The term methodology may be defined in at least three ways: (1) a body of rules and postulates that are employed by researchers in a discipline of study; (2) a particular procedure or set of procedures; and (3) the analysis of the principles of procedures of inquiry that are followed by researchers in a discipline of study. This entry will first discuss each of these definitions. It will then cover the debate among philosophers of science about general methodological assumptions. Finally, the entry will review some of the issues pertaining to the quantitative versus qualitative debate about methods.
A BODY OF RULES AND POSTULATES
Methodology refers to the behavior of scientists and scholars when examining phenomena relevant to their specific disciplines. The American Heritage Dictionary offers the following formal definition for methodology: "the theoretical analysis of the methods appropriate to a field of study or to the body of methods and principles particular to a branch of knowledge" (Pickett 2000, p. 2074). Methodology in the social sciences is usually characterized by the following: (1) it defines the information to be analyzed; (2) it provides the conceptual tools and procedures necessary to perform an analysis; and (3) it sets forth the limits of the analysis. Methodology necessarily encompasses the three facets of exploration, description, and explanation (Babbie 2001, p. 91).
When scholars undertake research projects, they usually follow a well-defined procedure known as the research process. The research process begins when a researcher determines his or her research topic and formulates the research question or questions. For example, the research question of a social scientist could be: Are minority children in the United States more profoundly effected by poverty than their white counterparts? Once the research question has been determined, it is necessary to construct a study design. This is where the researcher decides the type of research to be undertaken. In the social sciences, there are two broad types of research: quantitative and qualitative. The former relies on numerical and statistical techniques and data garnered through the analysis of large groups of subjects. The latter often involves in-depth interviews that are designed to probe and produce extensive information about small numbers of subjects. Inductive and deductive reasoning also come into play when the researcher decides to follow a predetermined framework for the duration of the project (deduction), or instead formulates the research question and allows the remainder of the research to unfold as it may (induction). Once the researcher has decided which method to use, the next step is to collect the data. Finally, the data are analyzed, interpreted, and put into a format accessible to others.
Many of the social science subfields (e.g., economics, psychology, sociology, etc.) have developed specific models for the collection and organization of knowledge. Economics and psychology were the first to develop mathematical models of inquiry. Many subfields have borrowed heavily from one another. Beginning around 1970, a general system of models for all the social sciences was developed following guidelines proposed by the Social Science Research Council (Hekman 1980). The social sciences have also borrowed heavily from statistics, as is evident in the quantification of information obtained in the research process.
A PARTICULAR PROCEDURE OR SET OF PROCEDURES
The social sciences use an assortment of research methods. These include, but are not limited to, experiments, surveys, field research, content analysis, analysis of existing data, comparative research, and evaluative research (Mouton and Marais 1988). Each of these types of analysis needs to be systematized.
The first and most important step in any methodology is the formulation of the research question. It is with this step that the researcher determines the direction and approach of analysis. This step involves exploration, is crucial to understanding the topic, gives an idea of the feasibility of the research, and identifies the methods to be used (Babbie 2001, p. 92).
Once the researcher determines what the research will entail, it is then necessary to address the study design. During this step, the unit of analysis is determined. The unit of analysis is often an individual or group of individuals that is sampled from the larger population to which the researcher wishes to generalize his or her findings. For example, if one’s research were to focus on the above question concerning whether American minority children are more profoundly effected by poverty than white children, the appropriate unit of analysis would be the child, and a sample of children would be drawn from the larger population of minority and white children. In this example, the use of quantitative methods would be most appropriate and would enable the researcher to develop generalizations about the general population.
However, even if the researcher chooses to use inductive methods, it would still be necessary to address many of the above issues. The study design portion of the process would involve determining sample size, who or what should be studied, and how the study would be conducted. Researchers must create a framework for their study for a number of reasons, including the need to coordinate the activities of more than one researcher and to obtain funding, among other things.
The next step in the research process is the collection of the data. This is accomplished according to the framework that has already been prescribed. In a quantitative study, researchers will most likely use a survey or some other type of questionnaire. In qualitative research, a list of questions would probably be employed, along with less-structured interviews. An important function of scientific inquiry is description. Qualitative studies in particular enable the researcher to describe situations and events in detail.
After the research data have been collected and organized, it is necessary to undertake the analysis. This may include statistical analyses of data gathered via quantitative methods, or more straightforward descriptive analyses of data obtained via qualitative methods. The data are then interpreted and summarized so the results of the research will be more accessible and available to others. Explanation is the natural by-product of research, and researchers hope that their projects provide information that answers the original research question.
ANALYSIS OF PRINCIPLES
All methodologies include a system of analysis that is used as a backdrop for organizing, collecting, and interpreting data. Data are usually systematized through either inductive and deductive reasoning. In most research projects, the system of analysis is determined in the first step of the research process because the researcher must at that point choose how he or she intends to collect data. Deductive reasoning begins with the idea that the researcher has a predetermined framework and uses it as a model to guide the research (Mouton and Marais 1988; Babbie 2001). Inductive reasoning, on the other hand, allows the researcher to begin a project with a general question but without a clear outline or framework.
METHODOLOGICAL ISSUES IN THE PHILOSOPHY OF SCIENCE
One of the major issues when conducting social science research is validity. Some argue that the social sciences do not deal in empirical fact and as such are not as valid as the so-called hard sciences. It is believed that researchers are human and can therefore never be fully objective. Max Weber (1864-1920) argued for value-free sociology and urged researchers to contribute information free from subjective opinions (Mouton and Marais 1988; Weber 1962). This is a central debate in social science research. Many researchers strive to separate research from value judgments, and the idea behind quantification within the social sciences is a nod toward value-free judgments. In fact, the idea of creating a methodology with clear procedures and principles is the embodiment of the necessity to make the social sciences more objective in the eyes of the public.
Approaches to methodology in the social sciences generally fall into three categories: positivist, interpretive, and critical social science. William Neumann writes that the positivist approach is the most widespread and is based on the methods of the natural sciences. In the social sciences, this approach was first used by Auguste Comte (1798-1857) and was later expanded by Emile Durkheim (1858-1917) (Neumann 2003; Smith 1983). The posi-tivist approach is most strongly linked to the quantitative realm of social science research and strives for objectivity. It is argued that only when the social sciences follow the models prescribed by the natural sciences can the findings be valid and reliable. According to Neumann, positivism as it relates to social science can be defined as "an organized method for combining deductive logic with precise empirical observations of individual behavior in order to discover and confirm a set of probabilistic causal laws that can be used to predict general patterns of human activity" (2003, p. 71).
Interpretive social science originated with Max Weber and is focused on discovering the meaning behind social action. There are many facets of interpretivism, including hermeneutics, constructionism, ethnomethodology, and qualitative sociology. This school of thought argues that the social sciences cannot be analyzed using the methods of the natural sciences because they are inadequate for studying the meaning behind human behavior (Lee 1991). As stated previously, this approach focuses on meaning. As such, qualitative work is often considered an interpretive method. Neumann defines the interpretive approach as "the systematic analysis of socially meaningful action through the direct detailed observation of people in natural settings in order to arrive at understandings and interpretations of how people create and maintain their social worlds" (2003, p. 77).
Critical social science is associated with Karl Marx (1818-1883) and Sigmund Freud (1856-1939). Practitioners of this approach criticize the positivist approach for its inability to focus on meaning in the human context, and the interpretive approach for "being too subjective and relativist"(Neumann 2003, p. 81). Neumann defines this approach as "a critical process of inquiry that goes beyond surface illusions to uncover the real structures in the material world in order to help people change conditions and build a better world for themselves" (2003, p. 81).
QUANTITATIVE VERSUS QUALITATIVE ANALYSIS
Quantitative researchers are mostly concerned with measurement and sampling and often use deductive reasoning. In addition, it is important to identify facts and laws that can be used as predictive tools. In contrast, qualitative researchers tend to be more interested in content and use induction with more frequency (Neumann 2003). In this case, the meanings associated with human behavior are taken into consideration. Though quantitative and qualitative research differ in many ways, both types of research make important contributions that benefit the other. Quantitative research provides data in numerical form and allows for the manipulation of the data using statistical procedures. That information can then be combined with the descriptive data provided through qualitative research to provide more meaningful results.
Many social science researchers use both methods to provide fuller and more complete explanations. Indeed, since the 1980s, there has been a convergence of the two approaches, and many analyses use both methods (for examples in demography, see Massey 1987, 1990; and Knodel et al. 1987). The argument here is that qualitative research enriches purely quantitative research by filling in gaps created by the sole use of straightforward statistical methods. On the other hand, quantitative research can enhance qualitative work by providing validity in the form of numbers. And qualitative work can enhance quantitative work by providing detail in the explanation of certain trends that is not possible through a strict analysis of the numerical data.
Qualitative methods examine social data without quantifying the data. Qualitative researchers often examine the links between theory and analysis and seek to discover general patterns among and between their variables. Some of the methods involved in qualitative research include grounded theory, semiotics, and conversation analysis (Mouton and Marais 1988). Qualitative researchers more frequently use an interpretive or critical approach and are interested in allowing the meaning of their work to develop as they conduct more research. They are further interested in using their research to explain and predict. Their data usually take the form of words, rather than numbers, and such researchers most often reason via induction (Neumann 2003).
In contrast, researchers using quantitative methods seek to transform the collected data numerically so the data can then be analyzed using statistical methods. Quantitative researchers most often use a positivist approach because accuracy is an important requirement. The data are first described using measures of central tendency, such as the mean, median, and mode. In most cases, the quantitative researcher next moves to multivari-ate analyses that examine several variables simultaneously. Quantitative research begins with the formulation of a research question and then a hypothesis or hypotheses. The researcher then identifies and operationalizes the desired variables, creates a standardized data set, defines procedures with which to analyze the data, and finally undertakes analysis using the statistical methods described above.