There are approximately 250 million cars on the road and they only used data from 8 million? That’s 3% of cars on the road to extrapolate into all the cars on the road. Seems like a huge flaw especially since we didn’t know how they got that subset. All seems like click bait as most articles related to Tesla are…
Another good way / better way to see what cars are dangerous are insurance rates. Since insurance companies take in way more data than 8M cars when determining rates.
Just honestly asking Im not a statistician. From a lay person looking high level this seems weird. The conclusion also does not match up with insurance prices that I’ve personally seen nor correspond with my experience.
I’m here for discussion not trying to put anyone down. Could someone just explain to me what I’m missing. No need to downvote. So if you take a non random sample of data how can you extrapolate that out so much? Does this data line up with other people’s data? What am I missing?
Since they just use the 8m for the normalisation it’d be interesting to know how sensitive the rankings are if they assumed some bias. Or maybe even just swap around some normalisation factors and see how robust the ranking is.
I guess they do have near complete data on the deaths, and pretty good data on the population of registered vehicles.
There are approximately 250 million cars on the road and they only used data from 8 million? That’s 3% of cars on the road to extrapolate into all the cars on the road. Seems like a huge flaw especially since we didn’t know how they got that subset. All seems like click bait as most articles related to Tesla are…
Another good way / better way to see what cars are dangerous are insurance rates. Since insurance companies take in way more data than 8M cars when determining rates.
That’s more than 3% of the user base. 0.5% is considered sufficient for statistical relevance.
Found the Tesla driver
Just honestly asking Im not a statistician. From a lay person looking high level this seems weird. The conclusion also does not match up with insurance prices that I’ve personally seen nor correspond with my experience.
I’m here for discussion not trying to put anyone down. Could someone just explain to me what I’m missing. No need to downvote. So if you take a non random sample of data how can you extrapolate that out so much? Does this data line up with other people’s data? What am I missing?
A sample of 8 million cars is more than enough to be representative.
… sufficiently random …
Since they just use the 8m for the normalisation it’d be interesting to know how sensitive the rankings are if they assumed some bias. Or maybe even just swap around some normalisation factors and see how robust the ranking is.
I guess they do have near complete data on the deaths, and pretty good data on the population of registered vehicles.