A lot of these neat or fun statistics are not incorrect but can be misleading or commonly misinterpreted. For example, one such shark based factoid compares the number of people bitten by new Yorkers to the number of people bitten by sharks. We need to keep several things in mind with such statistics. The sheer millions of people who will pass through New York and the number of individual people they will pass by compared to the number of people in the ocean each year and the number of sharks they will pass by for example. If you do a “tour” of New York City you might pass by tens of thousands of people, in the ocean it is common for any given person to encounter 0 sharks, especially in oceans where there are no sharks. So your chances of being bitten by a shark are generally lower because it is generally less common to encounter a shark, but if you do encounter a shark the number of people who are bitten compared to the number of man to shark encounters is much higher.
In short, any random trash can is much less likely to hurt you and any random shark is much more likely to hurt you, but you will generally have thousands or hundreds of thousands or infinitely greater interactions with trash cans than sharks for the average person. Of course this brings us to the second statistical issue. This is general, it doesn’t take into account circumstance. Shark researchers and marine biologists, certain divers etc. will regularly encounter sharks close up, often in controlled environments and with safety measures. So while someone working with sharks might generously have equal interactions with trash cans and sharks, they are unlikely to be killed by a shark in general.
So the number of chances you have for an accident to happen is only one factor but is circumstantial. Radiation poisoning is a rare cause of death but few people work closely with strong ionizing radiation in environments that make poisoning likely. Of the number of historical deaths by…
.. or relating to radiation exposure, most are of people who did not work closely with radiation such as those in proximity to nuclear detonations and aftermath, reactor release incidents, and of course exposure through medical means or “lost source” incidents and other secondary exposure.
So we do have to look at risk factors.
Sanitation workers might be a group at high risk for garbage can related fatality. They work in a fast paced environment with trash cans and have a high exposure to them. What more though is what that work looks like. It often utilizes many potentially dangerous machines, cans are often lifted high above the ground, placed or moved with high mechanical force. Calling cans, being crushed or concussed by a can that is manipulated by mechanical means, etc etc. many of these statistics also don’t factor or differentiate between poor/improper and proper/advised practices.
A statistic comparing vending machine deaths to other types of death often doesn’t factor that most vending machine deaths occur when people attempt to steal or abuse the machines in unsafe manners and the machines fall on them. If you remove those instances the statistics for vending machine deaths changes dramatically.
When we look at automotive accident fatalities, when we remove incidents where people were not following traffic rules (running lights, incomplete stops, excessive speed, driving the wrong direction, failure to maintain safe distance etc..) or other unsafe or prohibited practices such as driving tired, driving under the influence of drugs or alcohol, distracted driving etc. your odds of traffic fatality plummet- but are still high because of the other stat issue- in this example removing people who cause accidents drops the danger but we didn’t remove the fatalities they cause. In other words if a drunk driver kills themselves and a car with 4 people, if we remove the drunk driver from the statistics we still have 4 victims. The victims were following the rules but were killed by someone who wasn’t. So if we remove the victims of poor operators as well, your odds of traffic fatality become almost nil.
So it depends on what we want to know. If you want to know how likely you are to die while driving, and you follow the rules and good practices of the road, you would leave in victims of poor drivers but remove poor drivers themselves. If you wanted to know how safe the automobile was you would remove all Poor operators and their victims as the theoretical inherent safety of the automobile doesn’t include those who misuse the device or suffer from poor users as if we removed all Poor operators none of those incidents would occur. We can however use statistics to the inverse- since we can choose our data based on the question we want answered, we can also choose out data to produce the answer we want for a question.
That may confuse some so let me be clear:
If you have a question you need answered like what are your specific odds of being bitten by a shark in the specific beach you are vacationing at, you don’t want to consider a broad average that includes people swimming in areas that have no sharks. Those people would never be bitten because even though they are in the ocean they won’t encounter a shark to bite them at all. You also likely don’t want to include people working 13 hour shifts in shark infested waters unless you will be in the water 13 hours a day every day and the beach is shark infested. So your specific question needs specific data to give you the most accurate and useful answer.
However let’s say you owned the beach resort and wanted to make people feel safe, then you might include places where there are no sharks in your statistic so you can say that the odds of being bitten are lower than they are and it won’t be a lie exactly- not one you’re likely liable for in court.
So the number of chances you have for an accident to happen is only one factor but is circumstantial. Radiation poisoning is a rare cause of death but few people work closely with strong ionizing radiation in environments that make poisoning likely. Of the number of historical deaths by…
So we do have to look at risk factors.
A statistic comparing vending machine deaths to other types of death often doesn’t factor that most vending machine deaths occur when people attempt to steal or abuse the machines in unsafe manners and the machines fall on them. If you remove those instances the statistics for vending machine deaths changes dramatically.
If you have a question you need answered like what are your specific odds of being bitten by a shark in the specific beach you are vacationing at, you don’t want to consider a broad average that includes people swimming in areas that have no sharks. Those people would never be bitten because even though they are in the ocean they won’t encounter a shark to bite them at all. You also likely don’t want to include people working 13 hour shifts in shark infested waters unless you will be in the water 13 hours a day every day and the beach is shark infested. So your specific question needs specific data to give you the most accurate and useful answer.
However let’s say you owned the beach resort and wanted to make people feel safe, then you might include places where there are no sharks in your statistic so you can say that the odds of being bitten are lower than they are and it won’t be a lie exactly- not one you’re likely liable for in court.