Two equal-sized cohort trays with different event-token counts. Original AI-generated artwork produced for GreenCalc; not a measured diagram or empirical data display.
Compute a risk within each group first
Suppose group 1 has 20 events and 80 non-events, while the reference group has 10 events and 90 non-events. Risks are 20/100 = 20% and 10/100 = 10%. Relative risk is 0.2/0.1 = 2. The risk difference is 10 percentage points, a separate measure of absolute contrast.
Reversing the reference reverses the ratio
If the two groups exchange roles, the relative risk becomes 0.1/0.2 = 0.5. Neither arithmetic result is more correct without the intended direction. Label the numerator and reference group rather than reporting a bare “risk ratio of two.”
Use comparable follow-up and valid denominators
Risks refer to the proportion experiencing an event over a defined observation period. Comparing a one-month risk in one group with a one-year risk in another can mislead. Counts drawn from a case-control sampling design do not automatically estimate group incidence risks; an odds ratio serves a different purpose in that setting.
Keep association separate from a causal explanation
The Relative Risk Calculator reports group risks, RR, risk difference and a supported approximate log interval. It does not remove confounding or decide that exposure caused an outcome. Read zero-cell interval limits before interpreting a boundary estimate. The Risk Calculator elsewhere in this category is a probability-times-loss model, not another name for this two-group comparison.
Check the model behind your result
Use the Relative Risk Calculator for the supported calculation and review its stated assumptions before interpreting the output. Find related tools in the Statistics category, or browse the GreenCalc homepage for other calculation topics.
By GreenCalc Editorial Team · Updated 2026-10-26