On this concept: Formula

Not sure which lesson to do first? Find your level and get a plan →

Statistics · Grade 6-8

Standard Deviation

Standard deviation is a measure of how spread out data values are from the mean, representing the typical distance of data points from the average.

σ=∑(x−μ)2n

Quick Answer

Standard deviation is a measure of how spread out data values are from the mean, representing the typical distance of data points from the average. To find it, take each value's distance from the mean, square it, average the squares, then take the square root. For 2, 4, 4, 4, 5, 5, 7, 9 the mean is 5 and the standard deviation is 2. Use it to compare how consistent groups are and to judge whether a value is ordinary or unusual: in bell-shaped data, about two thirds of the values lie within one standard deviation of the mean.

Why This Matters

Standard deviation is the spread measure used almost everywhere: in factory quality checks, weather records, exam scoring and science experiments. It gives a standard way to say how unusual a value is, so a height two standard deviations above the mean is known to be rare while one just above the mean is not. Without it, "far from average" has no number that means the same thing in different data sets.

Intuitive Explanation

Start, as with the MAD, with how far each value is from the mean. For the data 4, 8, 6, 2 the mean is 5 and the deviations are −1,3,1,−3. Instead of dropping the signs, standard deviation squares the deviations: 1, 9, 1, 9. Squares are never negative, so nothing cancels. Their average, 20÷4=5, is called the variance. Its unit is squared, so take the square root to get back to the data's own unit: 5≈2.24.

Squaring gives far-away values extra weight. A value 3 away from the mean contributes 9, while a value 1 away contributes only 1. That is why the standard deviation is usually a little larger than the MAD of the same data: here the MAD is (1+3+1+3)÷4=2, while the standard deviation is about 2.24. In return, the standard deviation has properties that make it the main tool of later statistics, such as the rule that in bell-shaped data about 68% of values lie within one standard deviation of the mean.

Standard deviation is not the variance, and it is never negative. If a calculator gives a variance of 25 for some test scores, the standard deviation is 25=5 points, not 25. A value below the mean does not have a negative standard deviation either: the signed distance of one value from the mean is its deviation, while the standard deviation describes the spread of the whole data set.

Think of it like this. If the mean is home, standard deviation is the typical distance people wander from it, with long trips counting extra. One friend who strays 10 blocks adds as much to the squared total as 100 friends who each stray 1 block, so a single far-off value raises the standard deviation sharply.

Core idea. Standard deviation is the typical distance of the values from the mean, found by squaring the distances, averaging them and taking the square root; squaring makes far-away values count more, and the square root returns the answer to the data's units.


When to Use

Use standard deviation when a question asks how spread out or consistent data are around the mean, names σ or s, or asks how many standard deviations a value is from the mean. It pairs with the mean for roughly symmetric data. For a quick calculation by hand, the MAD gives a similar picture; if the data are skewed or have large outliers, use the median with the IQR instead.

Tip. Ask: Do you need the typical distance from the mean, found by squaring the distances, averaging, then taking a square root?

How to Recognize It

Before reaching for a calculator, check that the question wants a spread measured from the mean, and which kind of spread.

  1. Is the spread measured around the mean, with every value counted?

    Yes points to the standard deviation or the MAD. If the spread uses only the middle half of the sorted data, it is the Interquartile Range (IQR).

  2. Are the distances from the mean squared before they are averaged?

    Yes is the standard deviation, or the variance if no square root is taken. Plain distances averaged without squaring give the Mean Absolute Deviation (MAD).

  3. Is the answer one spread for the whole group, rather than a position for one value?

    Yes is the standard deviation. Saying how many standard deviations one value lies from the mean is a Z-Score (Standard Score).

  4. Must the answer be in the same units as the data, such as centimetres or points?

    Yes means take the square root to get the standard deviation. Squared units, such as square centimetres, belong to the variance, which is the step before.

Standard Deviation vs Mean Absolute Deviation (MAD) vs Interquartile Range (IQR) vs Z-Score (Standard Score)

Three of these are spreads and the fourth is built from one, so a problem with means and standard deviations can need any of them. Ask whether you need a squared-distance spread, a plain-distance spread, a middle-half width, or one value's position.

Standard Deviation

Meaning
The standard deviation is the square root of the average squared distance from the mean, in the same units as the data.
Key test
Squared distances from the mean; a single far value raises it sharply.
Formula
σ=1n∑i=1n(xi−μ)2
Example
Times of 3, 5 and 7 minutes have a mean of 5 and a standard deviation of about 1.63 minutes.

Mean Absolute Deviation (MAD)

Meaning
The MAD averages the plain distances from the mean without squaring them.
Key test
No squares or roots; every distance counts at its true size.
Formula
1n∑∣x−xˉ∣
Example
Times of 3, 5 and 7 minutes have a MAD of 2+0+23≈1.33 minutes.

Interquartile Range (IQR)

Meaning
The IQR is the distance between the lower and upper quartiles of the sorted data.
Key test
Ignores the top and bottom quarters; used with the median for lopsided data.
Formula
IQR=Q3−Q1
Example
House prices with Q1 at $200k and Q3 at $350k have an IQR of $150k.

Z-Score (Standard Score)

Meaning
A z-score tells how many standard deviations one value lies above or below the mean.
Key test
Describes one value, not the group, and has no units.
Formula
z=x−μσ
Example
With mean 70 and standard deviation 8, a score of 58 has z=−1.5.

Formula & Notation

σ=∑(x−μ)2n
For a population: σ=1N∑i=1N(xi−μ)2. For a sample: s=1n−1∑i=1n(xi−xˉ)2.

How to read it: σ is the population standard deviation, s is the sample standard deviation, σ2 is the variance. The units of SD are the same as the original data.

Worked Examples

Example 1 — Standard deviation of bus delays

Easy

A bus was 1, 3, 5, 7 and 9 minutes late on five days. Find the standard deviation of the delays (divide by the number of values).
Show the steps and the answer
  1. Find the mean: 1+3+5+7+9=25, and 25÷5=5 minutes.

    The standard deviation measures distances from the mean, so the mean comes first.

  2. Find the deviations: −4,−2,0,2,4.

    Each deviation is a value minus the mean.

  3. Square and average: 16+4+0+4+16=40, and 40÷5=8.

    Squaring removes the signs; 8 is the variance, in square minutes.

  4. Take the square root: 8≈2.83 minutes.

    The square root brings the answer back to minutes, the unit of the data.

Answer: The standard deviation is about 2.83 minutes.

Takeaway: Mean, deviations, squares, average, square root: five steps, and the answer is in the data's own unit.

Example 2 — Same MAD, different standard deviation

Standard

Data set A is 4, 4, 6, 6. Data set B is 3, 5, 5, 7. Both have mean 5. Find the MAD and the standard deviation of each.
Show the steps and the answer
  1. A: the distances from 5 are 1, 1, 1, 1, so MAD=4÷4=1.

    Every value in A sits exactly 1 away from the mean.

  2. B: the distances are 2, 0, 0, 2, so MAD=4÷4=1.

    The MAD weighs each unit of distance equally, so B ties with A.

  3. A: the squares are 1, 1, 1, 1, with average 1, so the standard deviation is 1=1.

    When every distance is the same, the MAD and the standard deviation agree.

  4. B: the squares are 4, 0, 0, 4, with average 2, so the standard deviation is 2≈1.41.

    Squaring makes B's two values at distance 2 count four times as much as a distance of 1.

Answer: Both MADs are 1. The standard deviation is 1 for A and about 1.41 for B.

Takeaway: Standard deviation is not just another MAD: it gives extra weight to values far from the mean, so it can separate data sets the MAD treats as equal.

Example 3 — Find the mistake: variance reported as standard deviation

Application

Four plants are 10, 14, 16 and 20 cm tall. Rosa works out: "The mean is 15. The squared distances are 25, 1, 1, 25, and their average is 13, so the standard deviation is 13 cm." Find the mistake and correct it.
Show the steps and the answer
  1. Check Rosa's steps: the mean is 60÷4=15, and the squared distances add to 52, with average 52÷4=13.

    Everything up to 13 is correct; 13 is the variance.

  2. Take the square root: 13≈3.61 cm.

    The variance is in square centimetres; the square root returns to centimetres.

  3. Sense-check: the plants are only 1 cm and 5 cm from the mean, so a typical distance of 13 cm is impossible.

    A standard deviation must fall between the smallest and largest distances from the mean, here 1 and 5.

Answer: The standard deviation is about 3.61 cm (13 is the variance).

Takeaway: Finish with the square root: the average of the squares is the variance, not the standard deviation.

Where You'll See It

  • How tightly a vending machine fills cans

    A machine aiming for 355 mL has a small standard deviation if nearly every can is 354-356 mL, and a large one if fills swing from 340 to 370 mL - the SD measures the typical wander from the target.

  • Consistency of two basketball shooters

    Two players both average 18 points a game, but the one with the smaller standard deviation scores near 18 every night while the larger-SD player swings between 4 and 40.

  • Test scores clustered vs scattered

    If a class averages 85 with a small standard deviation, most students scored near 85; a large standard deviation means scores were spread widely above and below.

Common Mistakes

  1. The slip: Standard deviation can be negative when data is below the mean.

    Why it's tempting: Many data points sit below the mean, so a negative spread feels possible, but the formula squares every deviation before taking a (positive) square root.

    Instead: Standard deviation is always greater than or equal to zero; squaring removes the sign, and the square root is non-negative.

  2. The slip: Standard deviation and variance are the same number.

    Why it's tempting: Both measure spread and the SD is just the square root of variance, so students report one for the other.

    Instead: Variance is the average squared deviation; standard deviation is its square root and carries the same units as the data.

  3. The slip: You can directly compare standard deviations from two data sets in different units.

    Why it's tempting: Both are 'just a spread number', so comparing them seems fair, but units (cm vs kg, or dollars vs seconds) make them incomparable.

    Instead: Only compare standard deviations when the data are in the same units; otherwise use a unitless measure like the coefficient of variation.


Mini Practice

Try these on your own. Tap Reveal when you want to check.

  1. Find the standard deviation σ of 2, 4, 4, 4, 5, 5, 7, 9.

    Hint: The mean is 5. Square each distance from 5, average them, then take the square root.

  2. Find σ for 6, 8, 10, 12, 14.

    Hint: Find the mean first, then follow the formula.

  3. Every value in 6, 8, 10, 12, 14 increases by 5. What happens to the mean and to the standard deviation?

    Hint: Does adding 5 change how far the values are from each other?

  4. Set P is 48, 50, 52 and Set Q is 40, 50, 60. Both have mean 50. Find σ for each set.

    Hint: Square the distances from 50 and average them before the square root.

  5. A factory wants every bolt it makes to be very close to the same length. Which number shows how consistent the lengths are: the mean length or the standard deviation?

    Hint: Which one measures distance from the center?

  6. A student finds that the variance of some plant heights is 16 cm² and reports: "The standard deviation is 16 cm." What is the standard deviation?

    Hint: How are variance and standard deviation related?

Want the full set?

50 practice questions for this concept — free to try, every one with a complete worked solution showing the why, not just the answer.

Frequently Asked Questions

Why do some calculators divide by n−1 instead of n?

Dividing by nn is right when your data are the whole group you care about. When the data are a sample from a bigger group, the distances are measured from the sample mean, which is the point the sample values sit closest to, so the spread usually comes out slightly too small. Dividing by n−1n - 1 corrects this. For 2, 4, 4, 4, 5, 5, 7, 9 the two answers are 2 and about 2.14.

What happens to the standard deviation if I add 10 to every value, or double every value?

Adding 10 to every value slides the whole data set along without changing any of the distances, so the standard deviation stays the same: 2, 4, 6 and 12, 14, 16 both have about 1.63. Doubling every value doubles every distance, so the standard deviation doubles too: 4, 8, 12 has about 3.27. The mean moves in both cases, but only stretching the data changes the spread.

Is a standard deviation of 5 big or small?

It depends on the size of the values. A standard deviation of 5 cm in the heights of adults, who are about 170 cm tall, is a modest spread. The same 5 cm in the lengths of newborn babies, who are about 50 cm long, would be huge. Compare the standard deviation with the mean, or with the range of values that make sense, before calling it big or small.

Learning Path

Standard Deviation

You are here

You need Mean as Fair Share (finding the mean) and Data Variability (what spread means) first. Standard deviation adds a precise measure of spread: square the distances from the mean, average them, and take the square root. Next, the Z-Score (Standard Score) uses it as a ruler, z=x−μσ, to say how many standard deviations a value lies from the mean.

See Also