Correlation DESIGN vs. Experimental Design: Coffee and Cigarettes (7-5)

In an experimental design, the researcher manipulates the X variable and measures what happens to the Y variable. This allows the demonstration of causality. In a correlational design, variables are observed as they occur naturally; no attempt is made to manipulate either one. Dr. Daniel uses an example of consequences of caffeine and smoking to compare and contrast the competing research designs.

[arisoda]

so we never proved smoking causes cancer, we only made the case more _likely_ to be true based on multiple correlational design studies?