"

Types of Research Designs

Common Research Designs

Although the process of research design refers to designing your specific study, there are common research designs that share characteristics or have a common approach. We’re going to spend the next part of this book delving into many of those designs, addressing how they’re set up and how they’re evaluated. You’ll eventually be able to name them from their descriptions; this will be helpful both for your own reading of research and for planning your research proposal. For the moment, however, let’s just explore the basics of different types so you can begin to consider how the factors you’ve just read about fit together.

Grouping research designs into categories isn’t an exact science. Sometimes people group according to the potential for causality (so you’ll sometimes see experimental, quasi-experimental, non-experimental as classifications). Others will simply group according to how the data themselves are collected. For this book, we’re going to group designs by whether they are quantitative (focused on numbers and broad patterns), qualitative (focused on subjective meaning and descriptive words), or have attributes of both (mixed-methods).

Quantitative Research Designs

Quantitative research designs focus on numerical data and structured measurement. These designs are typically used to test hypotheses, measure variables, and identify patterns or relationships that may generalize to larger populations. One of the most powerful quantitative designs is the experimental design, which is used to examine cause-and-effect relationships. In a typical experiment, researchers randomly assign participants to either a control group or an experimental group. The experimental group receives some kind of treatment or intervention, while the control group does not. By comparing outcomes across the two groups, often using pretests and posttests, researchers can determine whether the intervention caused a measurable change. Variations of this design include posttest-only studies, the Solomon four-group design (which accounts for testing effects), and quasi-experimental or pre-experimental studies, which relax some of the strict requirements of a true experiment when full randomization or control isn’t possible.

Another popular quantitative approach is the survey design, which is used to gather information from a large number of people through questionnaires or structured interviews. Surveys are especially useful for describing population characteristics, identifying trends, or exploring relationships between variables. They can be cross-sectional (a single point in time) or longitudinal (over time), and they are commonly used in social science, public health, and market research. Surveys are well-suited for answering questions like “How common is this belief?” or “What percentage of people engage in this behavior?”

Correlational designs, also quantitative, focus on identifying relationships between variables without manipulating them. These studies can reveal whether two variables are related—for example, whether higher levels of stress are associated with lower job satisfaction—but they cannot establish causation. Correlational research is valuable when manipulation would be unethical, impractical, or unnecessary, and it is frequently used in the early stages of inquiry to guide further experimental or qualitative research.

Qualitative Research Designs

On the other hand, qualitative research designs aim to understand the meaning and complexity of human experiences. These designs typically involve open-ended methods such as interviews, observations, or the analysis of texts and artifacts. Rather than measuring variables, qualitative research explores how people interpret events, make decisions, and navigate their environments. Qualitative research designs are generally categorized by their philosophical approach (e.g., phenomenology), data collection strategy (e.g., case studies), or type of analysis (e.g., thematic analysis). Often these three things go hand in hand, which we’ll discuss in detail in Chapter 9. Qualitative designs are ideal for answering questions like “How do people experience this phenomenon?” or “What is the process by which this outcome occurs?”  While qualitative research does not aim for generalizability in the same way quantitative research does, it provides depth, nuance, and insight that is often critical for understanding social processes and lived experience.

Mixed Methods Designs

Bridging the gap between these two traditions is mixed-methods research, which combines both qualitative and quantitative approaches within a single study. Mixed-methods designs aim to capture the strengths of both—using numerical data to measure scope or impact, and qualitative data to provide context and deeper understanding. For example, a researcher might use a survey to assess how many people benefit from a program and follow up with interviews to understand how participants experienced it. While powerful, mixed-methods research requires careful integration and often benefits from collaboration between researchers trained in each approach. When done well, it can offer a more comprehensive picture than either method alone.

Customizing a Commonly Used Research Design

Choosing a research design is much more than simply looking at the groupings, selecting one from the list, and applying it directly to your question. No single research design fits every question. The key is to match your design to your goals—whether you’re testing a treatment, describing a population, uncovering relationships, or exploring personal experiences. Commonly used research designs can be thought of as templates: you select one to start your project off, but then you are free (and encouraged) to modify bits and parts as needed. In the end, your design should be appropriate to answer your particular question with the resource you have (or would have, if you were actually going to do this study).

The Researcher’s Toolkit: How I Apply What I Teach
The first research study I worked on was quantitative. I was an undergraduate student, and I was offered a summer job doing a simple task: entering data from paper surveys into spreadsheets. At first, it was just a job. I learned to type ten-key and got good at double-checking my entries against the other research assistants’ work to make sure there were no errors in the data we entered. I vaguely understood what we were measuring with those numbers (it was a scale meant to quantify a parent’s stress about being a parent, and it had been measured before and after participating in a parent education program), but I didn’t quite understand what everything meant. I was curious, though – how did such simple numbers, 1 to 5, tell us anything important? My curiosity led my professor to ask if I wanted to help him analyze the data, and I said yes. He opened up his statistics program, clicked a few buttons, and suddenly had a data table in front of him (a t-test, I now know). He showed me each number in that table and explained what it meant. As we worked our way down the table, the results of the study started to take shape, and we could see that, indeed, there were positive changes observed after parents participated in the program. It was a simple analysis of a relatively simple question, but I was amazed! Behavior or attitudes could be measured? By numbers? And then used to tell stories and answer questions about people and families? I was hooked and had to know more. Following this experience as an undergrad (and a few more studies with this professor), I entered a quantitative graduate program under the tutelage of a family demographer. Demography (the study of populations), is not exclusively quantitative, but quantitative methods are most common for the kinds of questions demographers ask, and my professor encouraged me to delve into every statistics class I could. In that program, I continued to develop my quantitative skills, learning how to handle complex datasets, large datasets, and those with unique and tricky characteristics. Throughout my training and my continued career as a mostly-quantitative researcher (I have learned to use a couple qualitative methods now and have collaborated with qualitative colleagues on a few studies), I have continued to feel that there is a certain magic in being able to translate behaviors and stories into numbers and then back again through statistics. Perhaps the best part of this all is that I’ve never considered myself a math person, so I am living proof that you don’t have to love math to love research! You just have to be curious and open to a little bit of wonder. – Dr. Arocho
The Researcher’s Toolkit: How I Apply What I Teach
There’s something deeply human about qualitative research. It wasn’t the first research method I learned, but once I discovered it, I never looked back. For me, qualitative research is where the heart of social science lives—it’s in the stories, the voices, the lived experiences. It’s meaningful, grounded in context, and full of potential for discovery. One of the things I love most about qualitative research is its unpredictable nature. I’ve found that some of the most powerful insights in my work have come not from my original questions, but from the participants themselves—people in interviews or focus groups bringing up ideas, concerns, or patterns I hadn’t even considered. It allows me to explore areas I might never have thought to examine without their input. That openness to discovery keeps the work fresh, relevant, and deeply engaging. Over the years, qualitative research has allowed me to explore difficult topics, amplify underrepresented voices, and ask questions that can’t be answered with a checkbox. It’s not just a method—it’s a mindset, a way of approaching the world with curiosity, respect, and empathy. What I want students to take from this is simple: Don’t underestimate the power of a good question and an open ear. Qualitative research may unfold in unexpected ways, but that’s often where the most meaningful learning happens. Embrace the process—you might be surprised where it takes you. – Dr. Knight
The Researcher’s Toolkit: How I Apply What I Teach
Sometimes a purely quantitative or qualitative design just doesn’t answer your research question. That’s okay! For my doctoral dissertation (a huge, multi-study paper to earn my PhD), I wanted to know what attributes people value most in a romantic partnership. Trust and loyalty? A shared sense of humor? Independence? Closeness and connection? We all want all of these things—but I was curious: if people had to choose just three, what would they pick? Previous studies mostly used surveys, asking participants to rate each attribute on a scale, but that didn’t quite answer my question. I used Q methodology, which gives participants a full set of statements and asks them to rank them in relation to each other—similar to choosing the top three things you’d bring to a deserted island. No one wants to choose, but a person’s answers can tell you a lot about who they are. The set of statements is usually created from qualitative sources – such as focus groups or interviews – and the ranking process can followed by interviews to learn more about why participants ranked items the way they did. This method allows a researcher to gather rich, detailed perspectives (a strength of qualitative research) while still comparing patterns across participants (a strength of quantitative research). That’s just one example of how mixed-methods research can offer a deeper understanding when one approach alone isn’t enough. – Professor Munk

Aligning Your Research Question and Design

Now that you’re familiar with the components of research design, let’s discuss how to make decisions for your own study. As with most things in research, your choice of design depends on your research question. The question is what you want to find out; the design is how you plan to go about finding it.

Research designs are flexible and can be tailored to your specific study. That said, certain types of research questions often go along with certain types of designs or methods. A great way to get familiar with what this looks like in practice is to read the Methods sections of scholarly articles on your topic. Look at how participants are recruited, what kinds of tools are used to measure key variables, what the data collection process looks like, and how the data are analyzed. The more research you read, the more you’ll start to notice patterns.

For example:

  • If your question is something like “How do parents of children with autism experience support services?”, you’re interested in people’s lived experiences. That would likely call for a qualitative design. Studies like this often use interviews or focus groups and analyze the responses by identifying themes and patterns.

  • If your question is “What is the relationship between screen time and academic performance in high school students?”, you’re probably looking at a quantitative, non-experimental study. Researchers might collect survey data about screen time and GPA, then run statistical tests to see if the two are related.

  • If your question is “Does participating in a mindfulness program reduce stress levels in college students?”, you’re asking about cause and effect. That would require an experimental design, where participants are randomly assigned to either a mindfulness group or a control group, and their stress levels are measured before and after the program.

In all of these examples, the design lines up with the question being asked. That’s the goal. Your sampling, measurement, data collection, and analysis should all make sense together and help you answer the question clearly. Looking at how other researchers have done this in your field is one of the best ways to get ideas and learn what works.

If your research question and design  aren’t in sync, the study can end up being confusing or might not actually answer what you set out to explore.

Practical Considerations

Another important part of being a researcher is learning how to make practical decisions about your study design. For example, as you think about your own project, you might imagine that one data collection strategy seems easier to accomplish than another. As a researcher, you have to choose a research design that makes sense for your question and that is feasible to complete with the resources you have. All research projects require some resources to accomplish. Make sure your design is one you can carry out with the resources (time, money, staff, etc.) that you have. Even advanced researchers must sometimes consider their own limitations – a researcher trained in qualitative methods might not be equipped to carry out a complex experimental study. In that case, the researcher would need to either find a fellow researcher with qualitative expertise, or choose to approach their research question from a slightly different angle. Ideally, you should choose a research design that matches your research question, and not the other way around. However, it’s also important to consider your strengths as a researcher and your constraints (such as whether you have time to learn the techniques you want to use).

Can I use AI for this?

Show
You’re the Researcher: Let AI Help—But Don’t Let It Take Over

The final way I’d like to introduce AI’s usefulness is in helping you design your study. You can apply this information to the following four chapters. Remember, when you use  AI to support your research design, it isn’t cheating. The caveat is that you need to be the one making the decisions. AI is only here to assist you through the design process. Once you’ve written your research question, it’s up to you to consider what kind of design might best help you answer it. This includes whether to use qualitative, quantitative, or mixed methods and what tools or procedures you might use to collect your data. Once you have a general direction in mind, AI can help you develop your ideas further. For example, if you’re thinking about using qualitative methods, you might ask AI, “I’m considering using interviews in my study. What are some tips for writing effective interview questions? Please don’t write questions for me. I am just looking for information to get me started?” or “What sample size is typically used in qualitative research like this?” If you’re leaning toward surveys, you could ask, “Are there any existing measurements that have proven reliable and valid that measure [your concept]?”

The key is that you lead the process — you bring your design ideas to the table, and AI helps you refine them, offering examples, pointing out considerations you may not have thought of, and helping you align your method with your research question. In this way, AI becomes a support tool that enables you to grow your understanding and confidence in designing research while keeping you in complete control of your study decisions. As you use AI to help develop your research design, it’s important to remember that you are responsible for checking the accuracy and reliability of the information it provides. If AI suggests a survey or existing scale, such as a self-esteem inventory or a resilience measure, you need to look it up through reputable sources—such as peer-reviewed articles, academic databases, or publisher websites—to confirm that the measure actually exists, is appropriate for your population, and has been validated. In addition, academic integrity requires transparency. If AI assisted you at any point in your study design—whether in brainstorming data collection methods, outlining procedures, or generating sample questions—you should clearly state how AI was used. This models ethical scholarship and helps your instructors understand your learning process. Being thoughtful, honest, and diligent in documenting how you use AI will strengthen your work and credibility as a researcher. I would like to give you an example of how AI might be ethically used in the design process. If you are considering a qualitative design, you might approach AI like this:

  • “I’m thinking of using interviews in my study. Can you help me compare different qualitative approaches, like focus groups or ethnography, so I can decide what fits best?”
  • “I want to create a semi-structured interview guide. Here are a few questions I’ve drafted—can you help me refine them or suggest how to organize them?”
  • “For my qualitative study, I’m planning to interview around 10 participants. What should I consider when choosing a sample size?”
  • “I plan to use thematic analysis to code my interview data. Can you walk me through the steps or help me understand the process?”

If you are considering a quantitative design, here’s how you might engage with AI :

  • “I’ve identified my independent and dependent variables—can you help me double-check if I’ve labeled them correctly and if they make sense for my research question?”
  • “I’m planning to measure self-esteem in my study. I’ve heard of the Rosenberg Self-Esteem Scale—can you help me understand how it works and whether it would fit my population?”
  • “I want to include questions about Instagram use in my survey. Here are a few items I’ve drafted—can you help me refine them or suggest ways to improve clarity and consistency?”
  • “My study looks at the relationship between Instagram use and self-esteem scores. I’m considering a correlational design—can you help me decide which statistical test would best fit my data?”

If you are considering a mixed-methods design, here’s how you might approach AI:

  • “I’m thinking of collecting both interview data and survey results. I’m not sure whether to do them at the same time or one after the other. Can you help me weigh the pros and cons of a sequential versus a convergent design?”
  • “I want to use qualitative and quantitative methods in my study because I think they’ll give me different kinds of insight. Can you help clearly explain how these two approaches complement each other?”
  • “I plan to compare themes from my interviews with trends in my survey data. I have a general idea of how I want to integrate my findings (state what you are thinking)—can you help me make sure I’m thinking about that process clearly and logically?”

No matter which research approach you choose—qualitative, quantitative, or mixed-methods—you are the leader of your study. AI can be a helpful support tool, but it should never be the one making your decisions. It’s your responsibility to develop ideas, questions, and drafts and then use AI to help you refine, improve, or troubleshoot your design. By taking the lead, you protect your work’s integrity and strengthen your skills as a researcher. Let AI support your thinking, not replace it. Your creativity, critical thinking, and ethical judgment will make your study meaningful and credible.

License

Icon for the Creative Commons Attribution-ShareAlike 4.0 International License

Understanding Research Design in the Social Science Copyright © by Utah Valley University is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License, except where otherwise noted.

Share This Book