You know what’s really delicious? Of course—we all do! Maybe you love a nice cheesy pepperoni pizza with a spicy, garlicky sauce and crispy crust. Yummy. Or, maybe you’re a fan of sweets like a decadent chocolate mousse with the perfect smooth, creamy texture and deep, rich flavor. Yummy! Sorry in advance, but I’m going to ruin the moment for you now. It’s for a good cause, though, I promise! But, imagine that the waiter at your favorite pizza joint arrives at your table with what should be the pizza of your dreams, but when you take your first bite you encounter the distinct taste of peppermint in the sauce. Gross! So you seek refuge at a gourmet dessert shop, but the chocolate mousse that is your heart’s desire is lumpy and has a faint scent of vinegar. Not at all yummy!
The point of this horrifying hypothetical is to help with your dissertation or thesis by illustrating that your recipe—or research design—matters. Incorporating certain ingredients, and excluding others altogether, is essential for creating the perfect pizza or chocolate mousse. Further, ingredients that are great in one recipe might not mix so well with another recipe. Take our peppermint and garlic pizza sauce for example. If that peppermint were mixed with the dark chocolate used for our mousse, that might be tasty! But, peppermint and garlic? Not so much. The same goes for the vinegar flavor we sadly encountered in the chocolate mousse; if that had been a dill pickle or sauerkraut, the vinegar tang would be most welcome! But that flavor doesn’t really combine so nicely with chocolate, does it?
Hence the importance of recipes, and the same principle can help when putting together the perfect combination of elements for your thesis or dissertation. In fact, this is the essence of alignment, which you’ve undoubtedly been hearing about, whether you’re planning quantitative or qualitative research for your dissertation. Your taste buds help you to determine whether ingredients in a recipe fit with one another and with the dish as a whole, and you can also train yourself to detect alignment (or misalignment) among key ingredients in your study. Familiarizing yourself with the optimal composition of dissertation elements will help you to identify a misaligned study as easily as your palate detects peppermint in pizza sauce (ew!). To refine your appreciation for a gourmet study requires understanding the recipe behind the masterpiece.
This brings us to the focus of this article: research design. Research design is to your study what a recipe is to your favorite meal: a list of ingredients and procedures for bringing those ingredients together to create an end product that is harmonious. Let’s look at a definition of research design to get this discussion going, and a ResearchGate (2019) user offers a fabulous one:
The research design refers to the overall strategy that you choose to integrate the different components of the study in a coherent and logical way, thereby ensuring you will effectively address the research problem; it constitutes the blueprint for the collection, measurement, and analysis of data.
You can see how this definition brings to light the commonalities between research design and recipes. We can consider the study’s end result of effectively addressing the research problem as the final “dish” created by the recipe, and we’ll dig into some specific research designs in following sections to examine how key ingredients are assembled to create your quantitative or qualitative research masterpiece. Specifically, we’ll look at the purpose statement, research questions (RQs), theoretical framework, and data collection as essential ingredients to get just right to make sure that your finished study is an impressive success!
Research Designs in the Qualitative Research Paradigm
If we think of key elements of a study as ingredients in a recipe, we can see that picking out certain elements will be very important for creating a delicious and tasty study (well, figuratively anyway). But, just as we must do when selecting or developing a recipe, we must first have some sort of idea of what we want the end result to look like. Are we making a cake? A fruit salad? Maybe a hot dog on a stick? These basic questions are important to answer before embarking on your cooking efforts; and, we need to have the same sort of vision of the end product when selecting a research design as well. In the following sections, I’ll share a basic overview of several types of designs in the qualitative research and analysis arena, and then we’ll go over how to put together an aligned list of “ingredients” for each.
Phenomenology
In our work to help clients who are completing their dissertations, one of the most frequent research designs we encounter is phenomenology. In planning for a great recipe for a phenomenological study, here’s what we need to keep in mind:
The focus of this methodology is on understanding the unique lived experience of individuals by exploring the meaning of a phenomenon. From this descriptive data, further interpretation and analysis enables the researcher to uncover a description of the ‘essence’ of the phenomenon; the universal meaning for individuals. To derive the essence, the researcher puts to one side their own views of the phenomenon, referred to as bracketing, in order to deepen their understanding. (Petty, Thomson, & Stew, 2012, p. 379)
Understanding that the “meal” a finished phenomenological study represents needs to resemble the above will help us to determine how our ingredients need to look. In short, the study’s ingredients need to all work together to support qualitative analysis that yields an in-depth understanding of participants’ own particular experiences as viewed from their own (not the researcher’s) perspectives. Here is a sample list of ingredients for this phenomenological delicacy, with brief explanations of why these individual elements work so well together:
Purpose. The purpose of this qualitative phenomenological study is to explore young African American men’s lived experiences of encounters with police officers in the course of roadside pullovers.
Theoretical framework. Critical race theory.
RQ1. How do young African American men perceive their lived experiences of encounters with police officers in the course of roadside pullovers?
RQ1a. How do young African American men perceive their interactions with police officers regarding search and seizure during the course of roadside pullovers?
Data collection. Semi-structured individual interviews with African American men between the ages of 18 and 25.
Why do these ingredients work well together within this recipe? Qualitative analysis within the vein of phenomenology should yield understanding of specific lived experiences from the perspectives of participants, with an aim of privileging their accounts and interpretations of experiences that others might not readily understand. You can see how young African American men’s experiences with police officers might be important to look at through this lens, and given current media coverage and findings from the recent literature, there is reason to believe that racially disparate treatment might be discussed by participants. For this reason, critical race theory blends well with the topic and the purpose.
If the RQs look repetitive when compared with the purpose, it’s because they are and should be! This creates consistency from the purpose to the RQs, which then facilitates data collection to support qualitative analysis that in turn yields findings that truly address the problem and purpose. As for the data collection procedures, semi-structured interviews blend well with this recipe because they allow the participants to share the sorts of individual experiences and personal interpretations that are central to phenomenology.
Let’s take a moment and imagine that you instead used a survey with a Likert-type scale to ask these young men about their experiences. Do you think they would be able to share the nuances of their experiences and interpretations by checking off answers on a 1-to-5 scale? If the whole point of phenomenology is to hear the participants’ stories and thoughts on what they’ve experienced, a survey doesn’t really accomplish that, does it? Using a survey for phenomenology—or any qualitative research—would definitely create a bad recipe! Like adding vinegar to your chocolate mousse…not good!
Case Study
Another very common qualitative research design we come across in the course of our dissertation assistance with clients is the case study design. Here’s a description of the case study design:
The more that your questions seek to explain some present circumstance (e.g., “how” or “why” some social phenomenon works), the more that case study research will be relevant. The method also is relevant the more that your questions require an extensive and “in-depth” description of some social phenomenon. (Yin, 2014, p. 4)
Sounds like a delicious form of qualitative research and analysis, and let’s consider how we might assemble our ingredients for this tasty treat!
Purpose. The purpose of this qualitative single case study is to explore dynamics related to decision making regarding romantic relationships among adults with intellectual disabilities living in group home settings.
Theoretical Framework. Self-determination theory.
RQ1. How does decision making take place regarding romantic relationships among adults with intellectual disabilities living in group home settings?
RQ2. How do adults with intellectual disabilities living in group homes perceive decision making with regard to their own romantic relationships?
RQ3. How do paid supporters of adults with intellectual disabilities living in group homes perceive decision making regarding their clients’ romantic relationships?
Data collection. Semi-structured interviews with adults with intellectual disabilities who live in group homes, semi-structured interviews with paid supporters of these individuals, observations of interactions within group homes.
Why do these ingredients work well together within this recipe? If you consider that the case study design is all about using qualitative analysis to gain an understanding of how or why complex social phenomena occur, you can see pretty clearly how an examination of the decision making processes regarding romantic relationships for adults with intellectual disabilities might fit the bill. Although adults with intellectual disabilities are entitled to the same rights of social interaction as any other adult, they also live within a social and cultural context in which they are often infantilized and deprived of choice-making due to care providers’ perceptions of them as being “eternal children.” Given this context, along with paid supporters’ responsibility to promote client rights while also protecting their health and safety, you can imagine how the decision making process regarding romantic relationships can become quite complex. Hence, the topic is a perfect fit for the case study design.
Now, consider this general topic, adults with intellectual disabilities and romantic relationships, if your final vision for the study was somewhat different. Let’s imagine that you’re really interested in learning about whether adults with intellectual disabilities are more likely to have romantic relationships if they live in group homes, in their parents’ homes, or in their own homes. There is still complexity here, but you aren’t so much looking at the “how” and “why” aspects of these complex processes of decision making; you’re really looking at the relationships between variables—and, to answer this question you’d obviously need to recruit from a large variety of home settings to get a comprehensive picture of the relationship between home setting and romantic relationship status. If you’re thinking this sounds like a recipe from a quantitative rather than qualitative analysis cookbook, you’re right!
We’ll get to a discussion of quantitative methods and designs (i.e., those involving statistical analysis) soon, but before we do so, let’s first consider the remaining ingredients in our delicious little qualitative research masterpiece here. Given our focus on decision making and rights of adults with intellectual disabilities, a framework of self-determination theory blends well with the recipe because of its inclusion of autonomy as a core dimension. Also, self-determination theory includes relatedness and competence as key components, which makes the theory quite fitting considering our focus on relationships (and possibly the skills and knowledge—competence—necessary to make informed decisions about relationships). Finally, note that the multiple RQs allow us to look at the decision making process from multiple perspectives. This goes perfectly with our aim in a case study, which is to use qualitative analysis of multiple data sources to cultivate a rich and in-depth understanding of these complex social processes. This recipe is so masterfully constructed that if there were an Iron Chef for dissertations, you’d stand a solid chance of taking the crown!
Research Designs in the Quantitative Research Paradigm
As promised, we will now open the quantitative cookbook and peruse a couple of recipe options to help you create the dissertation dish of your wildest dreams. In other words, if Ina Garten set her sights on creating your dissertation, here’s what she might come up with!
Correlational
For our clients who are working on quantitative dissertations, the correlational design is definitely one of the most common choices. Now, it doesn’t take a seasoned statistician to construct a correlational study, but it’s important to understand what these studies can and cannot do. So, what kind of recipe is this?
Correlational designs involve the systematic investigation of the nature of relationships, or associations between and among variables, rather than direct cause-effect relationships. Correlational designs are typically cross-sectional. These designs are used to examine if changes in one or more variable are related to changes in another variable(s). This is referred to as co-variance. Correlations analyze direction, degree, magnitude, and strength of the relationships or associations. (Sousa, Driessnack, & Mendes, 2007, p.504)
As we did above with our qualitative research designs, let’s now look at a sample quantitative correlational study and then pick it apart to determine how and why these ingredients come together so harmoniously.
Purpose. The purpose of this quantitative correlational study is to examine the relationship between perceived organizational justice and organizational citizenship behavior among production workers in the Southeast United States.
Theoretical Framework. LMX theory.
RQ. Is there a relationship between perceived organizational justice and organizational citizenship behavior among production workers in the Southeast United States?
Data collection. Surveys using validated instruments for perceived organizational justice and organizational citizenship behaviors.
Why do these ingredients work well together within this recipe? First off, you can see that the purpose statement and RQ are quite consistent with the overall description of a correlational study. Unlike in our qualitative analysis examples, you’ll see that there are clear variables (perceived organizational justice, organizational citizenship behavior) that lend themselves to statistical analysis, which are stated along with the goal of assessing how those variables relate to each other. Also, note that there is no hint of causality implied in the purpose or RQ, which blends well with the correlational recipe. Even though we might guess that positive views of organizational justice might cause workers to behave in certain ways, we know that we can’t truly assess that through statistical analysis of data collected in a cross-sectional study. One easy way to ruin the recipe is to try and throw causality into the mix—like suggesting you’re going to examine the “impact” or “influence” of perceived justice on citizenship behaviors. That’s a correlational recipe no-no for sure!
Now, what about LMX theory as a framework? How does that mix with organizational justice and citizenship behaviors? Well, consider the root of LMX theory: the relationship between leader and follower. According to LMX theory, workers will be more committed and loyal to the organization if they perceive their relationships with their supervisors to be respectful and positive. We can imagine that employees who have such positive views of their supervisors might also perceive their organizations as just and fair; and we can also imagine that the commitment and loyalty this inspires might lead workers to behave in ways that are helpful or beneficial to their coworkers and the organization more broadly. If this were so—if our statistical analysis yields significant results, in other words—then LMX theory would go a long way in explaining why perceived justice is related to organizational citizenship behaviors. Another masterpiece!
Causal-Comparative
Another very common research design we encounter often in the course of our dissertation assistance to clients, when they choose to work within the quantitative paradigm, is the causal-comparative design. When we envision the finished creation that emerges from a causal-comparative recipe, what do we see?
Comparative studies are also called ex post facto or causal-comparative studies. These studies describe the differences in variables that occur naturally between two or more cases, subjects, or units of study. Researchers who use a comparative design normally pose hypotheses about the differences in variables between or among two or more units. (Sousa et al., 2007, p. 504)
Let’s evaluate another hypothetical concoction to see if we’re up to Ina’s standards (if Ina were inclined to quantitative research and statistical analysis, that is):
Purpose. The purpose of this quantitative causal-comparative study is to examine the differences in grades and degree completion across undergraduate students in face-to-face, online, and hybrid courses.
Theoretical Framework. Adult learning theory.
RQ1. Is there a difference in grades across undergraduate students in face-to-face, online, and hybrid courses?
RQ2. Is there a difference in degree completion across undergraduate students face-to-face, online, and hybrid courses?
Data collection. Course modality, course grades, and degree status, drawn from existing university data.
Why do these ingredients work well together within this recipe? You’ll notice right off that the purpose and RQs are very clearly focused on differences that can be assessed using statistical analysis—not relationships, not impacts or influences, not experiences or perceptions—differences. This is a key ingredient in a causal-comparative study. Notice also that, again in contrast with our qualitative research recipes, we also have clearly defined variables, grades and degree completion, that we will compare across the multiple cases that are represented by the undergraduate students in our three different learning contexts (face-to-face, online, hybrid). Important to note is that we’re collecting data from existing records, which means that our statistical analysis of this data will support identification of naturally occurring differences, if any. We’re not engaging in any type of random assignment to treatments as we might do if crafting a study in the experimental vein; instead, we’re looking for any differences that might be present across these groups of students in “real life.” As a final essential piece of our study, we have the theoretical framework of adult learning theory. As our undergraduate students will all be adult learners, this theoretical framework brings the other ingredients of our study together like the mozzarella on your pizza!
What About Mixed Methods?
Given that mixed methods studies require the blending of quantitative and qualitative research and analysis, requiring alignment of purpose, RQs, and data collection to both qualitative and quantitative aims and procedures, it takes an experienced chef to master such recipes! In our provision of dissertation assistance to a wide range of clients, we do come across those brave souls who are ambitious enough to try their hands at such a complex dish. So, having gone over the basics of a couple of quantitative and qualitative research designs, how about a thought exercise to close out our discussion? Mixed methods studies include both qualitative and quantitative data collection, which means that both statistical analysis and some form of content or thematic (qualitative) analysis are involved. Data may be completed simultaneously (mixed methods concurrent design) or with one type of data collected first, followed by the second type of data (mixed methods sequential). For the purposes of our thought exercise, we’ll use a concurrent design, which is described thusly:
Concurrent designs have been useful in studies where sequential designs were impractical, when the ordering of data collection was irrelevant, or when the need for multiple kinds of data for a given time period was pressing. (Small, 2011, p.68)
Purpose. The purpose of this concurrent mixed methods study is to examine the perspectives of blog readers who take carrot cake breaks regarding their subjective happiness as it relates to the proportions of cream cheese frosting, spice, and cake.
RQ1. Is there a relationship between subjective happiness and frosting-to-spice-to-cake proportions among blog readers who take carrot cake breaks?
RQ2. How do blog readers describe their affective experiences of carrot cake breaks as these specifically pertain to the components of cream cheese frosting, spice, and cake, and their relative proportions?
Theoretical framework. Optimal tastiness theory (OK, I made that up).
Data collection. Surveys using validated instrument for subjective happiness, semi-structured individual interviews with blog readers with histories of eating carrot cake with cream cheese frosting.
Now it’s your turn to answer – Why do these ingredients work well together within this recipe?
As a closing note I’ll say that if you made it this far without having a snack, you are a wonder! Bon appétit!
References
- Johnson, R. B., Onwuegbuzie, A. J., & Turner, L. A. (2007). Toward a definition of mixed methods research. Journal of Mixed Methods Research, 1(2), 112-133. doi:10.1177/1558689806298224
- Petty, N. J., Thomson, O. P., & Stew, G. (2012). Ready for a paradigm shift? Part 2: Introducing qualitative research methodologies and methods. Manual Therapy, 17, 378-384. doi:10.1016/j.math.2012.03.004
- ResearchGate. (2019). What is research design? Retrieved from https://www.researchgate.net/post/What_is_research_design
- Small, M. L. (2011). How to conduct a mixed methods study: Recent trends in a rapidly growing literature. Annual Review of Sociology, 37, 57-86. doi:10.1146/annurev.soc.012809.102657
- Sousa, V. D., Driessnack, M., & Mendes, I. A. C. (2007). An overview of research designs relevant to nursing: Part 1: Quantitative research designs. Revista Latino-Americana de Enfermagem, 15(3), 502-507. Retrieved from http://www.scielo.br/scielo.php?pid=S0104-11692007000300022&script=sci_arttext
- Yin, R. K. (2014). Case study research: Design and methods. Thousand Oaks, CA: Sage publications.