The research design is a framework for planning your research and answering your research questions. Creating a research design means making decisions about:
- The type of data you need
- The location and timescale of the research
- The participants and sources
- The variables and hypotheses (if relevant)
- The methods for collecting and analyzing data
The research design sets the parameters of your project: it determines exactly what will and will not be included. It also defines the criteria by which you will evaluate your results and draw your conclusions. The reliability and validity of your study depends on how you collect, measure, analyze, and interpret your data.
A strong research design is crucial to a successful research proposal, scientific paper, or dissertation.
Step 1: Consider your priorities and practicalities
For most research problems, there is not just one possible research design, but a range of possibilities to choose from. The choices you make depend on your priorities in the research, and often involve some tradeoffs – a research design that is strong in one area might be weaker in another.
As well as scientific considerations, you also need to think practically when designing your research.
- How much time do you have to collect data and write up the research?
- Will you be able to gain access to the data you need (e.g. by travelling to a specific location or contacting specific people)?
- Do you have the necessary research skills (e.g. statistical analysis or interview techniques)?
If you realize it is not practically feasible to do the kind of research needed to answer your research questions, you will have to refine your questions further.
Step 2: Determine the type of data you need
You probably already have an idea of the type of research you need to do based on your problem statement and research questions. There are two main choices that you need to start with.
|Primary data||Secondary data|
|You will directly collect original data (e.g. through surveys, interviews, or experiments) and then analyze it.
This makes your research more original, but it requires more time and effort, and relies on participants being available and accessible.
|You will analyze data that someone else already collected (e.g. in national statistics, official records archives, publications, and previous studies).
This saves time and can expand the scope of your research, but it means you don’t have control over the content or reliability of the data.
|Qualitative data||Quantitative data|
|If your objectives involve describing subjective experiences, interpreting meanings, and understanding concepts, you will need to do qualitative research.
Qualitative research designs tend to be more flexible, allowing you to adjust your approach based on what you find throughout the research process.
|If your objectives involve measuring variables, finding frequencies or correlations, and testing hypotheses, you will need to do quantitative research.
Quantitative research designs tend to be more fixed, with variables and methods determined in advance of data collection.
Note that these pairs are not mutually exclusive choices: you can create a research design that combines primary and secondary data and uses mixed methods (both qualitative and quantitative).
Step 3: Decide how you will collect the data
Once you know what kind of data you need, you need to decide how, where and when you will collect it.
This means you need to determine your research methods – the specific tools, procedures, materials and techniques you will use. You also need to specify what criteria you’ll use to select participants or sources, and how you will recruit or access them.
|Method||What to consider|
Step 4: Decide how you will analyze the data
To answer your research questions, you will have to analyze the data you collected. The final step in designing the research is to consider your data analysis methods.
Quantitative data analysis
To analyze numerical data, you will probably use statistical methods. These generally require applications such as Excel, SPSS or SAS.
Statistical methods can be used to analyze averages, frequencies, patterns, and correlations between variables. When creating your research design, you should clearly define your variables and formulate hypotheses about the relations between them. Then you can choose appropriate statistical methods to test these hypotheses.
Qualitative data analysis
Analyzing words or images is often a more flexible process that involves the researcher’s subjective judgements. You might focus on identifying and categorizing key themes, interpreting patterns and narratives, or understanding social context and meaning.
When creating your research design, you should consider what approach you will take to analyzing the data. The main themes and categories might only emerge after you have collected the data, but you need to decide what you want to achieve in the analysis.
For example, do you simply want to describe participants’ perceptions and experiences, or will you analyze the meaning of their responses in relation to a social context? Will your analysis focus only on what is said or also on how it is said?
Step 5: Write your research proposal
The research design is an important component of your dissertation or thesis proposal. It describes exactly what you plan to do and how you plan to do it, showing your supervisor that your project is both practically feasible and capable of answering your research questions.
Read the guide on how to write a research proposal and make sure you have included all of the steps above in the research design section. Note that, in a proposal, the steps of your research that have yet to be completed should be written in the future tense. The research design or methodology section of your completed paper, on the other hand, describes the research steps in the past tense.