144 QUALITATIVE INTERVIEWING When circumstances preclude building such relationships, you may have to modify your expectations and interviewing patterns. In focus groups and other multiple-person interviews, you do less questioning and pay more attention to how the interviewees interact; in telephone interviews, you try to get more information across in your introductory letter and preliminary phone calls. In the next several chapters, we discuss how to prepare main questions, probes, and follow-ups, and how to use these three types of questions to structure an interview. This structure helps make the pattern of questioning intelligible to the interviewee and helps you get the depth, detail, and accuracy that qualitative interviews aim for. But it is the stages of the interview described in this chapter that encourage open and thoughtful answers. If you match your questions to the stages of the relationship, your interviewees should enjoy the interview more, feel more satisfaction in what they have told you, and be more willing to talk to you again. ŠMäSEKí §Sf ff Assembling the Parts Structuring a Qualitative Interview n qualitative interviewing, you change the questions you ask depending on what you learned or failed to learn. Overall through the interview, though, you have to maintain balance between separate lines of inquiry and ensure that there is time to go into depth on each major subject. To allow flexibility to change questions while maintaining an overall structure, researchers pattern interviews around three types of questions—main questions, probes, arid follow-up questions. Further, by preparing conversational guides— written outlines, protocols, or checklists—researchers set up an overall framework for the interview to keep the interview on course yet allow sufficient flexibility for exploring uncharted paths. THREE TYPES OF QUALITATIVE QUESTIONS An interview is built up from three kinds of questions. Prior to talking with the interviewee, the researcher prepares a handful of main questions with which to begin and guide the conversation. Main questions change during the course of the research, as the researcher learns what to ask and to whom to ask it. 146 QUALITATIVE INTERVIEWING When responses lack sufficient detail, depth, or clarity, the interviewer asks a. probe to complete or clarify the answer or to request further examples and evidence. Skillfully done, probes also communicate that you are paying attention to what the conversational partners are saying without uttering "I hear you" a hundred times. Follow-up questions pursue the implications of answers to the main questions. Follow-ups examine central themes or events, or ask for elaboration about core ideas and concepts. Main Questions Before each interview, the researcher prepares several main questions to direct the discussion. Breaking the overall topic into several related questions is done in a way that provides unity to the interview. The wording of a main question should be open enough to encourage interviewees to express their own opinions and experiences, but narrow enough to keep interviewees from wandering too far from the subject at hand. In preparing main questions for topical interviews, the researcher usually works out a series of queries that together cover specific events or stages of a process. With cultural interviews, main questions often are little more than conversational devices to encourage interviewees to begin describing what is important in their cultural arenas. You prepare main questions in advance, customizing them to what you think the interviewee might know. But you may not get a chance to ask all of the preplanned main questions. For instance, our colleague Jim prepared 15 main questions for his interviews on civil rights suits with federal judges. In the actual interviews, he rarely reached the fourth question, and sometimes only asked the first one. Not getting through a list of prepared questions is usually not a problem. The interviewee may answer your questions before you ask them or take off in a direction you did not anticipate but find important. But if you are tracing a series of events or finding out what parts of a program work, you have to ask questions you have prepared on each of these topics or the work is incomplete. To learn how cities handled fiscal stress, Irene had to ask interviewees about a series of key events in the history of the city's finances. She gently moved the interviews Structuring a Qualitative Interview 14? along to cover the main questions she had prepared. On those occasions Säir when she still could not get through the questions, she arranged a second Jľfi£::ť interview to finish them or set up. an interview with someone else who " W"': could answer the remaining questions. Siří In the next two chapters, we discuss in detail how preparing main Sír. questions differs in cultural and topical interviews, but three concerns '*?;'''■ should be kept in mind, no matter what the purpose of the interview. :;£'■:. First, do the main questions covéŕ'the overall subject? When the ^*.: researcher wants to explore a social or political process, are there main V' questions on each of the major events and stages? Does the wording /V encourage a discussion of the separate components of a culture without i prejudging what is important and what is not? Second, do the main questions flow from one to the next? Are the - (■■ transitions between the main questions smooth? Will they make sense to the interviewees? For instance, in tracing out histories of events, the main questions might follow a chronology. What happened first? What ■'_.-'■ happened next? This is a common way that people share narratives and so should seem logical to the interviewees. The connection between the '■-■- questions is obvious. More generally, the separate main questions should cover an overall subject in ways that suggest an underlying focus. Suppose you were interviewing about life in a retirement home and prepared three main questions to cover different parts of life in the home. First you could ask, "How good is the care from the staff?" The second question might be, "What is the social life like at Golden Acres?" A third main question could be, "Do you keep up contact with family and old friends?" Each main question speaks to a major part of life in a retirement home. First you asked about basic care, next about a sense of community, and your last question explored how people handled separation from their past. The interviewee should be able to see the pattern of the questions. Finally, in choosing main questions, check to make sure the questions match the research design. Do the main questions cover the subjects that emerged from the iterative design? Are the main questions changing appropriately as the research progresses and you learn more? Have the main questions been selected to fit what separate interviewees should know? Have they been worded to accommodate the experiences and perspectives of individual conversational partners? 148 QUALITATIVE INTERVIEWING Probes Probes perform three main functions in an interview. First, they help specify the level of depth the interviewer wants. Probes signal the interviewees that you want longer and more detailed answers, specific examples, or evidence. Probes encourage the speaker to keep elaborating. Second, probes ask the interviewee to finish up the particular answer currently being given. The interviewer may ask the interviewee to clarify an ambiguity or fill in missing information necessary to understand the answer. The third function of probes is to indicate that the interviewer is paying attention. You can show interest through attention probes. One form of attention probe is to ask, "Can I quote you on that?" Whether or not you need permission, by asking for it, you are telling the interviewee, "I am listening intently, you phrased that point extremely well, I would like to get it down and use it just the way you said it." The way you take written notes, especially if you are also tape-recording, can also be an attention probe. Each time you mark something down, you communicate, "I heard something that is just too important to forget, that I might want to come back to." Such attention probes teach the interviewee what type of material you find especially informative. During one interview, Herb asked the interviewee if he could quote him right after the person described a core part of his philosophy of work. A while later, the interviewee offered, "Here is a big quote for you. 'The skillful plagiarism beats inept originality every time.'" The interviewee then explained what the quote meant to him. This conversational partner had figured out from Herb's interest probe which issues Herb wanted to learn about and picked up the format of a summary quote to highlight his points so Herb would not miss them. You can use continuation probes to signal the interviewee that it is okay to keep elaborating, that what he or she is saying is right on target and you want to listen to all the details. One form of continuation probe is to silently lean forward in a posture of respectful listening. The interviewee reads the body cues and continues without your having to say anything. Brief continuation probes such as, "Go on," or "what happened then?" also indicate that you want to be told more on this subject. If the interviewee stops in the middle of a narrative, you can repeat part of the last sentence. For example, in presenting a narrative about a ^Structuring a Qualitative interview 149 berime, the conversational partner might say, "I approached the rich- Iv-looking couple and thought about what might be in the lady's purse." ř The conversational partner then hesitates. You wait for a second, and if Sfethe interviewee doesn't continue, you repeat the subject and verb, "You ií§v approached ..." and then you pause. You hope the interviewee will ŠÍP'.- resume the narrative. If you repeat other parts of the sentence, such as SSfSp-.' "the rich-looking couple" or "in her purse," your probe applies to these ""^ .phrases and suggests you want to know more about the rich-looking couple or the purse. If the interviewee interrupts him- or herself, resulting in an incomplete thought, you can ask for a completion of that line of thought. "Before we talked about the strike, you were saying that the union was all absorbing ..." invites either an explanation of "all absorbing" or a completion of the original thought. Sometimes what the interviewee is saying is simply not clear. It may be garbled grammatically, overburdened with pronouns, ambiguous, or lacking in sufficient detail to give a clear picture. Then you use a clarification probe, a form of conversational repair. "You said she did not want to go to a nursing home. Was that the social worker who said that or your mother?" Sometimes it is not just grammar that is unclear, but technical vocabulary. Or the steps in a process may not be clearly outlined. When our colleague Jim asked a computer hacker how he got into a university computer, he got the following answer: They don't know their own system's security procedures, and we just got the book. We read it, and I ran it, and we got it. He saw us, I guess. We were using an ID of some guy whose account was canceled, and they asked who we were, but we were lucky and social-engineered it and he told me to get out, so I did. So we didn't do anything other than get in, ya know? But, that's what all we wanted. Jim then used a clarification probe. "Could you run that one by me again? I am afraid I still don't understand how you did that." Herb says in similar situations, "That's kind of technical, but I think I need to understand. Could you explain it again?" If you are unsure of what someone said—perhaps the person gave you a summary and maybe some key points were left out, or the material was garbled grammatically—you may be able to ask, "What were her 150 QUALITATIVE INTERVIEWING exact words, do you remember?" or "What did she actually say to you?"—"I don't remember exactly, but it was something like, 'I would rather be dead.'" In this probe you are asking for evidence (the exact words), but at the same time you get clarification of the meaning. Elaboration, continuation, clarification, attention, and completion probes are housekeeping probes. They ensure that you are getting a reasonably accurate and understandable answer while encouraging the interviewee to keep talking. They help teach the special rules of conversation that apply to interviews. But probing does more than keep the conversation going, it helps get the depth and dependability you need in qualitative interviews and the freshness of firsthand descriptions by encouraging narratives and stories, and requesting examples and evidence. If someone gives you a general answer such as, "There is a lot of conflict around here," you don't have enough yet to follow up on, so you probe to encourage the interviewee to continue. You can ask, "What do you mean by 'conflict'?" or "Could you give me an example of a conflict?" Sometimes you want to know how heavily to rely on a particular answer. You use an evidence probe to find out how the person knows what they are telling you. Evidence probes need to be phrased tactfully and can only be used a limited number of times before you sound like the grand inquisitor. One way is to phrase the evidence probe so it sounds more like a search for detail rather than checking up on how the person knows. For instance, in an interview about a hiring conflict, you can ask whether the interviewee was on the recruitment committee. How you probe helps define the interpersonal relationship between interviewer and conversational partner. Insufficient probing indicates boredom or inattention; too much probing and the researcher turns into an inquisitor. Fortunately, as the relationship develops, you need to probe less often, because interviewees learn the level of detail and evidence that you want and begin to frame their answers to match. Follow-Up Questions Main questions create a scaffolding for the interview, keep the questioning on the topic, and link what is asked in individual interviews to ".■.Structuring a Qualitative Interview I51 'the overall design. Probes clarify and complete the answers, making them intelligible, and signal the interviewees about the expected level .0f depth. They also show the interviewee that the interviewer is interested in the answers. The purpose of follow-up questions is to get the depth that is a hallmark of qualitative interviewing by pursuing themes that are discovered, elaborating the context of answers, and exploring the implications of what has been said. Some follow-ups are worked out in the periods between two interviews with the same conversational partner; others are thought of during the interview itself. In either case, follow-ups cannot be prepared prior to the initial interview, because they will be based on the interviewee's responses to the main questions. After each interview, and after each cluster of interviews, look over your transcripts to figure out what you should follow up on. Look for themes, ideas, concepts, and events and prepare additional questions on those that address your research concerns. Preparing follow-ups between interviews takes work and attention, but you usually have sufficient time to think about which issues should be explored in depth. Following up during an interview is more difficult. With experience, you learn to pay real attention to what the conversational partner is saying and figure out what seems to be missing or what you need to ask to elicit additional depth. The trick is choosing to follow up on only those matters that provide insight on the core matters of concern. Listen for partial narratives, unexplained lists, and one-sided descriptions of behavior that beg for elaboration. When a conversational partner provides a list of reasons, events, or explanations of the core topic, and then only explores one, you follow up by asking, "Would you talk a bit about some of the other items you mentioned?" You may follow up when a narrative or example has been left incomplete. For instance, if somebody describes a fight between the public works manager and the budget officer about what project to do next and then goes on and talks about the project, the obvious follow-up is to ask about the fight. Such follow-ups are also suggested when interviewees skip topics that were important in discussions with other people. Follow-ups cascade, because the answers to one follow-up can suggest new lines of inquiry that you want to follow up on in turn. Consider 152 QUALITATIVE INTERVIEWING the following example from our colleague Jim's interviews. In response to a main question about prison life, a prisoner explained that boredom was his worst problem. Jim followed up: Jim: What kinds of things do you do in prison that keep you from becoming bored? Prisoner: You can join gangs, play chess with your cellie, go to the [exercise] yard, drink, join the JayCees, get into the education program, or just lay around your cell watching tv, if you don't have a [job] assignment. Startled, Jim probed for clarification. Jim: Uh, "drink?" Did you say "Drink?" Prisoner: Yeah [laughs]. Drink. You know, drink. Jim: As in "drink, drink?" Like . . . Prisoner: As in drink, booze, hooch, firewater, joy-juice, alcohol, ya know? Jim: [laughs]. Prisoner: Got news for ya. Anything you can get on the outside, we can get in here. Jim then followed up again: Jim: Amazing. If I were a prisoner, how would I go about getting something to "drink?" Contradictions or puzzles, statements that sound guarded, incomplete answers, and the use of new words or unfamiliar terms should trigger follow-up questions. In the case of the prisoner and the drink, Jim followed up on the new information because it sounded puzzling. How could someone in prison get liquor, and how could they afford to buy it, assuming that it was available? A similar invitation to follow up occurs when interviewees explain how people should behave. The puzzle is, do they actually behave that way, and what happens when they don't? Structuring a Qualitative Interview-. 153 Asking good follow-up questions is a matter of trained curiosity, recognizing and pursuing puzzles while exploring emerging themes. Sometimes a theme virtually leaps out at you, as when Jim stumbled across the prison economy. But more often, an unanticipated theme is more subtle and you have to be exposed to it several times before you hear it and it dawns on you what people are saying. Follow-ups to these more subtle themes rarely occur in one interview, but depend upon your taking the time between interviews to study what your interviewees have told you. Although new words or concepts invite probes or follow-ups, you can't ask about every word you don't understand. If you ask about too many words in a row, the interview ceases to sound or feel like a conversation and you lose focus and depth. You might also appear so ignorant that interviewees feel you are not worth the effort to teach. You sometimes have to request follow-ups without seeming to ask about every word. For example, in one of Jim's interviews with a young computer hacker, Jim was overwhelmed with new vocabulary: Jim: You said earlier that you were "hacking" last night. What did you do? Young Hacker: I was running numbers, you know, wardialing 288s, and got a couple of hits. So I cracked into a local dialup, found a PBX and hacked into one of 'em and tried one of 'em. I got in, and ran one and hit. A Unix system. After a few tries, I got root, set myself up, logged out, then hacked out a coupla more. . . . The interviewee responded with what was to him everyday vocabulary that needed no translation, Jim wanted to know what the interviewee meant but was wary of appearing totally ignorant. So, Jim phrased his follow-up in the following way: Jim: Now, if you were talking to a reporter, say somebody from the New York Times like John Markoff (co-author of The Hacker Crackdown), how would you explain to him, so he could explain to the public, what "running numbers" means? Jim managed to keep the focus on one core term, expecting that an explanation would include definitions of some of the other puzzling 154 QUALITATIVE INTERVIEWING phrases without his having to ask a dozen boring questions about vocabulary. What terms you follow up on depends on the conceptual focus of the study. Suppose you are interviewing students about life in the dorms and find out that loud music is a particular problem. Interviewees mention particular vocalists and musical groups, such as Black Uhuru and Tracy Chapman. If you never heard of Black Uhuru, you should not stop and ask, lest you get into a discussion of contemporary music and lose the thread of the interview. A more appropriate follow-up would be, "Where do you go to study?" or "Is there a quiet floor?" Or "Are there rules against loud music?" Or you could follow up with a question about other problems besides loud music. Follow-ups are about more than learning the meaning of a core idea or concept, or completing a missing story or narrative. They are the way the researcher explores emerging themes with the conversational partner. When answering such follow-ups, the interviewee becomes a full-fledged research partner. You can ask this conversational partner what he or she thinks of an idea that a different interviewee has suggested. Or you can follow up by asking about themes you think are in the answers this interviewee has already given, but you are not sure about them yet. For example, when Irene was interviewing in Phoenix, the city manager responded to main questions of why the budget process changed by describing the professionalism of the staff and the political involvement of the citizenry. From her prior research, Irene was surprised that he had not discussed the possibility of financial deficits causing changes in the budget process, because other interviewees had done so. When the manager had finished his answer, she followed up by saying, "I noticed that you never mentioned deficits. Is that a factor in why the budget process changes?" The manager indicated that he thought it was a relatively unimportant source of change, and then he and Irene discussed the reasons why. The discussion stimulated by this follow up caused Irene to rethink an underlying theme. When you are following up on themes that emerge from the interviewee you are currently talking to, choose what to follow up on depending on what you have learned from other interviews and how well each follow-up helps build the picture you are forming of the research arena. I BSeF' Structuring a Qualitative Interview 155 > SSE For example, in the main question of a life history interview we 'haf"., asked a senior citizen to talk about her early work experience, especially i.-|v; jjgj contact with unions. Her narrative was rich and detailed. She had •Hf- worked in a clothing factory, because it was the only job that she could '/t*. get, but she longed for an office job. Eventually she was promoted into ■, "*?- the office and loved the work. Then there was a strike. During the strike, '" *i" she crossed the picket line, but her friends on the factory line protected ' '7,-~ her and did not call her a scab, because they said, "She is okay, she '. -- works in the office." Remembering back all those years, she said, "They really respected anyone who worked in the office." Neither she nor her - - friends saw promotion to office work as making her the enemy. But her background and her union connections made her suspect to the bosses. Fortunately for her, because she was such a good worker and they knew ; "" her already, management let her stay, as long as she promised not to pass any information to the union. There are a number of ways to follow up on such a complicated narrative, depending upon the themes being explored. The central tension for the woman was between being supported by and admired by union members and being proud of working in the office and winning the support of the bosses. You might want to examine this tension with follow-ups such as, "Were you ever tempted to pass information to the union?" or "How did you feel when you had to cross the picket line?" Or you could follow up by trying to find out what loyalty to the boss meant among the office staff or what "a good worker" meant. Her answer is so rich with thematic material that it is stunning, and you may not think of all the ways that you want to pursue it at the moment. You might want to come back to some of them later in the interview, such as how she felt about the union after she became an office worker, how she felt about the promotion, whether her pride in doing white-collar work resulted in her looking down on the factory workers, and whether her attitudes were widely shared among the office workers. Any of these would lead to deeper understanding and you choose among them depending upon the overall thrust of your research. When you get a complex answer like this, filled with new information and possible themes, you stop thinking in terms of single questions, but rather plan clusters of linked follow-ups. To help you keep these in mind, jot down the ideas as they come to you, and then gradually ask them, allowing for full discussion of each one, and sometimes following 156 QUALITATIVE INTERVIEWING up on the follow-ups before returning to the original list. Even if you are tape-recording you must jot down these follow-ups and where you are in the process of questioning lest you get lost in the cluster. Both the mental exertion of intently listening to what is being said and the excitement of the answers can make your mind go blank on what you intended to ask next. Sometimes the initial answer you get isn't as rich as the one in this example; in fact, it may be very lean, though you suspect from other interviews that there is hidden depth. In these circumstances, you follow up in stages, asking about the separate parts of what the person said, beginning first with clarifications of terms and then moving on to the deeper issues involved. The follow-ups here are meant to encourage the person to talk in more depth. Sometimes the initial question you ask is too abstract or too stressful, and the interviewee cannot come up with a good answer. For instance, we were interviewing a woman about her relationship with her mother and heard her express a feeling of guilt. To follow up, we asked what she felt guilty about. When the woman did not answer this broader question, we asked more specific questions about things she might have felt guilty about. "Do you feel that you should be spending more time with your mother?" "Does she blame you for her illnesses?" Such follow-ups are meant to encourage the interviewee to start talking. You don't need the answers to these questions specifically, you just made them up to give the interviewee an idea of what an answer might be. She might reject these questions completely, but now tell you what it is she does feel guilty about. In this case, the interviewee responded, "No, it is not time I feel guilty about, it is not having her in my house. And I feel guilty trying to get her to live within her budget, as if I did not want to supplement her income. It makes me feel cheap." Once the interviewee starts to respond in depth, you follow up on the responses, rather than continue to ask the questions you made up that were not based on her answers. For instance, you can continue by asking, "Did she want to live with you?" or "Does she have enough money to live on comfortably?" Follow-ups generally work better when they are grounded in the immediate topic at hand. When our follow-ups focus narrowly on specific events or themes, our interviewees can usually answer the specific question and then build toward their own generalizations based ŕ' Structuring a Qualitative Interview 557 ".'. on particular experiences they have had. They are telling you both what ■ '. they feel and what they have experienced. If you ask for generalizations .. first, without focusing on concrete experiences, you may get abstrac-. tions back that may not be connected to the interviewee's experiences. For example, in some of Herb's interviews with community developers, he followed up on an answer about funders by asking how the interviewees felt funders responded to the community needs. In re-: sponse, the interviewees answered that funders did not realize that : investments in low-income communities would pay off. That was an okay answer, as far as it went, but it wasn't concrete or grounded and provided few details. Herb changed the way he asked the question, posing a specific example instead of asking a general question. He asked, "What difficulties did LISC [a supplier of funds] create in packaging the project?" The interviewees provided grounded, concrete details about LISC and then described examples of difficulties their organizations had with other funders. Figuring out what to follow up on is usually more difficult than figuring out how to word a follow-up question. There is one basic principle of wording follow-ups: Try to make sure that your follow-ups correspond to what the conversational partner has said. The following conversation indicates how easy it is to ask a follow-up that doesn't match what the interviewee is telling you: Interviewer: How did you feel when your father died?. Answer: I was really torn up. Really bad. I couldn't sleep for a few days, missed a few weeks of classes, and I cried constantly. Interviewer: Has anybody else in your family died in the past few years? The follow-up ignores the content of the response and misses the opportunity to explore the consequences and management of grieving. Better follow-ups would stay closer to the material. For example, "What kind of things made you cry?" Or "Did you have anyone to talk to about what you were feeling?" You could even ask about the interviewee's relationship with her dad before he died. Sometimes the follow-up doesn't match the previous answer closely because the interviewer used a follow-up that was planned in advance without thinking about whether it fit the situation at hand. When the 158 QUALITATIVE INTERVIEWING interviewer is continuing a topic with the same conversational partner at a later time, preparing such follow-ups is quite helpful. But be careful not to mechanically use prepared follow-ups in circumstances in which they do not belong. Herb recently made such a mistake because he was too tired to listen carefully. (This is an argument for pacing your interviews to ensure that you get enough rest between them.) In several previous interviews, developers had complained that they hated the aggravation of convincing multiple funders to cooperate on one project. Herb had worked out a standard follow-up asking how developers handled this aggravation. In the next interview, a person described her main skill as getting funding agencies to cooperate. Herb, too eager to use the prepared follow-up, asked how she handled the aggravation. Herb's follow-up, of course, was nonsensical, and his interviewee told him so in no uncertain terms. PUTTING IT ALL TOGETHER Sometimes interviews just happen. You walk into a room, the conversational partner is attuned to your interests, begins talking, and the conversation flows. But you can't count on that occurring; rather you need to plan how to put your questions together to guide the conversational partner through a focused conversation. Metaphorically, putting together the main questions, probes, and follow-ups is a little like playing golf. You have a series of holes, the separate topics of the discussion, each of which you approach with a main question, a big long drive that you have planned out in advance. Depending on where your drive lands, you follow up with other shots, choosing clubs with different shapes that are most likely to get you close to the green from where you are. If you hit the ball into the rough—that is, get an answer that is vague or incomprehensible—you choose a specialized club, the probe, that helps you get past the difficulty. You continue the probing and following up until you get the ball into the hole. You repeat the process again on the next hole with another preplanned main question, follow-ups, and probes as needed. The holes are connected in a logical order and give structure to the whole game. Structuring a Qualitative Interview 159 The interview is like a golf game in that you think about the overall direction yet maintain flexibility to adapt to what you have heard. Just as the choice of club depends on where your ball lands, your choice of follow-ups and probes depends on what you hear in answer to your main questions. The details of sequencing and wording questions differ between cultural and topical interviews, but the need to maintain overall coherence in the interview remains the same. To do this, you generally plan the overall structure of the interview by figuring out roughly the balance you want between main questions and follow-ups. Then you prepare conversational guides for yourself, such as written protocols, outlines, and checklists, that remind you of what you want to ask without forcing you into a preset pattern. Framing the Interview As part of preparing for an interview, the researcher decides how to link the main questions to each other and determines the strategy for following up on what he or she hears. Two patterns of structuring the interview are quite common, although in practice there are a number of variations in between. With the first approach, the tree-and-branch model, the interview is likened to a tree. The trunk is the core topic; the branches, the main questions. You plan the questions to explore each branch with more or less the same degree of depth. With the second model, the interview is more like a major river that merges different currents into a single stream and then breaks into separate channels, possibly combining again later into a single stream. In this second case, the questions explore one current within the main river and follow it no matter where it goes. The researcher might use the tree-and-branch model if she or he knows, perhaps from previous interviews, observation, or background reading, that certain main questions must be asked in order to cover the entire subject. The interviewer follows up on the answers to each main question looking for depth and detail, but in doing so balances depth in one part of the subject with coverage of all the parts. The goal is to learn about the individual branches that frame the entire tree but still obtain depth and detail.