Chapter 3: Methods of Research

Section 2: Qualitative Methods

Introduction

In the last chapter we learned about the importance of data and theory in conducting sociological research and we were introduced to two sociological traditions in research, the Positivist Tradition, and the Interpretivist Tradition. 

Lecture 10 focused on the Positivist Tradition which emphasizes using deductive scientific methods for understanding social life. To do so, sociologists use what are called quantitative methods of research, or methods of research using numerical data and measurable variables. We then learned about three different methods of research, questionnaires, experiments, and content analysis, and reviewed their strengths and weaknesses. 

Lecture 10 was pretty full.

So quantitative data has many many uses, even when analyzing something relatively subjective like, say, racism. Can a quantitative instrument be devised to research something as subjective as racism? 

Yes. Racism isn’t easy to measure. It’s not like you can walk up to someone and ask them if they are racist. They’ll likely say no. However, there are ways to ask them questions in a survey that can measure their level of racial resentment.¹ Using these tools, you can get a profile of racism in a particular area. Then you can reproduce the research and compare different regions or cohorts. If you get consistent results over time, it’s likely you have a useful measure. It won’t be complete. There will be limitations, but it can be useful when trying to address race issues. 

That being said, given that you can reduce a complex concept like racism down to a number value for analysis, does that number help the researcher understand racism as a lived experience? If you determine that twenty percent of the people in a region demonstrate a high level of racial resentment, you must ask “why.” What is it about that twenty percent, or about the other eighty percent? What is it like to live in such a community if you are black, or if you are white? How do people respond to racism? How do they develop racist ideas? How do the victims of racism cope with living in a racist society? 

Some time ago Vox.com posted a video by Christian Picciolini, a former member of the Neo-Nazis, very racist. He has since left the movement and has become an equal rights advocate. What happened? He started out just a young man like any other. He became a racist as a teen but then changed his mind. So, this brings up questions about free will, or as sociologists call it, agency. Might understanding this gentleman’s life story and how he exercised his own agency help us find ways of convincing others to give up on racism and discrimination? 

Interpretivism and Qualitative Research

The Positivist tradition, using quantitative research is not adequate for answering these questions. Instead, Interpretivist sociologists believe that the best way to understand lived experience is by directly studying how people interact and analyzing how people assign meaning to social situations. Whereas the positivist tradition gives the impression that human beings are simply acted on by external social structures, the Interpretivist tradition understands that human beings can and do assert significant agency and interact with social structures rather than just respond to them. 

Interpretivists tend to use Qualitative Methods of Research. That is, Interpretivists are interested in getting the stories or narratives of individuals’ personal lives and how they navigate through and make sense of their social worlds. 

Qualitative researchers usually make use of an Inductive Method of analysis. In other words, the qualitative researcher is going to collect his data first before making any assumptions. The researcher will then look for patterns in the data. The researcher may apply a theory to understand the patterns that she sees, or may even develop a new theory that can lead to new directions in research. Qualitative research can also help evaluate the value of theory in understanding real-life situations, filling in some of the holes that all theories have, and elaborating new directions for research.

Methods of Qualitative Research: Interviews

Qualitative research uses two major methods, Interviews and Observations.

Unlike the structured interviews done by quantitative researchers, qualitative researchers use semi-structured or unstructured interviews. With this method, the interviewer asks open-ended questions or asks her subject to respond to a given topic. This may be done in one-on-one sessions or in groups. There are few if any specific questions. Instead, the interviewer follows the subject’s responses and allows the narrative to develop in its own way. This allows the researcher to ask contextual questions and prompt elaboration.

Whereas with structured interviews, the researcher must have a clear idea of what he or she wants to study. She must make certain assumptions and predictions of outcomes going into the interview in order to shape the questions. With unstructured interviews, the researcher minimizes the assumptions taken. This allows the researcher to be open to and responsive of new directions as the data develops. Respondents may go in directions that the researcher never anticipated. This can be much more revealing than structured interviews, increasing their validity. 

Such interviews, however, require a very skilled interviewer, one who can respond quickly to changing currents in the conversation. Issues with reliability occur as well. Respondents may lie or misrepresent themselves or simply have difficulties with recall. Memories are selective and often subject to interpretation based on how we feel at any given time. Also, the less structured an interview, the more difficult it is to reproduce the results. And, as with structured interviews, the Researcher Effect or the Hawthorne Effect is especially at play with unstructured interviews. In a group setting, group dynamics also impact the stories that are told.

Methods of Qualitative Research: Observation 

The next major method of qualitative research is observation, in which the researcher goes into a research setting and observes the interactions. Again, the researcher is looking for patterns from which he can understand the lived experience and how people attach meanings to their lives. This is a great way to get valid data, as the researcher is directly witnessing and recording what’s going on. Reliability becomes more of an issue, however, because it’s hard to reproduce the research in a different setting, and all observations, even by a trained researcher, are subject to interpretation. That’s why observation techniques require a highly trained observer who can ground his observations in sound social theory. It also helps if he can support his observations with more empirical quantitative data. 

There are two ways to conduct observations. Participant Observation is the most common, but researchers may also use non-participant observation. Participant Observation is where the researcher takes part in the behavior and interactions that he’s studying. This is consistent with Max Weber’s concept of Verstehen, or an empathic understanding of human behavior. In this case, the researcher has a deep understanding of the interactions he’s studying because he experiences those interactions, including the psychological and emotional ramifications. 

Sociologist Sudhir Vankatesh gained unprecedented access to a local gang. His observations were published in the 2009 book Gang Leader for a Day. Vankatesh was able to make use of a sponsor, a middle-ranking member of the gang who knew Vankatesh, to gain access to and establish trust with the group. 

Sometimes a researcher will spend significant time in the field, often many years, and make very detailed observations over a broad spectrum of the lived experience during that time. This kind of intensive participant observation is called Ethnography.  William Foote White set the standard for urban ethnography in his study, Street Corner Society. Whyte researched the social organization of what many considered a disorderly and disorganized Italian slum in the 1940’s identified as “cornerville”, really North End Boston. Another famous sociological ethnography was conducted by Robert and Helen Lynd in their study of Midwestern norms and values in Middletown, really Muncie, Indiana.

These two studies are examples of Overt Observation, or participant observation in which the subjects know that the participant is conducting research on them. This kind of observation is valuable because it allows the researcher to conduct follow-up research on the interactions observed at the moment. This is important because it allows the researcher to ask probing questions and get elaboration from regular participants in the field. Being asked to elaborate on one’s thinking with regard to a normal interaction would seem strange if those in the field didn’t know they were being studied. Normal interactants don’t usually ask probing, contextual questions. Imagine what it would be like if someone you were interacting with asked you to explain the details of your actions. That would be weird.  But if everyone knows that you are a researcher, they may be more accepting of your inquisitiveness.

Being known as an observer may also help you gain access to all levels of a research setting whereas regular participants only have access to one. For instance, conducting research in a large organization would be constrained if you were not known as a researcher. Most people only interact with a limited network in the day-to-day interactions on their company jobs. The people in the clerical pool will typically only associate with other clerics. They don’t often get to interact with folks at the top or at the very bottom of the ladder. An overt observer, however, can gain such access. 

On the other hand, if people are aware that they are being observed, they may alter their behaviors. This is predicted by the Hawthorne Effect discussed in the last lecture. In this case, is the researcher getting valid data on the lived experience? Maybe not. 

Another option for participant-observers is covert observation or observation in which the subjects do not know they are being observed. This controls for the possibility of a kind of Hawthorne Effect. My own research in the wilderness camp was covert observation. I had a role at the camp from which I kept notes and observations in the normal routines of my job, but the young people and most of the staff I was working with, other than my direct supervisors, did not know that I was conducting research. This helped with the validity of my research because I didn’t have to worry about my subjects changing their behaviors. But covert observation is limiting. If Sudhir Vankatesh had been conducting covert observation, he may have been expected to participate in the gang’s illegal and immoral activities, putting him in an ethical bind…even more so than he was already.  There are also ethical concerns with regard to privacy that I will discuss in the next lecture.  

Participant observation is great for developing that sense of Verstehen, but may actually be too good. When participating with other people it’s not uncommon to adopt the norms and values of those you are studying. You develop relationships with your subjects, even friendships and affinities. These emotional ties may bias your research. 

One way of controlling for this is through non-participant observation. In this case, the researcher gains access to a research setting, but only to observe. 

Observation studies are often the main focus of case studies. A case study is an in-depth investigation into a particular event, person or group. The case study will involve qualitative methods, but may also pull in quantitative studies that help the researcher get a more broad picture of what is going on in the sample. 

Case studies using multiple kinds of research is an example of Methodological Pluralism. Methodological Pluralism combines research methodologies like positivism and interpretivism in complementary ways to improve validity and reliability. If we look at Quantitative Methods as having high reliability but relatively low validity, and qualitative methods as being valid, but of relatively low reliability, then how do you, as a researcher, maximize the validity and reliability of your research? Well, you could combine quantitative and qualitative methods. 

When you use two or more research methods to understand a social phenomenon, you are using a technique called triangulation. This allows the researcher to compensate for the weaknesses of on study process by utilizing complementary strengths of another. Ideally, if multiple research methods reach similar conclusions, then those conclusions for which there is overlap should be considered valid and reliable. 

There are three different kinds of triangulation: Researcher Triangulation happens when multiple researchers using the same research methods in the same settings. Ideally, the researchers should represent different ethnic, age, gender and class groups to control for subjective influences or biases. Data Triangulation uses multiple sampling strategies to collect data. Again, those areas of agreement between multiple samples are more likely to be reliable and valid. Finally, there’s Theoretical Triangulation in which the same research setting and focus is looked at using different theoretical perspectives. After all, if methods are subject to bias, so are theories. 

Conclusion

So there it is. In the last two lectures, we’ve covered methods of research and their associated issues. Research is central to sociology as a science. Otherwise, all you have is editorializing, and that’s not the point. As with all sciences, the goal of sociology is to elaborate often hidden truths that can be applied to policy or to improve the real lives of real people. 

Social reality, however, is a complicated thing. It would be nice if there were just one tried and true way to analyze society. But there isn’t. On the one hand, having all of these different methods of conducting research is a great strength of sociology as a discipline. It’s a great way to focus the lens of your sociological imagination. Quantitative research is a really powerful tool for getting that big picture of society, like the wide-angle lens used to photograph landscapes. But this kind of analysis leaves a lot to be desired when it comes to understanding the lived experience of individuals. For this, you need a lens with a small frame, but great detail, like those used to take portraits or flowers. The bottom line is, if you only focus on one lens, you end up with an incomplete understanding of the social world. 

However, sociological research faces a serious complication. You see, as sociologists, the subjects of our research are human beings, often real, living human beings. This poses a problem because of the very nature of sociological research, prying into social interactions, and revealing the hidden constructs are also things that can cause harm to a research subject. Knowing this, it is extremely important for the sociologist to approach her research with a strict set of ethical principles by which the subjects as well as the reputation of the profession as a science are protected. Next lecture we will look into some ethical principles of sociological research. 

---

Notes

1. That’s not to say that there isn’t significant debate between social scientists about the validity of such measures. Social scientists are involved in a never-ending process of developing better tools for measuring subjective experience. ↩︎