0

I was making histograms and depending on the bin width the data either appeared skewed-right or roughly symmetric. Is there an actual mathematical way of proving skewness using a formula or calculation of some kind aside from looking at the graph?

a13a22
  • 163
  • 12
  • If you click on the "skewness" tag (gray box right below your question) there may be some answers that could help, such has [this](http://stats.stackexchange.com/questions/145159/how-to-tell-if-my-data-distribution-is-symmetric) (#2 on the vote-sorted list that comes up). – GeoMatt22 Sep 13 '16 at 01:14
  • 1
    Note that there are numerous measures of skewness. While you can do hypothesis tests with them, a skewness of approximately zero (by any of the measures) doesn't imply symmetry. Why do you want to *prove* skewness (rather than have a good way to investigate distribution shape, say)? What's the underlying problem? – Glen_b Sep 13 '16 at 03:04
  • On an introduction to Statistics quiz, I was marked wrong for describing it as right-skewed. Some students told me it was symmetric and when I changed the bin width my answer changed – a13a22 Sep 14 '16 at 02:11

0 Answers0