Currently, we have so many road safety policies to reduce the number of casualties and deaths. The WHO guideline recommends the use of CEA for policies, measuring the outcomes in terms of death reductions or casualty reductions. The thing is we don't know which specific policy is accounted for the decrease in that number. The question is how to use CEA to analyze the effectiveness of the policy while the resulting casualty reductions etc. came from a mix of interventions.
United Kingdom |
Replied: 13 Jun 2018 at 17:56
I dont think this is a question for CEA as such. What I would suggest doing is looking at the data to see whether you can isolate what has caused particular reductions. For example if you know when a particular intervention was introduced (eg new drink driving laws) you can look at before and after that date to see if there is a change (you can also do statistical tests such as regression discontinuity design tests https://en.wikipedia.org/wiki/Regression_discontinuity_design). Or if there is a policy which is implemented in a particular province or city you can see whether there is a change in accident rates in that locality while other things stay the same. There are more sophisticated things which you can do (eg synthetic controls) but a lot depends on the level of detail and aggregation of your data.