Thats not completely true. In statistics a bias is deffined as: A statistical sampling or testing error caused by systematically favoring some outcomes over others.arnisador said:Statistics aren't biased. (Bias actually has a technical meaning here, and it's very rare to use a statistic with nontrivial bias--the s.d. is the only commonly used biased statistic.) Most of what people are throwing around are descriptive statistics, but in any event, statistics are numbers, and inferential statistics gives conclusions (at a certain confidence level, with certain simplifying assumptions). It's what people make of those that is where the bias truly enters.
Nontrivial statistical biases are quite rampant. Inferential statistics do give a conclusions or prediction such as:
That was my point, statistics as a whole and especially inferential statistics are slave to their creators bias. Your talking about proven statistics. To say inferential statistics as a whole give conclusions at a certain confidence level and with a certain simplifying assumption is pretty naive if you ask me.upnorthkyosa said:2. Nationwide African Americans have 3x greater chance of being poor.
Yes, your correct that peoples interpretation to statistics offer a large margin for bias but even larger so may be the bias of the creator. Either way, there is quite a large margin for error in blindly stating statistics as proof of anything, especially something as furtive as "soft racism".
7sm