3

I often see statistical abominations in a business context (be it in advertisement, in the way market research is conducted, etc.). I am wondering whether or not the misuse of statistics is a general practice.

Is there any evidence (research papers for instance) highlighting the misuse (or good use) of statistics in business decision making ? Or more generally, evidence of a lack (or presence) of statistical "literacy" (within the business population) ?

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
Pholochtairze
  • 460
  • 4
  • 12
  • You are asking to prove a negative which is very nard to impossible. – forecaster Feb 01 '17 at 13:38
  • 6
    I think the question is rather broad but it seems in order. Possible answers could be research papers and examples from various media. But where do you stop? There are for example enough examples of lousy business graphics to fill many encyclopedias. I've reviewed several books that are intended primarily or largely for business people that are themselves often wrong or in my view misguided statistically. – Nick Cox Feb 01 '17 at 13:53
  • 3
    @NickCox I'd even say business is not the only field 'afflicted' with this problem. So if you really need sources for this, you could also look beyond the borders of economics, 'broadening' (is that an English word?) a possible answer even further. – IWS Feb 01 '17 at 14:02
  • 2
    @IWS Absolutely. No field is perfect. But broadening [sounds OK to me] a broad question is not the way to go. – Nick Cox Feb 01 '17 at 14:04
  • @NickCox Indeed, I agree my question is broad. The thing is that we can find lots of examples of errors online. Nevertheless, even if we find 1000 examples of errors, if there are millions of correct analysis then it's normal. That's why I was wondering if a more rigorous analysis had been conducted. – Pholochtairze Feb 01 '17 at 14:04
  • 1
    In medical statistics people have surveyed publications and pronounced retrospectively on what fraction of papers included correct analyses. Since even statistical people disagree on what is correct, I've always felt queasy about such exercises. I don't know about anything similar in business. One key problem is that in business many analyses are never published. – Nick Cox Feb 01 '17 at 14:16
  • @NickCox I don't know if you've read the book "Statistics done wrong" that describes lots of those studies about medical statistics. I saw that indeed there were lots of mistakes being made but was curious if such studies had been done in business. I couldn't find papers about statistical mistakes in a business context (for the reason you mentionned: it's generally kept within the company), that's why I asked the question on stats.stackexchange to know if someone had heard of such studies! – Pholochtairze Feb 01 '17 at 14:25
  • 1
    I don't think I read this right when I voted to close it. If it's references you're after (rather than asking CV users to provide examples), then it's in order, as @NickCox says. I've added the `references` tag & edited it slightly. – Scortchi - Reinstate Monica Feb 01 '17 at 14:46

0 Answers0