LIES, DAMNED LIES, AND DATA

There are currently a lot of ads on TV for video doorbell apps.

Cameras attached to your doorbell that have a two-way link to your smart phone.

So that, when you’re not at home, you can talk to whoever’s at the door, whether it’s a deliveryman or an intruder.

The ads stress how much more secure you’ll feel, being able to scare off burglars.

Which was pretty much the whole reason Amazon paid $1 billion for Ring, the American company providing doorbell security apps.

Ring launched their doorbell app in 2013 with the mission statement: “Reducing crime around the world”.

Ring reported that in a Los Angeles Police Dept pilot study, the devices had reduced burglaries by 55%.

Even though they were installed in just 10% of homes, burglaries were also down in neighbouring homes, showing a ‘herd immunity’ effect.

Ring CEO, Jamie Siminoff, said: “It taught us that Ring actually protects the homes around it, as much as it protects the home”.

These results caused Ring to offer a deal to neighbouring cities.

They’d give each home $100 discount if the city covered half the cost.

Several cities jumped at it, by 2018 Ring had contracts of $420,000 on their neighbourhood -rebate scheme.

But it wasn’t quite as simple as it seemed.

The 2015 study featured just 40 devices, but Ring weren’t willing to release detailed data of the locations.

Maria Cueller, assistant professor of criminology at the University of Pennsylvania, said: “I don’t see the decrease in crime.

The sample size is too small, it’s not enough to say whether the effect is something you see in data or just some random variation.”

2016 showed a 21% decline in burglaries, but again Ring wouldn’t release detailed data.

Wilshire Park Association said the area of the trial covered 3 districts which had actually seen a combined rise in burglaries, but without detailed locations how could you tell?

In 2017, Wilshire Park had more burglaries than in any of the previous 7 years.

Country Club Park said Ring had conducted a study of 200 devices there, but for some reason Ring never reported the results.

Public data shows that in 2015, 2016, and 2017, burglaries increased in Country Club Park.

John Rock, management analyst for West Valley City decided to conduct a test.

If Ring were telling the truth about their device creating ‘herd immunity’ effect it could save them a lot of money.

In August 2017 they fitted doorbell-camera apps to 10% of 764 homes.

Then they kept a control group of 754 identical homes without doorbell-camera apps.

By August 2018 the results were in.

The 10% of homes with apps had seen a 41% drop in burglaries, but the 754 homes withoutany apps had seen a 50% drop in burglaries.

The 10% of homes with apps had seen a 25% drop in property crime, but the 754 homes without any apps had seen a 32% drop in property crime.

In 2018, Ring donated 500 doorbell-camera apps to two neighbourhoods in Newark.

Figures show that burglaries were down 50% in those two neighbourhoods.

But figures also show that, at the same time, burglaries were down in other parts of Newark, without cameras, by 65%.

There is a belief that data is truth, that data equals facts.

But as Josef Stalin said: “Those who vote decide nothing. Those who count the votes decide everything.”