Hugnot
Well-Known Member
I suspect its using decimals that it isn't displaying on screen. With such a tiny sample size and such a tight grouping, rounding up can change the calculation
I suspect its using decimals that it isn't displaying on screen. With such a tiny sample size and such a tight grouping, rounding up can change the calculation.
View attachment 521164
Two different data sampling types. I interpreted the data as a population as there were 3 measurements. The other method would be to interpret the data as a sample, but no other data was available.
This resulted in the sum of squares, 4.047 being divided by N or 3 for a population or N-1 for a sample like 2.
My spread sheet shows a sum of squares value of 4.047, and a variance of 2.023 (estimated variance via division of population minus 1)
This agrees with your calculations. Gathering more data will reduce the 1.4 SD that was displayed by the Magneto Speed.
Actual MS data calculations were unknown, but MS treated it as a sample, probably expecting more data. The difference was not caused by "rounding up" it was caused by two different data sampling methods - population (like everything) vs sample (part of everything). Or dividing sum of squares by N-1 (3-1) or N (3).
I enter MS data into my spreadsheet(s) so I can look & compare. The results have adequate accuracy despite whole number data entries. Open Office is much easier to use than On-Line Excel.
So sorry about the pasting remark - I got a history of dealing with data treachery. A good example of which is faking electronic ballot images used for verification. My bad
Last edited: