We had this story from the Times’ Laura Nelson this past week:
Pedestrian deaths in L.A. rise sharply but officials remain committed to traffic safety program.
First off, this quote from Nat Gale is true:
Many of the changes were installed last summer and fall. Typically, cities wait for one to two full years of data before gauging their success, Vision Zero division manager Nat Gale said.
The city also saw a decrease in deaths involving cyclists, drunk driving and hit-and-run crashes, the Los Angeles Police Department said.
So Nat is right, you want to give programs time, but there are some other reasons why we want to have controls in place rather than just looking at rates and the timing of the program.
If people are walking and biking more in general, then backgrounds for collisions could easily go up; walking and biking could go up for any given year for a bunch of reason, other than just year-to-variation in numbers:
A) better weather; less heat, less rain etc;
B) more employment among those who walk and bike;
C) Improvements on high-demand corridors prompt people to walk and bike more (this would be bad, as it would mean that improvements are not zeroing out collisions. But there may be few collisions per hour flow than there were before, which means there is actually an improvement in the slope of incidence if not the raw numbers);
D) The passage of Vision Zero signaled a greater commitment to bike safety, and it prompted more people to engage in the activity;
E) More collisions are reported, even potentially less serious ones, because people are expecting different/better driver behavior or simply due to a random blip in reporting, or because traffic division officers have been urged or incentivized to write up more;
F) etc.
We should probably be thinking about a case-control method here–maybe Marlon could figure one out–but I don’t know how you control for a case like LA City–you have, I suppose, boundaries between LA City and LA County but I am not sure the policing effort is otherwise the same.