You’ve probably seen countless interviewers, microphone in hand, clamoring to ask their questions at the crime scene, after the big game, or on the red carpet. Sociologists also use interviews—face-to-face, information-seeking conversations—to gather qualitative data directly from research subjects, or respondents. When sociologists conduct interviews, they try to do so systematically and with a more scientific approach than that used for the kind of interviews you might typically see on TV or read in the news. Sometimes, interviews are the only method used in a research project, but sociologists may also combine interviews with other methods, such as participant observation or analysis of existing sources. Closely related to interviews are surveys, which we will consider in the next section. Both methods are concerned with asking people questions, usually very specific groups of people as well as particular kinds of questions. Interviews, however, are always conducted by the researcher, whereas surveys may be taken independently by the respondent.
When using interviews to collect data about a particular question or project, sociologists must first identify a target population, or group that is the focus of their study. If it is a large group, for instance, all parents with children under eighteen years of age, it might be impossible to study each and every one of them. Researchers, then, must select a sample, or a smaller group that is representative of the larger group. The sample will be used to make generalizations that can apply to the larger target population. The number of possible respondents in a sample depends on the type of study, the nature of the questions, and the amount of time and staff available. In most research studies, interviews can be administered to only a limited number of people, so the scope of such projects is usually smaller than for other methods, such as surveys. While most interviews are conducted one-on-one, some researchers will organize a focus group, in which a number of participants (perhaps five to ten) will be interviewed at the same time, also allowing for group members to interact with one another. This may be one means of increasing the sample size of a study. Researchers must get informed consent from those who will be participating in the study; in other words, respondents must know what they are getting into and explicitly agree to participate. This is particularly important because most interviews are audio or video recorded.
More information
A Black teenager sits on her bed wearing shoes. She is texting on her phone, but her eyes are looking at her mother, who sits on the edge of the bed with her legs crossed, smiling at her daughter.
Mothering While Black Dawn Marie Dow’s study of Black middle- and upper-middle-class moms found that race had to be part of their everyday calculus, even as they navigated a world of class-based advantages.
Sociologist Dawn Marie Dow used interviews for her study of how middle-class African American moms balance the demands of child-rearing with their experiences of race and gender. Her interest in this balancing act came from her own experiences as a child and, later, as a parent. In her 2019 book, Mothering While Black: Boundaries and Burdens of Middle-Class Parenthood, Dow looks at how middle- and upper-middle-class Black mothers navigate the advantages of class at the same time as they must navigate the disadvantages of racism.
Dow interviewed sixty moms, and many of them spoke about how they grappled with the realization that motherhood isn’t the same for Black moms as it is for white moms. For one thing, they often found themselves in all- or mostly white settings. One respondent described how she proactively coordinated with other African American moms so that their kids could enroll in the same ballet class and not be “the only brown girl” (Dow 2019, p. 32). And while the moms Dow interviewed acknowledged that their kids led pretty ordinary middle-class lives—heavily scheduled and highly sheltered—they also expressed concern that the shelter of class did not extend as far for Black children, especially Black boys. As mothers, they were responsible for both protecting their children and preparing them for a world in which racism put them at risk, despite their class-based advantages. Connecting with other African American mothers helped create a space where this balancing act was appreciated and supported by others who understood.
Arlie Hochschild used interviews to conduct her landmark study of parents in two-career families, The Second Shift (Hochschild and Machung 1989). In this book, Hochschild looks at how couples handle the pressures of working at a job and then coming home to what she calls “the second shift”—doing housework and taking care of children. Hochschild, who was herself in a two-career family, wanted to find out how couples were dealing with changing family roles in light of the fact that more women had entered the workforce. Were women able to juggle all their responsibilities, and to what extent were men helping their wives in running the household? Hochschild and her assistants interviewed fifty couples in two-career marriages and forty-five other people who were also a part of the respondents’ domestic arrangements, such as babysitters, day-care providers, and teachers. From this sample of households that Hochschild studied, we can now extrapolate to a much larger population; her findings should also be applicable to similar couples elsewhere.
When conducting an interview, how do you know what to ask? Composing good questions is one of the most difficult parts of interviewing. Most interviewers use many different questions, covering a range of issues related to the project. The best interview questions are typically open-ended, allowing respondents to talk about their own experiences in their own words. Well-known interview researcher Howard S. Becker (1998) often said that the best interview questions are simple, open-ended prompts such as “And then what happened?” and “So, how did that make you feel?”
Researchers must be careful to avoid biased or leading questions, those that predispose a respondent to answer in a certain way. Overly complex questions are a problem, as are double-barreled questions, those that involve too many issues at one time. It is also important to be aware of any ambiguous or inflammatory language that might confuse or spark an emotional reaction on the part of the respondent. Asking a single parent how difficult their life is will elicit data about the difficulties, but not about the joys, of parenthood. More neutral language, such as “Tell me about the pluses and minuses of single parenthood,” is preferable. In some studies, researchers will solicit the entire life history of a respondent, a chronological account of the story of their life from childhood to the present or of some portion of it.
Once the interviews have been conducted, they are usually transcribed so that researchers can analyze them in textual form, a process called “coding” the data. Coding involves sorting through the material looking for repeating categories and patterns among the answers. Some researchers use computer applications designed to help code such data; others do it “by hand,” writing codes down on index cards, for example, or cutting quotes out of printed data and making piles of excerpts for various code categories. For her analysis, Hochschild coded the types of household chores done by men and women and quantified the amount of time spent daily and weekly on those chores. She then categorized couples as “traditional,” “transitional,” or “egalitarian,” depending on how they divided their household labor.
Advantages and Disadvantages
ADVANTAGES
Interviews allow respondents to speak in their own words; they can reveal their own thoughts, feelings, and beliefs, internal states that would not necessarily be accessible by any other means. In many other instances, it is the researcher who tells the story. A book like The Second Shift, which features direct quotations from interview transcripts, provides the reader with an authentic and intimate portrait of the lives of married couples. Hochschild was able to get at the different subjective experiences of the women and men in her study and to see how each of them perceived the reality of their situation.
Interviews may help the researcher dispel certain preconceptions and discover issues that might have otherwise been overlooked. For example, before Hochschild began her project, many other studies had already been conducted on families with two working parents, but few seemed to examine in depth the real-life dilemma of the two-career family that Hochschild herself was experiencing.
DISADVANTAGES
Although most interview respondents are forthcoming and truthful, not all are. A few may be selective about what they say in order to present themselves in a favorable light. Sometimes they are difficult to talk to, and at other times they may try too hard to be helpful. An adept interviewer will be able to encourage meaningful responses, but it can sometimes be helpful to triangulate data gathered using interviews with data gathered using other methods. Hochschild, for example, also observed a few of the families she interviewed. She saw that what these couples said in interviews was sometimes different from how they acted at home—in particular, that couples’ actual divisions of household labor were less egalitarian than they thought.
Another problem is generalizability: whether the conclusions of interview research can be applied to larger groups. Because face-to-face interviewing is time-consuming, interviews are rarely used with large numbers of people. Can findings from a small sample be generalized to a larger population? In regard to Hochschild’s research, can we say that interviews with fifty couples, although carefully selected by the researcher, give a true picture of the lives of all two-career families? Hochschild was able to answer this question by comparing selected information about her couples with data from a huge national survey.
IN THE FUTURE
Emerging Methods in Sociology
Other methods are being developed and used by sociologists. These emerging methods may not yet rival the major methods of social research in popularity, but each has its own novel merits.
Action Research: Combining Research and Social Change
A growing trend in social science methodology, action research combines social science research with community problem solving and social change, in a way that calls into question some of sociology’s closely held beliefs about ethics, bias, and the role of the researcher. Action researchers see their research skills as problem-solving tools, and they view those whom others might call “research subjects” as active, collaborative, equal participants in the project. In other words, action researchers do research with people, not on people, and see their work as part of a “scholarship of engagement” (Rajaram 2007, p. 139), rather than one of erudite distance. While action research is not exactly new, it has been gaining popularity recently, across the social sciences as well as in practice-oriented disciplines such as nursing, public health, education, and urban planning.
An award-winning example of action research is the work of Chicago’s Community Organizing and Family Issues group (COFI). Its project “Why Isn’t Johnny in Preschool?” sought to answer this question, particularly among families in low-income, racially diverse neighborhoods, where kids are less likely to be enrolled in early childhood education programs. COFI sent community members, trained in interview methods, out into their neighborhoods to talk with more than 5,000 other parents about the barriers to preschool enrollment. These findings were then used to design outreach campaigns that promoted the importance of preschool attendance and provided information packets that helped families find solutions to some of the problems identified in the research. Preschool attendance increased in the targeted neighborhoods as a result (COFI 2009).
Digital Ethnography
Digital ethnography, or virtual ethnography, involves the use of participant observation methods to study online communities. We interact with others and engage in group activities online more and more, so it makes sense that sociologists would begin working to understand online social worlds using the discipline’s distinctive methods. While it’s true that our online communications are not exactly the same as our communications in real life (IRL), it seems appropriate to use adapted versions of real-life ethnography to study online social worlds and social interactions, rather than invent new methods with which to study internet life.
Online ethnographers study (mostly) written communication as it occurs in online locations—gaming environments, social media sites, and other web-based locations. They collect observational data as they participate in these online settings, and they code and analyze that data interpretively, just as traditional ethnographers do. In doing so, digital ethnographers can draw conclusions about the culture of online life and can use those findings to expand our knowledge of human interaction in all sorts of settings, from dealing with infertility (Lee 2017) to prescription drug use (Del Fresno and Peláez 2014) and beyond. For example, a team of ethnographers studied online celebrations conducted during the pandemic (Bascuñan-Wiley, DeSoucey, and Fine 2022). They explored the experiences of groups of friends and family members who had to dramatically change their traditional gatherings (for birthdays, holidays, and other special meals), moving them online during periods of isolation in 2020 and 2021. The meanings of shared meals changed when participants could neither cook nor eat together in physical copresence: If Grandma always bakes the cake but baby can only see it on Zoom, who blows out the candles? The researchers found that folks created new ways of “being together” even when they were prevented from being in the same room and that foodways—our beliefs and behaviors surrounding food—are flexible. Digital ethnography can open fascinating windows into online social worlds.
Mapping Methods
Another emerging method in sociology is GIS, or geographic information systems, a type of computer software that attaches social science data (like population demographics) to geographic locations. This allows sociologists not only to know that 13 percent of a city’s population lives in poverty, but also to know exactly where they live in the city. Maps of instances of illness, for example, can be overlaid on maps of pollutant exposure to examine the correlation between the two. The ability to include spatial and geographic data in sociological analysis makes GIS especially useful for studying diseases, matching populations with services, and answering questions about migration (Sianko and Small 2017). In fact, GIS was used extensively to better understand the Covid-19 pandemic (Franch-Pardo et al. 2020): For example, one study used data from Facebook to show that areas with a lot of social ties to early hotspots had higher numbers of Covid cases (Kuchler et al. 2021).
In another real-world example, sociologists Deirdre Oakley and Keri Burchfield (2009) looked at what happened to residents of Chicago’s notorious public housing projects when the buildings were demolished and the residents relocated in the early 2000s. These relocations were ostensibly meant to improve the situations of the former residents—to move them out of neighborhoods of concentrated poverty, crime, racial segregation, and general disadvantage. Residents were given vouchers to assist with their moves. Oakley and Burchfield used GIS to map the residents’ original locations with where they ended up after relocation, mapping each neighborhood’s characteristics, such as poverty and crime rates. What they found was that while the residents may have moved to different neighborhoods, those locations weren’t necessarily “better.” In fact, Oakley and Burchfield write, “The prospects of escaping high-poverty neighborhoods through relocation are very slim” (2009, p. 606), and it is through the use of GIS that their research shows this unfortunate outcome.
