A correlational design checks whether two things move together — when one goes up, does the other go up, go down, or stay unrelated? You measure both variables as they naturally are, without changing anything, and calculate how strongly they are connected. It answers "is there a relationship?" — not "does one cause the other?"
When to Use It
- Your research question asks about a relationship between two variables.
- You can measure both variables but cannot (or should not) manipulate them.
- You want to know the strength and direction of a link.
- Words like "relationship between" or "association" appear in your topic.
When Not to Use It
- You want to prove that one variable causes the other — use an experiment.
- You only have one variable to study.
- You simply want to describe a single situation — use a survey.
Nigerian Project Example
"The relationship between study time and academic performance among final-year students of [Department]." You measure how many hours students study and their CGPA, then use a correlation test to see how strongly the two are linked — without claiming that more study hours directly cause higher grades.
Undergraduate vs Postgraduate
Undergraduates typically test the relationship between two variables using a simple correlation (such as Pearson's r). Postgraduates often handle several variables at once, use regression to model how variables predict an outcome, and discuss possible confounding factors more carefully.
Common Mistakes
- Treating a strong correlation as proof that one variable causes the other.
- Ignoring an obvious third factor that could explain both variables.
- Reporting the correlation value with no interpretation of its strength or direction.
- Using a correlation when the question really called for an experiment.
Explain your variables and analysis in your methodology — see how to write Chapter Three and the full research methodology guide. Project Lab can help you state relationships precisely without overclaiming cause.
Frequently Asked Questions
Which test do I use for a correlational study?
Pearson's correlation (r) is common when both variables are numeric. Spearman's is used for ranked data. Your analysis should report the value and whether it is statistically significant.
Why can I not claim causation?
Because you did not control the conditions or rule out other explanations. A hidden third factor could be driving both variables. Only a controlled experiment supports a fair claim of cause and effect.