Is there something about statistics that lends itself to this sort of saying, or is it just that people will say anything to support their case, and this includes citing irrelevant or incomplete statistics?

  • $\begingroup$ I feel that your question is not precise enough to get a reasonable answer. $\endgroup$ – user28 Jul 26 '10 at 21:44
  • $\begingroup$ It could be rephrased as: In what ways are statistics misleadingly reported or cited? $\endgroup$ – hslc Jul 26 '10 at 21:51
  • 1
    $\begingroup$ Even if it is not off-topic, it should be community wiki. $\endgroup$ – user88 Jul 26 '10 at 22:08
  • $\begingroup$ Your re-stated question is much better. I would either suggest asking another question along those lines or better still edit the current one along the lines of your comment. $\endgroup$ – user28 Jul 26 '10 at 22:20
  • $\begingroup$ See the meta thread: meta.stats.stackexchange.com/questions/213/… where this question is proposed to be closed. $\endgroup$ – user28 Jul 29 '10 at 15:48

Statistics is about inferring something about a population, and that requires some level of interpretation.

More intuitively, "is the glass half full or half empty?". They both mean the same thing, but may have a different effect on the person who hears it.

So I would say it's the interpretation aspect which is the problem

P.S. There's an interesting article on the BBC website which may be worth a read.

P.P.S. If you meant this more generally, then there could be a case for saying that the frequentest approach to statistics can give a different result to the Bayesian approach.

  • $\begingroup$ About the glass, I think that just the boundary between phases lies in the half of its height. $\endgroup$ – user88 Jul 26 '10 at 22:10

Not the answer you're looking for? Browse other questions tagged or ask your own question.