If you mean normal distribution:If you have a set of data, how can you figure out if it's normal data?
Hi Cynthia. Were you able to follow the given example? If not, which parts do you need help understanding?How would I determine whether or not the data is "normal data"?
The word is "empirical", not "imperial"; read the paper carefully!I won't give you results or formulas (maybe a small one) .In this problem you have to:
68% range is calculated by using mean\(\displaystyle \pm\)deviation.
- calculate mean and median
- calculate standard deviation
- sketch a histogram to check for bell rule
- check for imperial rule. The imperial rule is I believe checked by first calculating the margin of error in ranges of data (68%,95%, and 99.7%). Senior members feel free to fix the theoretical jargon if I am wrong. See below.
For 95% range the deviation factor is 2 which translates to [Mean\(\displaystyle \pm\)2*deviation]
For 99.7% range you can already guess.
Later we count the numbers that fall into these three ranges. I'm not sure how to be mathematically certain but if almost all data falls into at least 2 ranges the data should be normal in my opinion. I would be worried if more than 5 numbers don't fall in the ranges though.
Ah sorry for mixing up "empirical" and "imperial" .I will be more careful next time! I wish I could edit it out.The word is "empirical", not "imperial"; read the paper carefully!
As for mathematical certainty, that's impossible. There are much more advanced techniques, discussed in the article in post #3. At this level, you are just making a largely subjective determination whether the data seem approximately normal. That word is repeated many times in the assignment.