Choosing Between Exploratory, Descriptive and Causal Designs
Your research design is the overall plan for answering your research question. Three broad types come up again and again in methods textbooks: exploratory, descriptive and causal. In this blog, we'll look at what each one does, when to use it, and how to decide which fits your study.
Exploratory Research:
- Used when a problem is new or poorly understood, and you need to find out what the key issues are.
- Common methods include literature reviews, in-depth interviews, focus groups and case studies.
- It is flexible and generates ideas and hypotheses, but its findings are not usually meant to be generalised.
- Example question: "What challenges do small retailers face when moving their business online?"
Descriptive Research:
- Used when you want to describe characteristics, behaviours or patterns accurately, often in numbers.
- Common methods include surveys, observation and analysis of existing records.
- It answers questions about who, what, when, where and how much, but it does not show cause and effect.
- Example question: "What proportion of small retailers in the city sell online, and which platforms do they use?"
Causal Research:
- Used when you want to test whether one variable causes a change in another.
- The usual method is an experiment or quasi-experiment, where you manipulate one variable and control others.
- It gives the strongest evidence of cause and effect, but it needs careful design and is not always practical or ethical.
- Example question: "Does a one-day digital skills workshop increase the number of retailers who start selling online?"
How to Decide:
- Ask how much is already known. Little knowledge suggests an exploratory design; established concepts allow descriptive or causal work.
- Ask what kind of answer you need: ideas and themes, an accurate picture, or evidence of cause and effect.
- Consider whether you can realistically control variables. If not, a causal claim will be hard to support.
Combining Designs:
- Many studies combine designs, for example exploratory interviews to identify key factors, followed by a descriptive survey.
- If you combine them, explain the sequence and how each stage feeds into the next.
The right design follows from your research question and from how much is already known about your topic. Be clear about which design you are using, avoid making causal claims from descriptive data, and explain your choice in your methodology chapter.