Re:Blizzard employees strongly support punishing HK advocate...
by diziet on Tuesday October 15, 2019 @08:39PM
(#21265817)
I think your math is off - with a 99% confidence interval, on a population of 4,700 employees of Blizzard a sampling of 60 people with a 64.27% approval rate gives us a confidence interval of +-16%~. At 95% confidence, the interval shrinks to 12%~
[ Parent
| on HN ]
Re:Blizzard employees strongly support punishing HK advocate...
by chongli on Tuesday October 15, 2019 @08:59PM
(#21265947)
You can’t construct that confidence interval unless you know the sample was selected randomly among all Blizzard employees. If the sample is self selected then you don’t have anything at all, really the data is useless.
In general, there is a lot of pressure for employees to conform to company decisions, regardless of their actual opinion, because their job is on the line. It does not matter that this “Blind” service purports to be anonymous. People will still demonstrate a bias because they have no guarantee that their communications are not being monitored and ultimately they have nothing to gain by giving their honest opinion.
Re:Blizzard employees strongly support punishing HK advocate...
by JumpCrisscross on Tuesday October 15, 2019 @08:54PM
(#21265924)
> with a 99% confidence interval
Confidence intervals assume a random sample. This wasn’t a random sample.