Answer the questions
Gathering Qualitative Data: Interviews & Focus Groups
Ok, so let’s talk about the way we conduct two types of methods for gathering qualitative data: interviews and focus groups. The primary objective of this lesson is to identify the appropriate steps to follow to gather data using these methods.
1
INTERVIEWS
INTERVIEW AS CONVERSATION
Interviews provide rich, descriptive data to understand and decipher meaning.
Interview protocols are more/less structured, but always have a purpose.
Allows for flexibility and discussion/conversation.
Focus on participant’s meaning and experience (i.e., the interview is a meaning-making process)!
Acts like a conversation, but not a conversation
The purpose of an interview is to ask questions and get answers from the participants involved in your study in order to get descriptive knowledge. And there is a great deal of work that goes into conducting interviews. And, its important for you to understand the theory behind the research because this will help you determine what to ask participants in the interview. The whole point of an interview is to get participants to ‘open up’ and share their experiences with the person who is interviewing them. So—you want the participant to see the interview as more of a conversation—it is a process of meaning making for both people—BUT, it is actually NOT a conversation. Remember: you want to understand things from the participants point of view—you want to understand meaning from their lived experience. Interviews are very good for Focus on understanding of meaning based on experiences and language of participants. They’re also very good for validating or exploring information in a mixed method design. And it is appropriate to USE interviews when you want to give VOICE TO THE PARTICIPANTS—when the participant’s perspective is important to the research goal. To help us get at this—there are different types of interviews or different approaches to interviews.
2
INTERVIEWS
| Interview Protocol Type | Description |
| Structured | All questions must be asked to every participant! Think: verbal survey-esque. Prepare all questions on interview guide/protocol ahead of time! Each participant: same questions, same wording, same order! Think: “YOU SHALL NOT DEVIATE!” |
| Semi-structured | |
| Unstructured |
We have structured, semi-structured, and unstructured interviews. Let’s talk about each of these for a moment. In a structured interview, all of the questions on the interview protocol—or this list of questions that the interviewer will ask the participant—all of these questions must be asked to every participant. Structured interviews are almost like verbal surveys. In a structured interview, the interviewer all the questions ahead of time, asks each participant each question on the interview protocol, and there are fewer open-ended questions included on the protocol. Then, of course, the interviewer does not insert his or her personal opinion into the interview, so, there’s no agreeing or disagreeing with the responses.
3
INTERVIEWS
| Interview Protocol Type | Description |
| Structured | All questions must be asked to every participant! Think: verbal survey-esque. Prepare all questions on interview guide/protocol ahead of time! Each participant: same questions, same wording, same order! Think: “YOU SHALL NOT DEVIATE!” |
| Semi-structured | More of a conversation between the interviewer and participant. Interviewer has a clear list of questions, but may skip some questions (or ask follow-up or probing questions) depending on the nature of the interview. Narrative interviews; collecting someone’s life history. All open-ended questions! |
| Unstructured | Consists of a clear list of topics or broad, open-ended questions to discuss in the interview, but no set question list. Interviewer may create questions in the interview. |
The next two types are semi-structured and unstructured interviews. Semi-structured are more flexible than structured—AND unstructured are MUCH more flexible than both structured and semi-structured interviews. In both of these types of interviews the interviewer and participant have a formal interview where they meet with one another and chat about a specific topic or topic(s). For both types the interviewer has developed an understanding of the setting and context to allow the majority of the questions to be more open-ended in nature. So, the interview protocol for both of these interview types is more flexible compared to the interview protocol for a structured interview. The difference between these two is that in an unstructured interview the interview protocol doesn’t consist of pre-determined questions. Instead, the interviewer may go into the interview with simply a broad list of topics and then make up questions depending on how the interview goes and the information that the participant discloses.
4
INTERVIEWS
| Things to consider… |
| What is the research question? Role of the researcher. Access and sampling. Negotiating access to the population. |
Regardless of the type of structure you will use, you always need to follow a few steps for planning interviews. So, here are things that you need to consider--you must ask yourself what are you looking to study? Where does it take place? Who are the folks that are important in what you want to study, why do you think they are significant, and how can you ask these folks about the topic you want to research—in other words, how are you going to go about getting at your topic with the sample population. Often, interviews are used in studying populations that are underrepresented in research or populations that are hard to reach. For instance, I recently published a study that used interviews to examine coping strategies for refugees post resettlement in the United States. The population was very challenging to recruit and I had to be very thoughtful about the way I worded interview questions—especially because the population had experienced a lot of trauma and there were also language and cultural barriers that I had to be aware of and plan for. The role of the researcher might look a little different depending on the type of interview you conduct—so, you need to consider how much do you interact with participants and will you do the things that the participants will do. So, if you are conducting an interview in the field, will you become more of a member of the group you are studying. This is somewhat reflexive, too, because there are various roles that continually develop for the researcher during a project. With underrepresented or hard to reach populations, one of the challenges is gaining access to the population and negotiating that access. So, gaining access is not only important at the beginning of the study, but it is a continuous process that takes place throughout the study. You have to be conginzant and respectful of the participant’s space—especially if you are conducting interviews in the field. Most of the time, you will only gain partial access to a population and when this happens you need to consider if all of the notes you take will serve as data or if you will only be using interview transcripts as data. In this instance, gatekeepers are the key to gaining and negotiating access-–especially for hard to reach populations. So, be considerate of how you are going to gain access and if you do gain access if you will exclude people because you may want to be more inclusive if you are sampling a population that is difficult to reach. With that, be mindful of your sampling strategy and be realistic about how many people you should recruit. Remember for interview studies, you need fewer participants than for studies where you are collecting quantitative data. The goal is to understand unique experiences, so instead of needing 200 participants—so you can generalize information—you only need 5-15. You only need enough participants that no new information comes up in each subsequent interview. So, you keep interviewing people until you don’t hear participants sharing anything new.
5
INTERVIEWS
| STRENGTHS | WEAKNESS |
| Rich, descriptive data! Participant has voice in the research process. Flexibility! | Desirability bias. Time consuming. Control, depth, and scope. Generalizability. |
As with every other method we learn about in this course, there are strengths and weaknesses—pros and cons—benefits and challenges to gathering data using interviews. For instance, the data that you gather is very rich and provides thick descriptions of the research topic you are studying—and these results are very interesting to read. Often, they’re easier to understand by the general public. So, that is a benefit of conducting interviews. Also, the participant has a voice in the research process—this is a much less clinical way of gathering data—and because you get information directly from the participant’s description, the participant feels like they have a more active role in the research process-–and this can alter the implications of the findings. You can also have more or less flexibility depending on the type of interview you conduct. So you have the flexibility to get more or less descriptive information depending on the type of interview you conduct. On the other hand, you are subject to desirability bias---this is due to the researcher’s reflexivity and the role of the researcher in the study. But, again, remember that the goal here is not to predict and control—so, we don’t mind as much that there’s some bias in the study on the part of the researcher. Interviews are also very time consuming to conduct because you have to transcribe the audio recordings from the interviews before you can analyze the data. We also don’t have much control when conducting interviews and the depth and the scope of the topic limits generalizability. So, we cannot generalize our findings. The findings are only descriptive for the sample in the study.
6
FOCUS GROUPS
Interview with more than one individual—group is the unit of analysis!
Number of focus groups.
Homogeneity of group (to what extent do you want the groups to be the same?)
You aren’t just getting one participant’s perspective.
Interaction Key for focus group!!!
Role of Researcher
Facilitating discussion vs. Guiding Discussion.
Notetaking, recording.
Alright, now let’s talk about focus groups! So, first, what exactly is a focus group? Well, focus groups are simply group discussions that are guided by a moderator. The discussion is on a predetermined topic that the researcher decides. Focus groups are very popular in mass media and advertising and they’re often times used if you don’t necessarily have a good idea of what-exactly—you need to ask about a topic. But, you do have a general idea of a broad topic. When we gather data using focus groups, we’ll usually conduct more than one focus group—so, we might conduct five focus groups with 5 to 7 people in each group. One of the benefits of conducting a focus group is that you can gain some variability in your participants perspectives or opinions because you aren’t just getting one participant’s perspective. However, often, the challenging part of a focus group is making sure that all participants engage and interact. Sometimes this can be challenging simply because you may have one or two people who monopolize the focus groups and one or two people who are extremely quiet and perhaps don’t participate. So, the goal of the researcher is to moderate and facilitate the discussion, but not guide the discussion too much.
7
FOCUS GROUPS
| 1. Research question. |
| 2. IS a focus group the right approach? |
| 3. Number of focus groups to hold? |
| 4. Sample participants? |
| 5. Setting? |
| 6. Types of data? |
| 7. Moderator—make sure you have a good one! |
If you are going to conduct a focus group there’s 7 important steps to follow. First, you develop your research question. Second, you decide if the this is the best way to answer that research question. So, why would you want to use focus groups? Well, when conducting a focus group, you can tap into the ‘group effect’ with participants. Because you are getting input and generating input from numerous points of view, each focus group can kind of take on a life of its own and may generate unexpected information from participants. Next, you will determine how many focus groups you will hold and how many participants you aim to have in each focus group. You want to have enough participants that you can get a good variation of information, but not too many that it is difficult for folks to share their opinions—and this number might vary a little depending on the discussion topic. Finally, you will need to decide how you are going to select participants. So, let’s return to our restaurant example, perhaps one of the main selection criteria to participate is that participants must have eaten at the new fancy restaurant you are gathering information about within the last four weeks. In addition, you will need to decide where you are going to conduct the focus groups—are you going to have participants come into a lab or will you hold the focus group in a natural setting? Next, you will need to decide how you are going to record the data—often, the best way to do this is to either record audio for transcription or to record the focus group using video. To do this, you’ll need to decide what type of data you are gathering—verbal or nonverbal? If you just want verbal data, then an audio recording will suffice. If, however, you are interested in nonverbals—then you’ll video record each focus group. Finally, you will need to find or train a skilled moderator. Sometimes, the researcher will moderate the discussion or they might train a research assistant to serve as the moderator.
8
FOCUS GROUPS
| PROS | CONS |
| Fast and cheap! | Moderators |
| Participants at ease—often feed of each other’s responses. | Participants—sometimes they can be ‘flakey’ |
After you’ve selected a location for the focus group, prepared your recording devices, and used your discussion guide to lead the focus group—then you will analyze the results of your focus group! There are different ways to analyze focus group data, but mostly it can be analyzed using grounded theory, content analysis, or conversation and discourse analysis—And we’ll talk more about the analysis part later. But, right now let’s wrap up by talking about some of the benefits and challenges of focus groups. First, focus groups are much faster than conducting individual interviews—which are extremely time consuming. They also cost less than individual interviews and if you are discussing a topic that folks might be hesitant to talk about one-on-one, then a focus group can provide an atmosphere that will make participants feel more at ease and they will often feed off of each other’s responses. Something one person says might remind another participant of something they would like to add. In an individual interview, they might not be as likely to disclose this information. On the other hand, there are some drawbacks and challenges to conducting focus groups. Two of the big ones are moderators and participants. Moderating a focus group can be challenging and if you aren’t careful it is easy for the moderator to become over-involved. Further, sometimes participants don’t show up for one reason or another. So, if you aimed to have 6 to 12 participants in each focus group, but only 3 participants show up to one of the focus groups—this is a problem. So, there’s always a bit of uncertainty regarding whether or not the participants will show up.
9
Trustworthiness
Trustworthiness of positivist research designs (e.g., experiments, most surveys)
Internal and external validity
Measurement reliability
Measurement validity
For Interviews and Focus groups, data is text not numbers!
Interviews and focus groups: TRUSTWORTHINESS.
Credibility
Dependability
Confirmability
Transferability
For credibility we care about internal validity—so, “do the conclusions ring true for the people being studied?” Dependability and confirmability are similar to measurement reliability-–only when we gather qualitative data we aren’t using measures. So, in this way dependability refers to whether the conclusions are “trackable” or whether the conclusions could be verified. Confirmability refers to ate the conclusions based on data or are conclusions based on data or on the researcher’s interpretation of the data. Finally, we have to be concerned with transferability and this has to do with external validity. So, transferability means does the research provide enough information and description of the individuals, groups, and experiences that it accurately describes what is going on if we apply the findings to other contexts. The way that we make sure the findings are trustworthy is by conducting things like member checks, data conferences, and transcriptions. So, with a member check you provide a sample of the study results to one of the participants and see if the results accurately reflect what they shared. Data conferences happen when a few people who are working on the project look at the data individually, then come together and share the findings and see if they agree on what they found. Finally, transcribing the data allows for the people who are analyzing the data to be precise in understanding what the participant was saying—so, we aren’t relying on the interviewers memory—we have a transcript that is word-for-word what was said in the interview.
10