```Ugh.  Those stats should be shared.  I think I’ll do some tweeting.  Someone want to propose the best stats to tweet?

***************************************
Julie Brown, MD, MPH
Associate Professor, University of Washington
Attending Physician, Pediatric Emergency Medicine,
Seattle Children's Hospital, MB.7.520,
Seattle, WA 98105-0371
Office: (206) 987-4016
Fax: (206) 729-3070

Sent: Monday, February 26, 2018 9:35 AM
To: Brown, Julie
Subject: Re: statistical help

The calculation is very simple if you want to do it yourself. Using the Poisson distribution any time period which on average yields 3 events will have a 95% chance of yielding AT LEAST 1.

http://stattrek.com/online-calculator/poisson.aspx<https://urldefense.proofpoint.com/v2/url?u=http-3A__stattrek.com_online-2Dcalculator_poisson.aspx&d=DwMFaQ&c=aBkXpkKi7gN5fe5MqrMaN-VmRugaRb1IDRfSv2xVRy0&r=LdGMXlGHIBSSsmyYogxTzCzExtS2cqT3uVO-vyNVnd8&m=lf14hN8Qv3j6hNjyxMriH30ZMLN6fdlYQHJX5dqScTQ&s=hQ_P711w7mRzQWEk2Si8A-zia3ERuoLp003uQOfZF8A&e=>

Example: there were 70 school shootings with at least 1 death in the past 10 years in the US. Rate = 7/year. On average there are 3 every 5 months. So there is a 95% chance there will be at least one school shooting with a death in this country in the next 5 months. On average there will be 3.

And yes this is morbid. So is the situation. Of course we could all move to Canada where apparently they have solved the problem of mental illness.

MW

Not to be morbid, but ...

1.  Why would you base it on >10 deaths? I would say any with >2 deaths would qualify as “mass” in my book, assuming that with 3 deaths there are also multiple injuries.

2.  20 years seems like a long reference window given how steep the increase has been, why not make the reference time period something like the last three years?

What do you get then?

Julie

Sent from my phone, please forgive brevity and typos.

Assuming your definition of mass school shooting, a poisson distribution, and Wikipedia's list of school shootings (0.3 per per year over the past 20 years  with >= 10 deaths), there is a 95% chance that there will be a mass school shooting within the next 10 years.
Wait, was that a rhetorical question??
MW

I’m not a numbers guy, so I could use a little statistical help:

What is the nearest date, with 95% probability, by which the next mass school shooting (arbitrarily >10) occurs?

Julian Orenstein