Quote:
Originally Posted by Danzig
you think? based on what exactly?
Those internal figures show at least 1,800 police killings in those 105 departments between 2007 and 2012, about 45% more than the FBI’s tally for justifiable homicides in those departments’ jurisdictions, which was 1,242, according to the Journal’s analysis.
The full national scope of the underreporting can’t be quantified. In the period analyzed by the Journal, 753 police entities reported about 2,400 killings by police. The large majority of the nation’s roughly 18,000 law-enforcement agencies didn’t report any.
http://www.wsj.com/articles/hundreds...ics-1417577504
and then the article you posted has this in it:
His results, posted last week on his blog Cop in the Hood, arrived with several caveats, notably that 25 percent of the website’s data, which is drawn largely from news reports, failed to show the race of the person killed.
you're wanting this bs study with god knows what numbers from a site, where one quarter of the stats included don't even know the race of the person shot, to prove something??
if you wish to believe what the person in the article you cited tries to conclude, go for it.
but good luck getting anyone to read it and glean anything useful from it. |
If they didn't know the race of the person shot in 25% of the cases, that means they did know the race of the person in 75% of the cases. As I said before, when you look at data on any subject, you never need to have 100% of the data to be able to draw accurate conclusions. If there are 2,400 police shootings, if they only have data on half of the shootings, the results are still going to be the same (give or take 1%). Even if they only had data on 240 of the shootings (10%), the numbers are going to be the same. You don't need anything close to all the data. I'm not sure what is so confusing to you about that.
If I do a poll of 1,000 random people on any subject, and 80% of the respondents answer a certain way, is the poll not reliable since there are 300 million people in the country and I only polled 1000 people? Good luck if you think polls and data aren't accurate unless they cover the whole population. If a new medication is tested on 1,000 people and it works on 90% of them, would you say we need to test it on 300 million people to know whether it works? Of course not. If you have a good size sample of something, that is all you need. I don't know why that is so hard for you to understand. Although I don't think you would have a problem understanding it if you liked the conclusion. But since you don't like the conclusion, you say they need more data.