- standard deviation: typical distance of the values from their mean, the square root of the variance ( the average squared deviation )
- low: values hug the mean; high: spread over a wide range
- from a sample divide by , for a whole population by
- deviations from $\bar{x}$, which was fit to the same data, come out a bit small; $n - 1$ ( degrees of freedom ) undoes that and makes $s^2$ an unbiased estimate of $\sigma^2$
- the TI-84 shows both, `Sx` is $s$ and `σx` is $\sigma$ ( [[1 Variable Stats On Calculator]] )
- example: 2, 4, 6, 8, 10
- , squared deviations 16, 4, 0, 4, 16, sum 40
- as a sample: ,
- as the whole population: ,
- shows up in every spread-based tool: standard errors, Confidence Intervals, tests in Hypothesis Testing, feature scaling in machine learning
- the TI-84’s 1-Var Stats in python, with a boxplot and a histogram marking the mean and 20% trimmed mean
Python
Output