Scientific concept · BEGINNER LESSON
Causation
Causation is the relation between two events or variables when one event is a necessary part of the chain that produces the other. In these sources, epidemiology uses it to mean that the disease increase seen in exposed people would not have happened without the exposure, while many sciences also describe causation as one variable changing another through a mechanism.
What causation means
Causation is about a cause and an effect. In epidemiology, it means that an observed rise in disease among exposed people would not have happened if they had not been exposed. More broadly, scientists may describe causation as a change in one variable producing a change in another variable. This is different from simple correlation, which only means two things vary together. [0,1]
[1] Reference Guide on Epidemiology - Reference Manual on ... [2] Best practices for moving from correlation to causation in ecological research
How scientists look for it
A key requirement is time order: the exposure must happen before the outcome. Epidemiology also treats causation as an inference from circumstantial evidence, because cause cannot be directly seen with the five senses. Researchers therefore look for supporting features such as repeated findings, biological plausibility, and dose-response patterns, but these are supportive rather than always required. [0]
[1] Reference Guide on Epidemiology - Reference Manual on ...
Mechanism and common causes
One source describes a causal relationship as a perturbation in a cause X producing a change in an effect Y, sometimes through intermediate variables. Another source says statistical dependence between two variables can come from direct causation or from a third variable that causes both; that third variable is called a confounder, and it can make a non-causal link look causal. [1,2]
[2] Best practices for moving from correlation to causation in ecological research [3] 20th Century Theories of Scientific Explanation (Stanford ...
What causation does not automatically mean
Correlation alone does not prove causation, because the same pattern may arise from reverse causation or from a common cause. Epidemiologic methods also cannot deductively prove causation; they can support a strong inference instead. Another limit is that epidemiology addresses group-level disease incidence, not whether one agent caused one particular person’s disease. [0,1]
[1] Reference Guide on Epidemiology - Reference Manual on ... [2] Best practices for moving from correlation to causation in ecological research
Why this matters in explanation
In scientific explanation, causal information is often important, but statistical relevance by itself is not enough to capture all causal structure. Some theories of explanation require the cited cause to be part of the mechanism that makes the result happen, not just statistically associated with it. [2]
[3] 20th Century Theories of Scientific Explanation (Stanford ...
SEE THE IDEA
From exposure to causal claim
The factor thought to make a difference.
The exposure must come before the outcome.
The change being explained or measured.
Other variables may create the same pattern.
Researchers judge whether the evidence supports a cause-effect link.
Exposure → timing before outcome → compare with alternative causes → infer or reject causation
Causation is inferred by checking time order, mechanism, and rival explanations, not by correlation alone.
TRY IT YOURSELF
Test your understanding
Choose an answer, then check the explanation.
PUT IT TOGETHER
Match each concept to its meaning
Drag a label onto its description, or select a label and then select a description. This also works with a keyboard.
4 pairs to match
Sources and further reading
This general lesson was generated from public source material. It explains a concept; it does not interpret a private paper or provide medical advice.