Standard deviation
|
Formula to find mean of a set of data | |
|---|---|
Ungrouped Data: | \(\bar{x} = \dfrac{\sum x}{n}\) |
Grouped Data: | \(\dfrac{\sum fx}{\sum f}\) |
Formula to find standard deviation \(\sqrt{\dfrac{\sum x^2}{n} - \bar{x}^2} \quad \text{or} \quad \sqrt{\dfrac{\sum(x - \bar{x})^2}{n}}\) |
Example:
Find the mean and standard deviation for the following set of data.
\(2 \qquad 3 \qquad 3 \qquad 4 \qquad 6 \qquad 7 \qquad 8 \qquad 9 \qquad 9\)
For the example above, \(n=9\), \(\sum x=51\) and \(\sum x^2=349\). So \(\bar x=\dfrac{51}9=\dfrac{17}3\approx5.67\) and \(\sigma=\sqrt{\dfrac{349}9-\left(\dfrac{17}3\right)^2}=\sqrt{\dfrac{20}3}\approx2.58\).
Why the standard deviation is built this way
Set A, \(4,5,6,7,8\), and Set B, \(2,3,6,9,10\), both have mean \(6\), but Set B is more spread out. Measure each value’s distance from the mean, the deviation \(x-\bar x\).
| \(x\) (Set A) | \(x-\bar x\) | \((x-\bar x)^2\) | \(x\) (Set B) | \(x-\bar x\) | \((x-\bar x)^2\) |
|---|---|---|---|---|---|
| \(4\) | \(-2\) | \(4\) | \(2\) | \(-4\) | \(16\) |
| \(5\) | \(-1\) | \(1\) | \(3\) | \(-3\) | \(9\) |
| \(6\) | \(0\) | \(0\) | \(6\) | \(0\) | \(0\) |
| \(7\) | \(1\) | \(1\) | \(9\) | \(3\) | \(9\) |
| \(8\) | \(2\) | \(4\) | \(10\) | \(4\) | \(16\) |
| Sum | \(0\) | \(10\) | Sum | \(0\) | \(50\) |
A frequency table
In a frequency table each value \(x\) occurs \(f\) times, so \(\sum f=n\), \(\sum fx\) replaces \(\sum x\) and \(\sum fx^2\) replaces \(\sum x^2\). The table shows the number of pets, \(x\), owned by \(20\) pupils.
| \(x\) | \(f\) | \(fx\) | \(fx^2\) |
|---|---|---|---|
| \(1\) | \(3\) | \(3\) | \(3\) |
| \(2\) | \(6\) | \(12\) | \(24\) |
| \(3\) | \(8\) | \(24\) | \(72\) |
| \(4\) | \(2\) | \(8\) | \(32\) |
| \(5\) | \(1\) | \(5\) | \(25\) |
| Sum | \(\sum f=20\) | \(\sum fx=52\) | \(\sum fx^2=156\) |
\(\bar x=\dfrac{\sum fx}{\sum f}=\dfrac{52}{20}=2.6\) and \(\sigma=\sqrt{\dfrac{\sum fx^2}{\sum f}-\bar x^2}=\sqrt{\dfrac{156}{20}-2.6^2}=\sqrt{1.04}\approx1.02\).
A calculator’s statistics mode gives the same values: enter each value with its frequency, check that the count shown is \(20\), and read the standard deviation that divides by \(n\) (often labelled \(\sigma_x\)), not the one that divides by \(n-1\).
Checks and common errors
Video solution locked
A free account saves your progress; this video needs a course plan.
680+ questions here come with free worked solutions. A free account saves your questions and progress; this one needs a course plan.
Create a free account to practise similar questions.
Need help? Join our E-Maths tuition for the SEC, the exam that replaces the O-Levels.
Learn more