Building the methodology and research design for my empirical study has been the most challenging part of my PhD so far. It’s not that I haven’t conducted qualitative or quantitative research before; it’s just that the techniques used in industry differ significantly from their academic application. After the last catch-up with my supervisors, we agreed that a workshop would be necessary to determine the best methods for my proposed study. It’s crucial to understand the pros and cons of each method and assess whether it’s the right one for the job.
We have established that conducting interviews in the first stage of the study, followed by a survey in the next stage to evaluate the findings, is the best way forward. During the workshop, we will share examples of qualitative data, narratives, and experiences that matter, and explain how they differ from official, positivist, or institutional descriptions. The goal is to determine what can be gained from an interview or account that cannot be obtained otherwise.
So, I am writing this blog post (it was supposed to be a mini-essay!) in preparation for the workshop. I plan to weave in examples from my work experience, where I have used interviews, how I led them, and the outcomes. I will also research how and why interviews are conducted in academia.
I have an excellent book by Sharan B. Merriam and Elizabeth J. Tisdell on qualitative research, which has been my go-to resource for research design. According to the authors, interviewing is a technique used to gather data for qualitative research. It’s also worth mentioning that qualitative research centres around words, whereas quantitative research focuses on numbers and answers questions about “when” or “where.” Qualitative research, on the other hand, addresses questions like “how” and “why” and aims to understand human behaviour.
In qualitative research, interviews are often the only way to collect data. The most common type is person-to-person, where the researcher and participant engage in a conversation focused on study-related questions. This is a valuable method for uncovering things that are difficult to observe, such as feelings, thoughts, and intentions. It allows the researcher to understand another person’s perspective without experiencing it in real-time. Interviews are also useful for understanding how people interpret the world around them.
It’s important to use this technique purposefully. Interviews are a great tool when other data collection methods are too expensive or fail to gather sufficient insights. For example, surveys can capture certain data, and while they are primarily used for quantitative data collection, I’ve used them in the past to gauge sentiment on specific topics to shape other stages of research. However, surveys cannot provide the richness of a story that a participant can reveal through a less structured approach. They also mean you might only gather reactions to the researcher’s preconceived notions rather than accessing the participant’s perspective and understanding of the world. Additionally, surveys assume respondents share a common vocabulary, though they are useful for gathering sociodemographic data such as age, income, and marital status.
Interviews can be:
structured (where you ask all participants a set of specific questions),
semi-structured (where you have a few key questions but allow the conversation to flow into unexpected areas), or
unstructured (where no questions are prepared and the participant guides the conversation).
The choice of interview format depends on the level of structure required to answer the research question.
Less structured approaches assume that individuals define the world in unique ways. It’s useful to have some questions ready that can be flexibly adapted, allowing you to explore the paths a participant may lead you down. I like to have a few questions prepared in case a respondent is shy and needs some help, but it’s best to let the participant guide the conversation, exploring how they feel, think, and interpret things. A fully unstructured interview works well when the researcher doesn’t know enough about the phenomenon being studied to ask specific questions. However, fully unstructured interviews are not easy and take time to master. For those new to conducting interviews, semi-structured interviews are often the best option. It’s also possible to combine all three methods. For example, you can start with unstructured interviews, then move to semi-structured as you gain more insight into the phenomenon, and eventually conduct fully structured interviews.
In my case, the area I am researching is a phenomenon that isn’t well defined, so starting with an unstructured approach will be best. Depending on the approach you choose, you will also ask questions differently. The authors of the book I mentioned provide an example of asking about the role of mentoring in the career development of master teachers. In a structured interview, you would provide a definition of mentoring and ask the respondent to think of someone who is a mentor. In a semi-structured interview, you would ask the respondent to describe what mentoring means to them. In an unstructured interview, you might simply ask how they became a master teacher.
In one of the projects I worked on, we were asked to consolidate two servers into one to make it easier for users to find documentation. One of the account managers felt we needed to learn more before migrating all the documentation, so he asked me to interview a few users. I conducted 20 interviews with stakeholders at different levels in the organisation, all of whom used the servers to access various types of information. During one interview, a participant mentioned audit trails and how moving all the documents would result in losing vital audit information, which could be an issue if an audit were conducted in the future. Some documents had been in use for years and required their version history to remain intact. After probing further on the topic and asking other participants, we realised this issue needed to be addressed in the project scope. Instead of merging both servers, we built a user interface that acted as a gateway between the two. Users believed they were accessing one server, but both were working in the background. All new documentation would be stored on a single server, allowing us to shut down the other once no new documents were being added.
This example highlights the importance of giving users a voice and allowing them to contribute their expertise. The person who initially mentioned the audit trail issue had tried to raise it with someone but didn’t know who to approach or how to frame the concern. It was only after confirmation from others that we realised this was a significant issue.
The key to good interviews is asking good questions, which isn’t always easy. This takes practice. A great way to refine your questions is through pilot studies. These studies help test your thinking and refine your questions. They allow you to see if your questions are confusing and whether they generate the data needed to answer your research questions.
The way you phrase questions can yield different results, so it’s crucial to consider what you’re looking for and what participants might know about the topic. Ask yourself if the way you phrase questions or your knowledge of the subject could influence participants’ responses. Sometimes, you might be working in areas where the terminology differs from common usage, so you need to consider this. For example, I am conducting a study on Design Ethics, but people rarely use the term "ethics." They tend to talk about responsibility, sustainability, or transparency. So, should I use the term “ethics” in my studies? Perhaps if I provide a definition first, but this could influence participants with my version of the truth and lead to more subjective data. It may be better to ask questions surrounding the topic without directly using the term.
It’s also important to phrase questions in a way that makes sense to participants. Avoid jargon and use familiar language. When asking questions, remember not to ask multiple questions at once. The questions should be clear and easy to follow. If a participant asks you to repeat the question, it may be too long or convoluted. It’s also crucial to avoid leading questions, as they reveal your bias or assumptions. For example, asking “What emotional problems have you had since losing your job?” assumes that everyone who loses their job has emotional problems. Yes-or-no questions should also be avoided. If a question can be answered with a simple “yes” or “no,” it’s likely better suited for a survey. These questions can disrupt the flow of the conversation. Instead of asking, “Do you like the program?” ask, “What is it about the program that you like?”
In my work, I’ve conducted interviews where yes-or-no questions were necessary because we were working with other teams and combining our needs into one interview. I try to leave these questions until the end to allow the interview to flow, asking more structured questions last. The type of interview you choose will depend on the project and audience. I’ve used journey maps to gather specific data about a product or service experience because I knew exactly what I was looking for. Could I have done this with open-ended questions? Possibly, but I had to weigh up whether I would gather the data I needed.
Probing is a crucial technique. If a participant says something that catches your attention, it’s a good idea to probe further. You may be surprised by the direction the conversation takes. While it’s helpful to have an interview guide, even for semi-structured interviews, remember that questions don’t have to be asked in a specific order or with specific language. Nonetheless, having backup questions is useful, especially for less verbal participants or when probing deeper. Silence is also an effective probe. People tend to avoid silence and will often try to fill it, which may lead to interesting insights.
It’s impossible to predict exactly how the conversation will unfold, so the interview guide can only take you so far. If you sense that something important is emerging "off script," follow the lead. It might take you to new and unexplored areas. The guide will evolve with each interview, and that’s perfectly acceptable. Consistency is important, but if you’re only looking for answers to specific questions, a survey or highly structured interview might be more appropriate.
Interviewing is my favourite method during discovery in a project. It’s an excellent technique for gathering information through direct dialogue. People generally enjoy sharing their stories, and I’ve found that many appreciate being given the space to talk about a subject. Through interviews, you gain a deeper understanding of people and their views by eliciting their true feelings, desires, struggles, and opinions with carefully crafted questions or probes.
It’s important not to interrupt the participant. Listen carefully to what they are saying, and probe for more information where necessary. Avoid the temptation to put words into their mouth—the goal is not to hear what you want to hear, but to understand how the participant views the world. Don’t offer your opinions or analyse their responses during the interview.
Location also plays a role in interviews. The setting can influence the conversation. Whether the interview takes place in someone's office while discussing their work or outdoors in a park, it can lead to different outcomes. Ensure you choose a location with minimal distractions where participants feel comfortable. In a virtual setting, while you may not control the participant’s environment, you can check at the start to ensure they are in a suitable space for the interview.
With the internet and online communication software so readily available, online interviews have become more common, especially with the rise of remote working due to the COVID-19 pandemic. I have conducted many successful online interviews, allowing me to reach a broader audience, such as people with disabilities or those in different countries. However, building rapport virtually can be challenging, and observing non-verbal cues is more difficult. I’ve also had participants who felt uncomfortable turning on their cameras or had technical issues, which meant I couldn’t observe facial expressions. Even so, during the COVID pandemic, the team I worked with on the NHS Test and Trace project conducted all user research virtually, and we were able to gather the necessary data.
Interviews are fundamentally about listening. In my experience, people really appreciate being heard, especially when they are passionate about the topic. Making participants feel comfortable, safe, and heard is key to a successful interview.
Interviews are an excellent technique for gaining direct insights. They challenge preconceptions, help uncover the real situation, deepen empathy for others, and bring diverse voices into a project. The data and insights gathered will help build credibility with stakeholders.
Participant interviews are essential because they provide direct access to the thoughts, feelings, and experiences of the people you're designing for. They allow you to gather authentic insights that cannot be obtained from secondary data or assumptions. By engaging with participants, you gain a deeper understanding of their needs, challenges, and desires, which helps inform better design decisions. Interviews also offer an opportunity to uncover nuances that might be missed through other methods, such as surveys. They foster empathy, ensuring the design process is centred around real user experiences and not just theoretical concepts. Ultimately, participant interviews help create solutions that align with user needs, improving a project's overall quality and impact.
References:
Qualitative Research: A Guide to Design and Implementation, 4th Edition https://www.wiley.com/en-sg/Qualitative+Research%3A+A+Guide+to+Design+and+Implementation%2C+4th+Edition-p-9781119003618
LUMA Workplace: https://www.lumaworkplace.com/methods/interviewing
Comentários