Step 1: The phrase risk in exposed over risk in non-exposed points to a ratio measure of association from cohort studies.
Step 2: That ratio is the relative risk, written as $RR = \dfrac{\text{risk in exposed}}{\text{risk in non-exposed}}$, which tells how many times more likely the outcome is with exposure.
Step 3: Attributable risk is a subtraction of the two risks and the odds ratio compares odds, so neither fits a division of risks.
Step 4: Hence the definition given corresponds to relative risk.
\[\boxed{\text{Relative risk}}\]