Semi Structured vs Unstructured Interviews

Week 9

Description of Tasks

Give a detailed description of what you did this week regarding research and community engagement.

This week, I reviewed the feedback I obtained from the participants in the interview I conducted. The piloting process appeared to be successful insofar as I was able to obtain relevant answers to the questions I asked. Those questions allowed me to obtain data that could help me to answer the research question. But the data also let me see that there were other questions I could ask as well. I view this as a success because through this process I was able to see where I could probe a little more deeply to obtain a better and fuller understanding of the issues pertaining to why some parents choose to send their children to private schools instead of public schools.

One modification I may make to facilitate better interviewing technique will be to switch from a structured interview process to a semi-structured interview process. That way I can ask questions that might occur to me during the interview and I can probe for deeper data based upon the information supplied at the time. Burgess (1982) points out that the “unstructured interview as a conversation” can be beneficial for deeper analysis (p. 107).

After reviewing ways to support interviewing, I also decided I should ask participants for their permission to conduct follow-up inquiries through email. This will be helpful for when it is time to analyze the data. I can confirm with the participants that the way I am reading their responses is appropriate to what they intended to communicate. This method is known as member checking, or respondent validation (Barbour, 2001).

Another option is the survey instrument, but I have decided against this because what I want is to investigate this issue deeply and that means having the option to explore what is going on in the lives and minds of parents, what their perceptions are, and what factors contribute to their decision-making. The semi-structured interview process will give me the opportunity to do the kind of investigative work that needs to be done. A survey instrument is best suited for confirming or rejecting a hypothesis. But since I have not formed a hypothesis, it does not suit my needs. My research and the analysis I do will lead me to have a better understanding of the situation.

Next Steps

Give a brief description of what you plan to do next.

What I plan to do next is pilot a semi-structured interview with three additional participants from my community. The original participants were helpful in that they gave me references that I could contact to gain more insight into this issue. I have already contacted three new participants based on the references and they have agreed to do a semi-structured interview with me over phone. I am setting aside time for each one, and estimate that 30 minutes will be sufficient to test the piloting of this new approach. I will provide information to the participants so that they can give their informed consent, as is appropriate.

Capstone Questions

What questions do you want to bring to the learning community or professor

When conducting a semi-structured interview, is it better to record the interview or to take notes during the process and refer to the notes later?

My Reference Notes

Barbour, R.S. (2001). Checklists for improving rigour in qualitative research: a case of the tail wagging the dog. BMJ 322, 1115-1117.

This source was helpful for giving me information on ways to improve qualitative research, using the interview process. It suggested member checking as a way to make sure my analysis of data given by participants is actually aligned with what the participants meant to say. It requires obtaining feedback from participants to validate my analysis. It is a good way to bolster qualitative research and make it more credible and as objective as possible.

Burgess, R. G. (1982). The unstructured interview as a conversation. Field research: A

sourcebook and field manual, 107-10.

This source was helpful in that it provided me with support for revamping my interview approach. I went into the first pilot with a structured interview, but I felt like while the answers I obtained were good the interviews could have gone a little deeper. With the semi-structured approach, there is more room to ask questions on the fly. There is also the unstructured interview approach, which can yield very good results by allowing the interview to proceed more like a conversation. This can let organic ideas rise to the top. I may have to consider the unstructured process and might try to pilot an interview using the unstructured approach to see if it is more helpful than the semi-structured. At this point, I feel like my interviews need some sort of guidance, which is why I want to have a few questions ready to go, but that may change if I do an unstructured interview and find it to be most helpful in terms of what I am trying to accomplish here.

There are some more materials I want to read over the following week in terms of understanding how to properly analyze interview data. I think that what I need to do, whether it is taking notes or recording and printing out a transcript, is to focus on ideas or themes that come up and use these to support my understanding of the picture that emerges from the data.

Week 10

Description of Tasks

Give a detailed description of what you did this week regarding research and community engagement.

This week, I modified the interview I conducted by inserting new questions I wanted to ask and also I allowed myself the opportunity to probe a little more deeply with the three new participants I interviewed this week. I found this method to be even better than the first, because with the structured interview I felt like I was not obtaining deep enough data. With the semi-structured approach, I was able to probe further and get interviewees to expand upon their answers.

Each interview went about twenty minutes and I took notes throughout. When I compared these notes to the first pilot’s notes, I found that this new method did indeed yield improved results. However, now I am wondering if I may be steering my interviewees in certain ways that I might not realize. It is important to be as un-biased as possible when gaining information from participants (McCambridge et al., 2014).

At the same time, Collins and Cooper (2014) state that an interviewer should have a high degree of emotional intelligence in order to conduct an interview effectively. If I am going to do an unstructured interview, I want to make sure I am polite and can respond appropriately to emotional cues given by the participant. I will need to read some more on emotional intelligence and test out my own emotional intelligence skills before I conduct an unstructured interview—but I am excited about doing one. I think it will be helpful to compare notes from an unstructured interview to a semi-structured interview and a structured interview. I will then be able to see which pilot yields the best results and I will use that approach for conducting the rest of my research.

For the three participants for this round of piloting, I interviewed three parents recommended by the first round of interviewees. They all had children in private schools and all of them were in high school. The parents said they chose private schools because they did not trust the public school system, they wanted their kids to get a good education, they were happy with the private school education system, and they felt it was the right approach for them but may not be for everybody.

Next Steps

Give a brief description of what you plan to do next.

For my next step I will test my own emotional intelligence skills for an unstructured interview by observing people in public and listening to tones when I converse with others. Part of what makes emotional intelligence so important is that one can hear tones or read body language and respond in supportive ways that satisfy the other person. I will need to be able to do this if I want to have a successful unstructured interview for my next pilot.

The next pilot will be unstructured, but since I have already done two pilots with six participants total, I have a good idea of what questions I would like to ask. However, what I will do with an unstructured interview is this: I will begin by identifying myself and asking if I have permission to interview the person. I will explain the purpose of the interview and what I am trying to do with the information I obtain. I will then ask the person if he or she has chose private school and, if so, what factors led to that choice. Based on the responses given, I will add in additional questions so as to get the respondent to elaborate more deeply. I do not expect that the interviews will proceed in the same manner for this next set of participants. But it will be interesting to see what happens.

Capstone Questions

What questions do you want to bring to the learning community or professor

When conducting an unstructured interview, is it a good idea to have notes on hand that can help you to guide the interview or does this go against the purpose of an unstructured interview? If the idea is to let the conversation develop organically it seems that one should not rely on notes, but I am a little nervous about trying to obtain information without them.

My Reference Notes

Collins, C. S., & Cooper, J. E. (2014). Emotional intelligence and the qualitative researcher. International Journal of Qualitative Methods, 13(1), 88-103. doi:10.1177/160940691401300134

This source was helpful for explaining how important emotional intelligence is when it comes to conducting qualitative research. Emotional intelligence is described as the ability of being able to interpret other people’s emotions as well as your own so as to control feelings and give positive reinforcement. The idea is that the more one understands human behavior, how humans communicate in ways that aren’t verbal, and how to control emotions, the more likely one is to be able to navigate any situation successfully. This was a good resource for giving me the confidence to want to try to do an unstructured interview to see if it would be any more helpful than the other two pilots that I conducted over the past two weeks.

McCambridge, J., Witton, J., & Elbourne, D. R. (2014). Systematic review of the Hawthorne effect: new concepts are needed to study research participation effects. Journal of clinical epidemiology, 67(3), 267-277.https://doi.org/10.1016/j.jclinepi.2013.08.015

This source was helpful for explaining how bias can affect research. It uses the example of the Hawthorne effect in which subjects in a study realize they are being watched so they alter their behavior in ways that they think will satisfy the observers. This means that whatever data is obtained is not an accurate reflection of reality. I wonder therefore it would be better to interview a participant without letting that participant know what I am doing. But of course that would go against the principle of obtaining informed consent, so I can’t do that. It does raise an interesting question, however, about how accurate one’s data can be if there is always a risk of some type of bias emerging.

Week 11

Getting Organized to begin conducting your study – Set timelines for study

-Maintain a record of written and verbal reflections related to the challenges posed by the project which is supported by the feedback with the organization, community, or college mentors.

-Self-reflect on one’s own learning, achievement of goals, and contributions to learning.

Description of Tasks

Give a detailed description of what you did this week regarding research and community engagement.

This week, I compared notes from the three pilots I conducted and found that the semi-structured approach worked best for me. When setting timelines for my study, my goal is going to be to give 30 minutes to each participant that I interview and to have a total of 12 participants interviewed for the study. Once I have obtained the data, I have to begin analysis of it. I want to give myself two weeks to go over the data and conduct member checking so that I can validate my analysis of participant responses. This will give me time to identify themes that occur among the responses and to develop an idea of what factors contribute to parents wanting to pursue private school for their children. During this time I also intended to further my studies on relevant literature that has been published on this issue.

Another thing I want to do is to consider what public schools are doing differently from private schools and to what extent public schools are aware of this difference. If the goal is to help public schools catch up to where private schools are, I need to know where both are and how parents are perceiving each. Therefore, it might be helpful to interview parents of students who attend public school as well.

So far in the community, I have spoken to 9 parents about private school. The last pilot I conducted was unstructured and the interviews went for about 15 minutes each. I did not go off any notes and let the conversation develop organically. Each one was very different. Some participants were more talkative than others, but since I was not trying to guide the interview, I let it go where the participants were willing to take it, only prompting them to add more information when something was said that I wanted more information about.