Speed Limits Kill ?

[denny]

Montana state press release: "In 1999, after 4 years of no numerical or posted daytime speed limit on these classifications of highways, outside of urban areas, Montana recorded its lowest fatality rate. For the last 5 months of no daytime limits in Montana [...] reported fatal accident rate declined to a record low. Fixed speed limits were reinstated on Memorial Day weekend 1999. Since then, fatal accidents have begun to rise again."

Sounds like the statistics might have been a bit selectively interpreted to me - why only the last five months, odd number to choose - but still interesting.

Not sure how old this is - I saw it here: http://www.pistonheads.com/speed/default.asp?storyId=11256

Comments

[ciphergoth.livejournal.com]

A statistic this poorly sourced is best ignored.

[zotz]

This page is a report written during the period. It states that at between 1995 and 1996

From 1987 to 1995 the number of accidents remained fairly constant at around two per million vehicle miles. There was an increase in 1996 to 2.64 accidents per million vehicle miles.

Presumably the reason the experiment was abandoned was that it wasn't seen as a success, and from this I would guess that it wasn't seen as a success in safety terms.

Since 1975, the trend of fatalities has been gradually decreasing. When combined with the increase in the number of vehicle miles, it causes a significant decrease in the number of fatalities per million vehicle miles traveled.

From that, I'd guess that you'd expect to see record lows simply from the longterm trend, and there'd be no automatic reason to credit them to the new speed limit policy. Indeed, if it managed to temporarily interrupt that trend, then the new record low level might have been achieved earlier.

As Paul says, it's best ignored. Without more information, there's no reson to credit the changes with the drop in casualty rates, and while it's entirely circumstantial, my suspicion would be that speed lmits were possibly reintroduced because people who did have this extra information decided that they weren't any improvement. Having said that, if you can find more information, I'd be interested to see it.

We could always just ask steer, of course, traffic being his field.

[zotz]

After further reading, it seems that the return to a clearly-defined speed limits was for other legal reasons. If you look at the graph on page 6, though, you'll see that while the period without a numerical speed limit was characterised by a stable fatality rate, it also signalled the end of a falling trend. Was it responsible for the end of that trend? I think we'd need to know a lot more to say.

It would also be interesting to see what has happened since. Somewhere in that report it says that the rate in the next year after the graph finishes was expected to have fallen. It would be good to know whether it actually did.

So, basically, the original statement was based on a highly partial reading of the statistics.