2

I have been tasked with determining if my observed data follow a normal distribution.

With 119 observations (continuous data from -4 to 4), I was able to create a histogram with 25 bins, and upon observation it looks bell-shaped.

I was recommended to do a chi-squared for goodness of fit test to determine if the data is normally distributed, but in the examples I've seen the data is categorical. So, I'm stuck.

gung - Reinstate Monica
  • 132,789
  • 81
  • 357
  • 650
user3719974
  • 121
  • 1
  • The first half of my answer to the duplicate question explains the procedure generally. For examples of its application, please [search our site](http://stats.stackexchange.com/search?q=chi-squared+goodness+of+fit+test). – whuber Nov 30 '14 at 02:14
  • (1) If your data are real data, they aren't going to be normal (exactly? how would that work?); why try getting a noisy answer a question you already know answer to? $\,$ (2) don't use a chi-square goodness of fit test to test normality; if you must, there are much better choices (Shapiro Wilk is far better, for example). $\,$ (3) If you're assumption checking, formal testing of goodness of fit before selecting an analysis on the basis of its outcome may be counterproductive (compared to simply not assuming normality). – Glen_b Nov 30 '14 at 02:34

0 Answers0