|
|
| Re: Read these three articles about cycling (324984) | |||
 |
|||
| Home > BusChat | |||
|
[ Read Responses | Post a New Response | Return to the Index ] |
|
||
|
Re: Read these three articles about cycling |
|
|
Posted by JerBear on Wed Aug 16 07:58:55 2017, in response to Re: Read these three articles about cycling, posted by Stephen Bauman on Mon Aug 14 20:51:48 2017. I never took statistics, so bear with me if I get this wrong or extrapolate too much. But the percentage of deaths as a greater part of deaths and serious injuries is a pretty steady number. Ignoring the 2011-to-201-whatever numbers, the number of fatalities is between 4.1% and 5.2% of people killed or seriously injured. The fatalities number in the report for 2011-2015 does not have an asterisk next to it, so it is the true average of number of people killed each year during those time periods. To estimate the highest likely number of KSI between 2011-2015, take the lowest percentage of deaths per KSI in the recent past and apply it to the known deaths between 2011-2015. You get 418 as the average. For 418 to be the average for the 5 years between 2011-2015, knowing already that there were 1568 KSIs between 2011-2014, it would mean that there were 521 KSI in 2015. That would be both a huge number in and of itself, and DOT would not want that number to mix in with the data if they argue that bicycling is becoming safer per ride.Of course if you instead apply the highest percentage of deaths per KSI in the recent past and apply it, the 5-year average becomes 330 KSI, and only 84 KSIs would be the number for 2015. Now if we apply a bell curve to all of the numbers between those extremes, what is the likelihood that the 5-year average was 490? Again, I didn't even know what statistics was in college or that there was a class in it, so I could be wrong. But I think the correct answer is something like "less than a 1% chance." |
|
|
|