0

A friend of mine was talking about how method of data presentation can influence people's interpretation of the data, and I was wondering how much truth there is to it. The specific example was a controversial diagram, showing the difference in IQ scores between Black and White Americans as two separated bell curves, featured in 'The Bell Curve: Intelligence and Class Structure in American Life'. The difference looks large but having seen the data on a scatter plot, it's hardly noticeable. Ignoring the moral issues of the example, does anyone know anything about how choice of data presentation may seem to exaggerate results/ is there any research on the subject?

Guy4164
  • 11
  • Yes data presentation can influence people's interpretation of data. However this effect has little to do with mathematics. I don't know what SE site this question would be best suitable for (Politics maybe?), but it's very likely not this one. – Winther Oct 17 '16 at 10:45
  • As Winther commented, depending of what you want to prove, you can selection a representation supporting your argument. Take for example the price of a share on Stock Exchange; duringa year, it varied between $100$ and $10$. Make a plot of the prices as a function of time, the $y$ axis starting at $0$: big fluctuations, indeed. Do the same but, now, he $y$ axis starting at $1000$. Thia is avery good question. $+1$. – Claude Leibovici Oct 17 '16 at 10:55
  • A nice starting point for these kind of question is the book \emph{How to lie with statistics} by D. Huff. It's old but far from being obsolete. See https://en.wikipedia.org/wiki/How_to_Lie_with_Statistics for more on this topic. – mlc Oct 17 '16 at 13:11

0 Answers0