3 ms·
1% isn't that small a sample here, this is rather significant statistically.
by fooker 7y ago
1% isn't that small a sample here, this is rather significant statistically.
- uoaei 7y agoSurveys always have a self-selection bias. It's entirely possible people were worried about retaliation if they said anything different.
- seattle_spring 7y agoExcept Blind self-selects only the most disgruntled of employees.
- fooker 7y agoWouldn't employees against this be equally disgruntled?
- wutbrodo 7y agoNot even close. 1% of the company, selected not randomly, not only by who chooses to use Blind, but by employees on Blind who choose to respond to the survey about this issue.
- msbarnett 7y agoThe 1% is a self-selecting sample (people who signed up for Blind, and voluntarily chose to respond to the survey after seeing it), not a random sample. That leads to all kinds of bias issues that make it a pretty poor sample to extrapolate from.