General > General Technical Chat

Amazon accuses customer of racism & shuts down their smart home!

<< < (47/56) > >>

Zero999:

--- Quote from: james_s on June 30, 2023, 10:16:56 pm ---
--- Quote from: gnuarm on June 30, 2023, 10:10:50 pm ---Where?  In the US it is not "markedly higher".  What exactly are you referring to?  Can you give a reference?

--- End quote ---

I have not personally confirmed this to be accurate but here it is:
https://en.wikipedia.org/wiki/List_of_countries_by_traffic-related_death_rate


Road deaths per 1 billion km driven, USA 8.3, UK 3.8, that is certainly what I would call "markedly higher", it's more than double. Just scanning down the list the USA is higher than the majority of other developed nations.

--- End quote ---
As you've said, I suspect it's due to more younger drivers and perhaps the road layouts, which are responsible for the higher number of road deaths per mile in the US, compared to the UK.  I don't know about the US, but here it's quite difficult to pass the driving test, which consists of a theory, as well as a practical test, there's a blanket speed limit of 30 mph in residential areas, dropping to 20 mph in high risk areas and separate lanes, even roads for cyclists.


--- Quote from: TimFox on July 01, 2023, 04:32:28 pm ---I doubt that measured "IQ" is a truly valid quantitative measurement of anything beyond the ability to score on IQ tests.
However, in current psychometry the scores are normalized to a normal (Gaussian) distribution with mean = 100 and standard deviation = 15.
In that distribution, (the median) = (the mean) = (the mode).

--- End quote ---
IQ correlates very well with success in many areas of life. Interestingly sex makes a difference: the male population have a greater variance in IQ, which is less n females i.e. there are more male, than female, geniuses, but there are more males with learning disabilities. This explains why there are disproportionately more boys in gifted student programmes, as well as special schools.

coppice:

--- Quote from: TimFox on July 01, 2023, 04:32:28 pm ---I doubt that measured "IQ" is a truly valid quantitative measurement of anything beyond the ability to score on IQ tests.
However, in current psychometry the scores are normalized to a normal (Gaussian) distribution with mean = 100 and standard deviation = 15.
In that distribution, (the median) = (the mean) = (the mode).

--- End quote ---
In all of the study of human capability and behaviour, IQ scores have the best correlation with lifetime outcomes. The second best correlation is things which are measuring industriousness in some form. Someone who scores high in both IQ and industriousness generally does very well. IQ doesn't measure what most of us would consider "smartness". People seem to use that word as some hybrid of IQ and wisdom, which are rather different things. An industrious, high IQ person who lacks the wisdom to pursue meaningful goals can have a very bad outcome. A high IQ often allows people to rationalise stupid things a lesser IQ person would just consider dumb.

IQ values are on a very compressed scale. Is a person with an IQ of 140 only 40% more capable in problem solving? No, they are far more capable than that. Similarly, is someone with an IQ of 80 only 20% less capable than the average? No, they are barely able to take care of themselves. Perhaps the scaling was chosen specifically to make people feel better about their differences.

gnuarm:
Why are people saying US fatal accident rates have increased lately?  I am not able to find any data to support this.  Of course, the total death count is much more readily available. 

Does anyone have a source to support this claim?  Or did I misunderstand what was being said?

Monkeh:

--- Quote from: gnuarm on July 01, 2023, 05:44:38 pm ---Why are people saying US fatal accident rates have increased lately?  I am not able to find any data to support this.

--- End quote ---

Because they have. You can't have looked very hard.

The quick version:
https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813124
https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813321
https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813454

It's curious to me how they limited the context in the most recent publication..

Oh, and let's put the preliminary 2022 numbers in play:
https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813428

gnuarm:
Interesting.  If you look at the most recent data, you see that not only did the numbers increase in 2020, but the step function occurred in 2nd quarter, 2020!  I can't think of anything that happened in 1st or 2nd quarter of 2020 to cause such a step function. 

This would indicate a change in the way the data is collected and/or processed, not an actual change in the thing being measured.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod