INTRODUCTION

In contrast to most survey research and all experimental research, in this chapter we concentrate on techniques that are almost always inductive, where theory is primarily generated from data. Rather than using data to confirm or support existing theory, researchers use observational field research, or participant observation, to discover theory. They try to gain an understanding of human action and social process by entering, as far as possible, the worlds of those whose behaviors they are trying to understand. Participant observation lets us see the world as others see it – “in their own terms.” When we do that we learn how they achieve a coherent, ordered existence. Participant observation is, therefore, based on the presumption that by studying people in the natural settings where they live and work, social scientists will maximize the ability to grasp the motives, values, beliefs, concerns, troubles, and interests that underlie human behavior.

The Chicago School and Beyond

The famous researcher Robert Park is quoted as having issued the following methodological directive to his students:

INTRODUCTION

In Chapter 1, the scientific method was described as a process that contains theory and the testing of hypotheses or hunches suggested by the theory. Indeed, in all of the various social science disciplines, and no matter what method of data collection is employed, theory is used as a guide in the collection and analysis of data. Moreover, in many studies where the researcher does not begin with a thoroughly defined theory, we can create theory by searching through the data for recurring patterns of behavior.

The application of social theory to research is one of the prime areas for the use of the research imagination. This chapter explains the relationship between theory and methods in detail. It describes the various forms that theory may take. It compares and contrasts research that is designed to verify existing theory with studies that are designed to create new theory. Then, the source of research ideas or how researchers decide what to study is discussed. Finally, the idea of research as a “craft,” or set of related skills that can be practiced and learned, is introduced.

THEORY AND SOCIAL RESEARCH

There are numerous options open to the investigator in choosing a theoretical perspective. Moreover, the many different kinds of social theory vary in terms of the specific concepts they contain, how complete or well articulated they are and in the way they are expressed on paper, with words or diagrams.

INTRODUCTION

The first four chapters of this book set the stage for learning about research by introducing you to a vocabulary that specifies the various parts of the process and presenting the basics of research design and measurement. These chapters are a foundation for investigating the social world. However, before the various methods are presented in detail, this chapter summarizes the ethical and political context for research.

Growing up, many of us have been taught to abide by the Golden Rule in church, school, and community. Social researchers, however virtuous they may be in a personal sense, are concerned about ethical behavior in a more specific context. Because of the public nature of science and the increasing visibility and accountability of researchers, they must pay attention to ethical principles or suffer the criticism of those who read their work and who provide them with data. As we will see in this chapter, in research, doing the “right thing” is not always easy or obvious!

Suppose that a team of politically moderate social scientists has received a research grant from the federal government to investigate political extremism on both the left and the right in America. The researchers will interview members of the Communist Party and the American Nazi Party to find out what their attitudes are on a variety of political and social issues.

INTRODUCTION

If we become ill, or our parents are going through a divorce, or someone in our family loses a job, we have troubles at a personal level. However, over time, as sociologist C. Wright Mills (1960/2000) pointed out, history and biography intersect. That is, the private troubles that we experience at a microlevel are connected to public issues. In the United States today, there is a growing crisis in health care, both in terms of affordability of and having access to health insurance. The divorce rate has grown so high that in many places half of all marriages do not survive. In addition, as the economy shifted from mostly industrial to mostly postindustrial, pockets of permanent, structural unemployment dot the national map. These are examples of public issues that require historical analysis if we are truly to understand their origins and scope, as well as future trends. In fact, it is no exaggeration to say that without skillful historical analysis we would be deprived of a powerful tool to understand the context that helps us to cope successfully with our private troubles.

Historical analysis is a process calling for multiple methods, both quantitative and qualitative, microlevel as well as macrolevel, and the attention of all the social sciences. In this sense, it cuts against the grain of academic specialization that characterized the twentieth century.

INTRODUCTION

In the last chapter on survey research, the focus was on asking the right questions and getting them put down on the printed page (or the Web-based medium) in a clear and concise way. Much of the research imagination employed in survey research occurs before data are actually collected and is apparent in the overall design of a survey, the operationalization of variables, and the wording of the specific questions to be asked. In this chapter, the focus is on intensive interviewing, a technique in which, by contrast, appropriate questions are often determined as data are being collected. This means that creativity and spontaneous decision making are integrated at every step of the research process. In this up-close, conversational technique, clarity of communication is a mutual effort between the researcher and those being studied.

Interviewing is increasingly well established as a methodological alternative that can help us explore what to ask and how to ask it. Sometimes such information is used only to formulate structured questionnaires for surveys, but interviews are also immensely valuable in their own right. There are many research challenges for which intensive interviewing is the ideal investigative technique. Let us look at one example. Beyond a few memoirs (Delong and Petrini, 2001; A. C. Gray, 2003), not much is known sociologically about female crime-fighters – private investigators, police detectives, and FBI agents. We do know that relatively few women engage in these traditionally male-dominated occupations. How could we study them?

By now you have spent many hours reading about social investigation. You have been introduced to a new vocabulary of key terms and have become familiar with the search for “objectivity,” reliability, and validity. You have examined some of the ethical dilemmas of research. It is now time to take stock of what you have learned.

A CRITICAL PERSPECTIVE

Although this textbook may have increased your awareness of research methodology, you may not be pursuing a career in the social sciences. Many of you will never carry out your own research outside the context of a college classroom. Does this mean that the knowledge of methods is of no use to you? To the contrary, the same criteria used to evaluate social science research equip you to consider critically the wide range of assertions about social life that all of us read and hear daily. Knowledge of the obstacles to a sure understanding of society should enable you to develop a healthy skepticism toward the generalizations offered in the mass media, in literature, and in informal conversation.

Of course, we all make sense of our individual and collective experiences by making generalizations about the world. But it is important to realize that individuals are, in some measure, limited by the boundaries of their own experience. Our knowledge of social life emerges from the distinctive social positions we occupy, and our “realities” are constrained by our own life situations.

These ideas have inspired us to write this book about research methods. After decades of experience in planning and carrying out social research projects of all kinds, we are convinced that research is indeed a craft requiring judgment and creativity, not simply learning the rules of science and applying them. Whether one is doing the most intimate one-on-one interviewing or large-scale examinations of how entire societies make public policy, human imagination and scientific principles of inquiry go hand in hand. To that end, this book emphasizes scientific method but also acknowledges its critics. It covers a wide variety of data collection techniques but presents them as reinforcing, rather than competing with, one another.

A Balance between Qualitative and Quantitative Methods

This is a book for students and instructors who want a comprehensive treatment of a variety of research techniques but with special attention to qualitative approaches. We are committed to a balanced approach that gives a variety of qualitative methods full exposure alongside more mainstream quantitative strategies. Joe R. Feagin, a past president of the American Sociological Association, has commented on the almost exclusively quantitative emphasis of articles accepted for publication in leading sociology journals. He has advocated more realism in recognizing the methodological diversity within the discipline. Feagin (1999) also noted that many sociologists who study, for example, race, ethnicity, gender, class, and sexuality oppose a heavy emphasis on quantitative social research.

INTRODUCTION

The first two chapters have explained the elements of the scientific method and examined the connection between theory and the research process. After reading these chapters, you are now ready to look at how research is actually designed, step-by-step. research design is the overall process of using your imagination as well as the strategy and tactics of science to guide the collection and analysis of data. Have you have ever wondered: “What is a proper problem to write about in my term paper?” This issue, selecting a topic for study, is the first and most crucial one to be addressed in social research. The dilemmas of research design are sometimes difficult to resolve. Professional social scientists with years of experience sometimes wonder if investigating a problem that happens to interest them personally will produce a worthy addition to knowledge in their field of inquiry.

A related and equally thorny issue to be explored in this chapter is whether it is legitimate, or desirable, to be wholly descriptive in one's work, rather than explanatory. In other words, should all research make some theoretical contribution? In addition, we will examine the question of how much information is needed to substantiate an argument. The old axiom that “the more we know, the more we realize our own ignorance” certainly applies in social research. Are there guidelines to help us determine when we should stop collecting data and begin to analyze it?

INTRODUCTION

Having learned about sampling in the preceding chapter, we are now ready to examine social surveys, a quantitative technique. Surveys are the most common form of research in the social sciences. We are all regularly exposed to surveys in one form or another. We may be stopped on the street or in a shopping mall by interviewers who ask our opinions on current issues or our preferences for consumer items. We are telephoned by surveyors who ask how we intend to vote in an upcoming election or why we chose one airline over another. We receive a variety of questionnaires in the mail – from government agencies, business corporations, and community organizations. Newspapers report the latest public opinion polls; the major television networks employ their own pollsters. All these are forms of survey research: systematic attempts to collect information, mostly from individuals, to describe and explain the beliefs, attitudes, values, and behavior of selected groups of people. Nearly every topic of interest in the social sciences has been studied through surveys, and the broad adaptability of this research method is one of its greatest strengths.

The characteristic of surveys that best illustrates their value and explains their extensive use is their ability to produce a representative distribution, or cross section, of the “target” population, whether that population is the entire American voting public or the people who stayed at a certain hotel during a given month. So, surveys are most often coupled with probability sampling techniques.

INTRODUCTION

The last two chapters of this book cover the field of statistics, from a basic introduction through multivariate analysis. Social scientists regularly encounter quantitative data that have been summarized and presented in statistical form. They find statistics in almost everything they read, from articles in popular newspapers and magazines to scholarly journals. Published reports of the results of quantitative research usually include charts, diagrams, graphs, and tables. To evaluate such work, it is essential to develop a thorough understanding of statistical procedures, including the assumptions and limitations underlying each technique. The purpose of this chapter is to outline the basic procedures, ideas, and issues associated with the analysis of quantitative data.

UNIVARIATE ANALYSIS

A distinction is often made between descriptive research, which highlights the outstanding characteristics of a sample, or of the population from which the sample was drawn, and explanatoryresearch, which concentrates on cause–effect connections among those characteristics. In practice, most quantitative studies involve elements of both description and explanation, beginning with a descriptive statistical summary of the data and progressing toward testing hypotheses and causal relationships. Descriptive analysis typically involves consideration of one variable at a time rather than the relationship between two or more variables. For this reason, it is often referred to as univariate analysis.

INTRODUCTION

In this chapter, the detailed examination of the various methods tools and approaches in research that will be the focus for the remainder of this book begins, starting with sampling, the selection of a relatively small group of individuals from whom we obtain data in order to be able to generalize about a larger group. Sampling demonstrates the idea that science is a blueprint for research because there are indeed some rules for proper sampling that are essential. However, as we will see, in many cases, research imagination is also an important ingredient in successful sampling.

Let us take a look at public opinion polls. Nowadays, the results of polls are readily available in newspapers, in magazines, and on television. Strategies have been developed to ensure that these polls can be completed affordably and yield accurate results. Most public opinion polls in America survey only 1,000 to 1,500 respondents. This is a surprisingly low number considering the population of the United States is more than 300 million. How can such a small number accurately reflect the opinions of these millions of people? It is a matter of good sampling.

Consider Table 6.1, which reports findings from a major ongoing poll of public satisfaction. Over a three-year period, the public appeared rather divided about “the way things were going,” although there was a slight up tick in dissatisfaction. Only 1,003 adults were contacted; yet, they are taken to represent the opinions of the nation as a whole.

INTRODUCTION

This chapter is about the measurement of complex behaviors and attitudes. We know from earlier discussions that many of the variables that researchers manipulate are complex; for example, a number of separate mental and physical operations are required to produce the attitude of prejudice as well as discriminatory behavior, or the attitude of religiosity as well as pious behavior. Indexing and scaling are techniques for measuring these and other complex phenomena in social science.

Before some concrete examples are presented and the logic of index and scale construction is explained, it will be helpful to recall the basic principles of measurement outlined in Chapter 4 and the illustrations of questionnaire items in Chapter 7. An important way to increase both the reliability and the validity of abstract constructs – such as “happiness,” “alienation,” “tolerance,” and “anxiety” – is to operationalize them by using multiple indicators of the same phenomenon. A number of survey questions may be combined to assess the strength of a particular variable, the degree to which it is present, or its intensity. This idea will be explored in more detail.

Indexes and scales are devices for creating a single composite measure of behavior and attitudes out of a number of related indicators. Thus, they are particularly useful for summarizing complicated activities and orientations such as human mental capacity and people's perceptions, interests, and intentions.

INTRODUCTION

Social scientists want to understand the behavior of individuals and how this behavior is affected by membership in social groups. Many of the methodologies we have discussed in earlier chapters (for instance, survey research, experimentation, participant observation) are used to examine peoples' attitudes, beliefs, and values. If we want to discover why people vote as they do, exhibit prejudice, or engage in criminal behavior, we often proceed by interviewing or observing an appropriate sample of individuals. We try to show the distinctive characteristics of people who engage in these behaviors by taking individuals as the units of analysis, that is, the source of the data from which we are able to make generalizations.

However, individuals are not always the focus of social research. We may want to understand the nature, character, and dynamics of social structures, as well. We frequently wish to compare institutions according to some attribute. Social scientists sometimes take as their unit of analysis such organizations as universities, business corporations, prisons, or hospitals. As we will see, much investigation is also concerned with geographical, or areal groupings, of people. We might be interested in comparing rates of suicide in various countries. Although it is true that individuals are responsible for taking their own lives, the focus of our research need not be on the particular or separate motives, beliefs, or personal life conditions of these individuals.

INTRODUCTION

In this chapter, we consider one of the most deductive of all research techniques, the experiment. As a data collection technique, experimentation seems ultrascripted and not at all spontaneous. However, the elaborate laboratory environment that is prepared in advance in many experiments, and the rigorously practiced research roles assigned to those conducting the experiment, actually require a great deal of research imagination. In this case, the creativity is required at an early stage in research design, long before the research subjects appear in the laboratory to begin the experiment itself.

The idea of an experiment should not be foreign to most of us. We all use loose or incomplete forms of experimentation in our daily lives. Husbands and wives experiment with different ways of preparing food, noting their spouse's reaction to each modification of the prepared dish. Salespersons intent on finding the most persuasive pitch to use in selling a product may systematically test out a number of approaches until they find the one that works best. Teachers experiment with different formats for presenting course material, selecting one that allows students to learn the most. Vacationers in Las Vegas try out many systems at roulette or blackjack, hoping to find the one that will make them winners.

INTRODUCTION

Content analysis is a systematic attempt to examine some form of verbal or image communication such as newspapers, diaries, letters, speeches, movies, or television shows. Usually this communication already exists, and the researcher wants to discover its implications for the study of human behavior. Content analysis can be either inductive or deductive theoretically, but it is primarily quantitative because the examination of communication usually occurs through counting its content. Like other quantitative techniques, the most creative part of the method is concentrated toward the early stages of research design, when the categories to be used for counting have to be defined. In developing their ingenious data collection tools, practitioners of content analysis demonstrate their research imagination.

The analysis of communication content has a long tradition in the social sciences. It has revealed significant information about the values of both communicators and their audiences. This chapter summarizes the major elements of this methodology, offers some examples of its application, and discusses some of the reliability and validity issues that arise for the investigators who use it.

Consider the following example from the work of Simon Davis (1990), who analyzed personal advertisements to see what motivates mate selection. Do men and women differ in the factors seen as important in choosing a partner? Davis pursued this question by looking at the “personals” section in a major daily newspaper.

INTRODUCTION

In contrast to the chapters on survey research, experimentation, or content analysis that described a distinct set of skills, in this chapter, a variety of comparative research techniques are discussed. What makes a study comparative is not the particular techniques employed but the theoretical orientation and the sources of data. All the tools of the social scientist, including historical analysis, fieldwork, surveys, and aggregate data analysis, can be used to achieve the goals of comparative research. So, there is plenty of room for the research imagination in the choice of data collection strategies. There is a wide divide between quantitative and qualitative approaches in comparative work. Most studies are either exclusively qualitative (e.g., individual case studies of a small number of countries) or exclusively quantitative, most often using many cases and a cross-national focus (Ragin, 1991:7). Ideally, increasing numbers of studies in the future will use both traditions, as the skills, tools, and quality of data in comparative research continue to improve.

In almost all social research, we look at how social processes vary and are experienced in different settings to develop our knowledge of the causes and effects of human behavior. This holds true if we are trying to explain the behavior of nations or individuals. So, it may then seem redundant to include a chapter in this book specifically dedicated to comparative research methods when all the other methods discussed are ultimately comparative.