What do you mean by the spread? Are you looking at the distribution of 1's and 2's? You should be getting about 5000 +/- 150 ones.or twos. It is rare to go beyond this. One quick way to check is to take your column of data and fx =sum(A1:A10000) - 10000 (This is the equivalent of using 0 and 1 and adding up the 1s) in a cell. This will tell you the number of twos. Because this is symmetric, you can analyze the twos or you can do 10000 = N(twos) to get N(ones). A "300 spread" if you are looking at the difference between ones and twos is within the 3 sigma. I'm sorry if I misunderstood the problem.
Kim N.
Thanks a lot for the response. Yes, you understood the problem correctly. And yes by 'Spread' I was referring to the difference in amount of 1s versus 2s. Thank you for your clarification. I am very surprised to see that excel seems to big varians between the amount of 1s and 2s, since it should be fairly evenly distributed (50%). One last thing I am a bit unsure about: I am looking at the difference in amount of 1s vs 2s throughout the 10.000 cells. Here is a good example. https://ibb.co/BBLqq1C . (For every "1" the graph adds +1. For every "2" the graph substracts -1) . and the data shows around 200 more 1s than mean value, at the peak point on the graph. but then it ends a bit lower, just around +100 1s, above the mean value. So should I concider the peak point relevant? or should I simply base my conclusion on the ~100 surplus of 1s in the end? I hope it makes sense, I apologize for any confusion due to my english. It's not my first language.04/03/22