Forum Topic

Don't believe LTNs halve road injuries in London

Don't believe this, someone is massaging some data.

Rae Lewis ● 1918d7 Comments

Oh, OK then. Don't you think you ought to provide some statistics to back this up?

Carl Wynne ● 1918d

I haven't done any intensive research.  But in broad terms it's obvious that LTNs just displace traffic onto other roads.

Rae Lewis ● 1918d

RL: "I haven't done any intensive research."Was your use of "intensive" really necessary?

Richard Cathcart ● 1918d

I don't think any research is needed to disprove the notion that LTNs have halved road injuries in London because road injuries haven't halved but that isn't what the research is claiming. It is only badly worded headlines that misinterpret the data that do that.

Mark Evans ● 1918d

Direct from the paper: "Inside LTNs, total injury numbers approximately halved after LTN implementation."Of course, that's not claiming that LTN's halved road injuries across London. They did drop by 16% elsewhere, but that's much more likely to be because of fewer cars than anything to do with LTN's. The drop of one half inside LTNs is in addition to that fall outside the LTNs.There was no significant difference between boundary roads and outside LTN's. If LTNs produced any traffic displacement to their boundary roads, it's not evident in a significant increase in injuries.

Tom Pike ● 1918d

The data doesn't show that.What it shows is that injuries that are more than 25m inside the LTN boundaries halved and that change is almost certainly as result of the LTN if but only if you start from the assumption that there only 2 explanations for the change - either the change was chance or it was because of the LTN.Because we do not know how many injuries happened within the LTN but less than the arbitrary 25m from the boundary line that the researchers used we cannot say with any certainty whether injuries fell or not within the LTN.  For me the biggest surprise was that there was no change in cyclist injuries within the LTN.  That doesn't make sense to me, but I suspect it may be a quirk of limited data.It would be interesting to know why the researchers chose to define "inside the LTN" as being at least 25m inside the LTN - it could be a limitation of the data they were working with.I would also highlight, and I commend the researchers for specifically pointing this out, that the research only deals with absolute numbers of injuries and does not take into account whether the number of journeys changed, so it cannot give any insight into whether the relative risk of any particularly journey increased or decreased.

Justin Stephenson ● 1917d

Those boundaries certainly had a large number of casualties, before and after the introduction of the LTNs. They're close to 10% the total injuries for the rest of London, so most certainly not representative of an LTN! It's possible to explore what happens if some of those boundary numbers are added to the LTN numbers. If you add in 10%, the numbers of injuries about doubles, 147 rather than 81, but overall they're still down 25%, with a 56% reduction for pedestrians, 12% for cyclist and 29% for cars. So the answer will depend somewhat on whether injuries are considered to be due to traffic representative of the LTN, not the very busy boundary roads, but adding in a proportion of the boundary injuries still gives a significant reduction even if the original injury count is doubled. You of course correct that the risk does depend on the actual change in the number of cyclists and pedestrians, but other data certainly indicates that cycling numbers increased, so the reduction in cyclist injuries is likely an underestimate of the reduction in risk, as the authors themselves suggest.

Tom Pike ● 1917d