# Royal Statistical Society Statistic of the Year - comparing lawn mower deaths with terror acts

I was reading an article on the RSS website where they discussed some top statistics of 2017. One of them was an apparently viral tweet from Kim Kardashian that compares deaths from lawn mower accidents and falling out of bed with terror related acts. The link for this article is here.
I've seen this type of comparison many times on the news; their main point was that we don't need to be afraid of terrorist attacks because when you look at the data, the probability of being killed by a terrorist is so much lower than being killed by a lawn mower (in this case).
However, I feel like this comparison is not correct. I'm not a statistician, and my knowledge of probability is very shaky, but there must be another way of making this comparison.

I would like to know if there is indeed a more accurate way of comparing these things.

For example, a death from a lawn mower is most likely an accident. There are many additional factors about each event that we don't know about, such as was the person drunk at the time, was the machine modified etc. Also, for so many people who do not have lawns, this is almost a non-risk.

However, a terrorist related event is very random, it's intentional, the aim is to kill as many as possible in a single time period and there is a whole international effort to prevent these events. I'm sure there are not many people out there whose job it is to prevent deaths from lawn mower accidents.
So taking some of these factors into account, it seems that there should be a better way to compare these events.

• If you live in a condo or apartment in the city and like to compete in marathons and other mass races, then lawn mower statistics is irrelevant to you, while terrorist threat is relevant. Also, following that logic we shouldn't worry about Ebola. We're not going to get sick, some folks in remote places will, so who cares? Let them deal with it. Right? Commented Jan 25, 2018 at 3:29
• I'm not sure what is your point in the comment? If Ebola is not active in my region, then I would not worry about. I would still care, but that's not the point of the question. My question is more about how one would compare such events as death from a lawn mower accident with deaths from terrorist attack. If these type of comparisons should not be made, then is there a technical/statistical reason why not? Commented Jan 25, 2018 at 3:48
• many reasons. if you look at the argument made in these comparisons it's usually: death rate from terrorism is low -> the threat is low -> don't worry about it. the flaw in the argument is what economists call endogeneity. I could argue that the death rate is low precisely because we worry a great deal about terrorism and actively fight it. so the threat is actually quite significant, but because we fight it the death toll is low lately in US. However, 3000 dead and 6000 wounded in 9/11 should remind us of what may happen if we stop worrying about the threat Commented Jan 25, 2018 at 4:09
• It is always the statistician and never the statistic that does the talking. (for instance, the counting has been started, very conveniently, after nine eleven) Commented Jan 26, 2018 at 15:50
• A part of your question has answers in psychology and sociology, and is not necessarily only related to risk evaluation in general or to the specific kind of risk you mention here, which may be why you feel the comparison is not correct. A similar question has been asked on the psychology.SE website, you may find it interesting: psychology.stackexchange.com/q/9200 . However your question could be also a great one for a sociology or anthropology Q&A website (but to my knowledge, there are no such websites on stackexchange, so maybe you'd have to look somewhere else). Commented Mar 18 at 7:22