Two economists are blaming Pokémon Go for causing traffic accidents and likely fatalities:
For a brief, shining period last summer, Pokémon Go reigned supreme. It brought obsession, joy, and, according to a new paper, injuries and death.
This working paper, appropriately and evocatively titled "Death by Pokemon Go," shows the darker side of the massively popular augmented reality game. Purdue University economists Mara Faccio and John McConnell combed through accident reports from Tippecanoe County, Indiana, in the first 148 days after the game was released in July 2016. In that county alone, the total value from injuries, damage, and the two lives lost is between $5.2 million and $25.5 million. If you scale this to cover the entire US, it would suggest that $2 billion to $7.3 billion were lost just in those few months.
The reports showed during those 148 days, 286 additional crashes occurred in the county, compared to the same period before. Of these, 134 were near pokéstops. In this scenario, it's crucial to determine that Pokémon Go caused these damages directly, as opposed to just causing people to be outside more, thus more likely to be hit by cars.
Also at PC Magazine.
Related: Peak Pokémon Go?
No Pokémon Go or Other AR Games in China
Russian Prosecutors Seek 3.5 Years for Blogger Who Played "Pokémon Go" in Church
Trial Will Decide Whether Milwaukee Can Require Permits for Using Locations in Augmented Reality
(Score: 3, Insightful) by beckett on Monday November 27 2017, @05:27PM
In order for the extrapolation to hold true, the rest of the counties, suburbs, cities must meet the same geographical structure as Tippecanoe, IN, all police across the expanded area must fill out their forms the same way and report at the same rate, the reports must be consistent rather than reports of convenience, and with regards to fatailities the expanded area must have comparable emergency response and emergency healthcare resources that Tippecanoe county had. yet i see none of these variables controlled in their methodology.
What i see is findings that only apply to Tippecanoe County from March, 2015 to November, 2016. While it's mentioned that controlling for "pokemon caused these damages directly", but i do not see in the study how they were able to achieve this, let alone extrapolate this assumption to cover the entire US.