey Press.
- Johnson, J. B., Reynolds, H. T., & Mycoff, J. D. (2007). Political science research methods (6th ed.). Washington, DC: CQ Press.
- Kinder, D. R., & Palfrey, T. R. (1993). Experimental foundations of political science. Ann Arbor: University of Michigan Press.
- King, G., Keohane, R. O., & Verba, S. (1994). Designing social inquiry: Scientific inference in qualitative research. Princeton, NJ: Princeton University Press.
- Leege, D. C., & Francis, W. L. (1974). Political research: Design, measurement, and analysis. New York: Basic Books.
- Levy, P. S., & Lemeshow, S. (2003). Sampling of populations: Methods and applications (3rd ed.). New York: Wiley.
- Lohr, S. L. (1998). Sampling: Design and analysis. Pacific Grove, CA: Duxbury Press.
- Mack, N., Woodsong, C., MacQueen, K. M., Guest, G., & Namey, E. (2005). Qualitative research methods: A data collector’s field guide. Research Triangle Park, NC: Family Health International.
- Manheim, J. B., Rich, R. C., Willnat, L., & Brians, C. L. (2007). Empirical political analysis: Quantitative and qualitative research methods (7th ed.). Essex, UK: Longman.
- Marshall, C., & Rossman, G. B. (1999). Designing qualitative research (3rd ed.). Thousand Oaks, CA: Sage.
- McNabb, D. E. (2009). Research methods for political science: Quantitative and qualitative methods (2nd ed.). Armonk, NY: M. E. Sharpe.
- Miles, M. B., & Huberman, M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage.
- Miller, D. C., & Salkind, N. J. (2002). Handbook of research design and social measurement (6th ed.). Thousand Oaks, CA: Sage.
- Ott, L. R., & Longnecker, M. T. (2001). An introduction to statistical methods and data analysis. Pacific Grove, CA: Brooks/Cole.
- Pollock, P., III. (2008). The essentials of political analysis. Washington, DC: CQ Press.
- Salmon, W. (1998). Causality and explanation. New York: Oxford University Press.
- Shiveley, W. P. (1990). The craft of political research. Englewood Cliffs, NJ: Prentice Hall.
- Silverman, D. (2001). Interpreting qualitative data: Methods for analyzing talk, text, and interaction (2nd ed.). Thousand Oaks, CA: Sage.
- Singleton, R. A., Jr., & Straits, B. C. (2004). Approaches to social research. New York: Oxford University Press.
- Weisberg, H. F., Krosnick, J. A., & Bowen, B. D. (1996). An introduction to survey research, polling, and data analysis (3rd ed.). Thousand Oaks, CA: Sage.