Enter the eight values and the card lists n 8, sum 40, mean 5 — and a population standard deviation line that reads 2 with no rounding at all. The reason is that the deviations from 5 are −3, −1, −1, −1, 0, 0, 2, 4, whose squares are 9, 1, 1, 1, 0, 0, 4, 16 and add to 32. Divide by n = 8 and you get 4, whose root is 2. The example is chosen; nothing about real data is that tidy.
The card prints both denominators, and the difference between them is the whole of Bessel's correction: divide the same 32 by 7 instead of 8 and the root is 2.1380899352994. σ answers 'how spread is this set', s answers 'how spread I estimate the population to be from this sample', and on eight points the two differ by 7%, which is not a rounding issue. Note also that the mean is 5 while the median of the same eight numbers is 4.5 — a set can be perfectly ordinary and still have its two centres in different places.
Answer
2
How to read it
The card's population line, σx with denominator n. The sample line, sx with n − 1, prints 2.13809.
Source
The worked example that opens most textbook treatments of standard deviation; σ = sqrt(Σ(x − x̄)²/n), with Bessel's correction giving s = sqrt(Σ(x − x̄)²/(n − 1))
Work it in the calculator
These links open the right calculator with this problem's numbers already typed in. The line above is what the screen should show — check it yourself.