What is needed to establish causation rather than correlation in an evaluation of a road safety program?

• A. Longer observation period
• B. Randomized controlled design or a controlled before-after study
• C. More participants
• D. Qualitative interviews

Answer: (B)

To establish causation in evaluating a road safety program, you need a study design that can isolate the program’s effect from other factors by comparing outcomes with a similar group that did not receive the program. A randomized controlled design or a controlled before-after study does this by creating a treatment and a comparison group and measuring outcomes before and after. Randomization helps balance both known and unknown confounding factors, so any difference in safety outcomes can be attributed more confidently to the program itself rather than to other differences between groups. A controlled before-after design uses a similar non-treated group and compares how outcomes change over time in both groups, helping to account for broader trends or external events that could affect all sites. This setup is far more capable of establishing a causal link than simply observing results over time without a proper comparison.

Longer observation alone can improve detection of effects but doesn’t by itself prove causality because other influences may be at play. More participants increase statistical power but don’t address internal validity or confounding. Qualitative interviews add context and understanding of implementation, but they don’t provide the rigorous evidence needed to claim a causal effect.