There are many ways to design your study, but some will answer your research question better than others. By choosing the right research design, you can minimize your margin of error and get more accurate and useful results. It all starts with the aim of your study, which will help you determine the best approach to take when it comes to your research design.
Research design is the methods and procedures of a study, which vary depending on the type of study, research question, variables, and hypothesis.
While there isn’t a clear consensus on how to categorize the many different types of research design, most survey research can be categorized as either exploratory, descriptive, correlational, or experimental.
Exploratory research aims to uncover new ideas and insights from participants who have some familiarity with your research subject. This type of research study design can yield powerful insights but has limited applications. As the name implies, exploratory research focuses on exploration and belongs at the beginning of your research project. The insights you get will help define the direction for the rest of your research, rather than provide conclusive answers.
Exploratory research is all about qualitative, not quantitative data. An exploratory research survey includes open-ended questions, where respondents can share impressions and ideas in an open format. While these responses can’t be quantified (except maybe with a word cloud), there are some helpful methods for collecting and analyzing open-ended data that you can apply to exploratory research.
Descriptive research sheds light on the current characteristics of a research subject by collecting, analyzing, and presenting feedback from those familiar with the subject. This type of transparent research design asks participants to give their thoughts and opinions on the research subject, so that the researcher can describe the state of the subject with more detail and accuracy. The focus of descriptive research is always on “what” rather than “why.”
This type of research study design leans on both qualitative and quantitative data. For example, a descriptive research survey might collect qualitative data with open-ended questions, while also collecting quantitative metrics with multiple choice, rating scale, ranking, or demographic questions. Both types of data will help you paint a clearer picture of your research subject.
Correlational research looks at whether or not variables in the study are correlated with each other. It’s a non-experimental type of research study design that uses factor analysis to determine whether, for example, the way respondents answer one question in your survey is related to the way they answer another question. Correlational research can help you develop models that predict things like medical conditions and consumer behavior.
Many observational studies use correlational research designs, particularly if the goal is to construct a predictive model. These studies use quantitative data derived from multiple choice, rating scale, ranking, or demographic questions to calculate the correlation coefficients between two variables.
Experimental research (or causal research) aims to establish a causal relationship between two variables by changing an independent variable to see what effect it has on a dependent variable. Experimental research design is ideal for very specific and practical research questions. An experimental research survey might, for example, examine how price affects purchase intention by testing people’s willingness to buy a new product at different price points.
Controlled experiments, field experiments, and natural experiments all utilize experimental research design. To be valid, these experiments must adhere to strict research methods and procedures that ensure the integrity of the experiment.
Exploratory, descriptive, correlational, and experimental studies each have their own research methods and procedures—and the approach that’s right for you will depend on what you’re seeking to learn. Think about the purpose of your study, and follow best practices for every type of survey design.