2

I tend to use histograms of continuous variables adding estimated density curves in order to compare several charts easily. However, I find difficulties when I try to explain what density is and the interpretation of the curve's height to non-statisticians.

The stalwart Wikipedia provide us a good definition:

In probability theory, a probability density function (pdf), or density of a continuous random variable, is a function that describes the relative likelihood for this random variable to take on a given value.

Do you think the concept "relative likelihood" would be clear enough for non-statisticians? What is your approach in such situations?

Emer
  • 241
  • 2
  • 6
  • 1
    Attempting to interpret the curve's height is problematic, because **the height is meaningless:** density curves depict probability by means of *area* rather than height. (Select any location along the curve. Change the height of the curve there--and only there--to any different value. This does not change *any* areas, so it gives an equivalent density curve.) Examples: http://stats.stackexchange.com/questions/14483/intuitive-explanation-for-density-of-transformed-variable/14490#14490, http://stats.stackexchange.com/questions/4220/probability-distribution-value-exceeding-1-is-ok/4223#4223. – whuber Jan 14 '13 at 22:48

0 Answers0