[JangoSteve]

I would have to assume that this absolutely is the case. As far as I can figure, there'd be no way to get percentage decrease in speeding without already having a speed camera installed as a baseline. Without a speed camera, the only way to measure speeding is by number of tickets issued by police at that exact intersection. But police would only ever be able to catch a small percentage of all speeders, whereas speed cameras could potentially catch all speeders. So installing a speed camera would likely show an increase in the number of speeding tickets issued, whether or not the actual number of speeders changed.

In fact, a cursory googling turns up no studies showing whether speed cameras actually decrease the number of speeders. Only inference can be made, as studies DO show that speed cameras decrease the number of injuries and deaths.

[mikeknoop]

Excellent point your research suggests:

There are few if any prior case-studies regarding the effectiveness of speed cameras.

But they are good at generating money.

[jrockway]

What's wrong with generating money? Why not let the people disobeying the law pay for the roads?

[rmc]

The best way to measure the effect of speed cameras on speeding to have some hidden cameras monitoring the section of road, recording speeds. Motorists should not be able to see them. Then collect some baseline data. Then install the speed camera. Collect more data (from the original camera). Compare the two.