algebra911.com

Standard Deviation

Q: Why do we divide by n-1 when calculating the standard deviation for a sample?

We divide by $n-1$ (Bessel's correction) to get a better, unbiased estimate of the true population standard deviation. A sample tends to have slightly less variability than the full population, so dividing by a smaller number makes our result a bit larger and more accurate.

Q: My calculator has two standard deviation buttons (σ and s). Which one do I use?

Use the $\sigma$ button (often labeled $\sigma_n$ or $\sigma_x$) when your data represents the entire population. Use the $s$ button (often labeled $s_n-1$ or $s_x$) when your data is a sample of a larger population. The $s$ button uses the $n-1$ formula.

Download as PDF

Have you ever wondered how we can measure the consistency of a set of numbers? Standard deviation is a powerful tool in mathematics that tells us exactly that. It measures how spread out or clustered together data points are, giving us a single number to describe the variability in a data set.

Standard Deviation — an original Algebra911 reference diagram defining standard deviation with its key formula and a worked example. — What Is Standard Deviation? A Beginner's Guide

What Is Standard Deviation?

Standard deviation is a measure of how spread out the numbers in a data set are from their average, or mean. A low standard deviation means that the numbers are very close to the average, indicating consistency. A high standard deviation means the numbers are spread out over a wider range, indicating variability.

Imagine two different basketball players, Alex and Ben, who both average $10$ points per game. Their average is the same, but that doesn't tell the whole story. Let's look at their scores over five games:

Alex's scores: $9, 10, 10, 11, 10$
Ben's scores: $0, 20, 5, 15, 10$

Just by looking, you can see that Alex is a very consistent player. His scores are all clustered right around his average of $10$ . Ben, on the other hand, is all over the place. Sometimes he scores a lot, and sometimes he scores very little. Even though their average is the same, their performance is completely different. Standard deviation is the mathematical tool that captures this difference. Alex would have a very low standard deviation, while Ben's would be very high.

What Are Mean and Variance?

Before we can master standard deviation, we need to understand its two key building blocks: mean and variance.

The Mean (Average)

You've likely calculated the mean many times before. It's simply the average of all the numbers in your data set. To find it, you add up all the values and then divide by the count of how many values there are.

In statistics, the mean of a whole population is represented by the Greek letter mu, $μ$ . The formula is:

μ = \frac{\sum x_{i}}{N}

Here, $\sum$ (sigma) means "sum of", $x_{i}$ represents each individual value in the data set, and $N$ is the total number of values.

Variance

Variance is the next step, and it measures the average of the squared differences from the mean. That sounds complicated, but let's break it down.

First, we find the deviation for each data point, which is just how far it is from the mean ( $x_{i} - μ$ ). Some of these will be positive, and some will be negative.
Next, we square each of these deviations. Why? Squaring does two important things: it makes all the values positive (so they don't cancel each other out) and it gives more weight to values that are farther from the mean.
Finally, we find the average of all these squared deviations. This average is the variance, represented by $σ^{2}$ (sigma squared).

The formula for population variance is:

σ^{2} = \frac{\sum (x_{i} - μ)^{2}}{N}

Variance is a great measure of spread, but its units are squared (e.g., points-squared), which isn't very intuitive. To fix that, we move to the final step: standard deviation.

How Do You Calculate Standard Deviation?

Standard deviation is simply the square root of the variance. Taking the square root gets us back to the original units of our data (e.g., from points-squared back to points), making it much easier to interpret. The symbol for population standard deviation is $σ$ .

σ = \sqrt{\frac{\sum_{i = 1}^{N} (x_{i} - μ)^{2}}{N}}

Here is the full, step-by-step process to calculate it:

Step 1: Find the Mean ( $μ$ ). Add up all your data points and divide by the number of points.
Step 2: Calculate the Deviations. For each data point, subtract the mean from it.
Step 3: Square the Deviations. Square each of the results from Step 2.
Step 4: Find the Variance ( $σ^{2}$ ). Add up all the squared deviations from Step 3 and divide by the total number of data points.
Step 5: Find the Standard Deviation ( $σ$ ). Take the square root of the variance from Step 4.

Example 1

Let's calculate the standard deviation for a set of five quiz scores: ${8, 9, 10, 11, 12}$ .

Step 1: Find the Mean ( $μ$ )

μ = \frac{8 + 9 + 10 + 11 + 12}{5} = \frac{50}{5} = 10

Steps 2 & 3: Calculate and Square the Deviations

We can organize this work in a table:

Score ( $x_{i}$ )	Deviation ( $x_{i} - μ$ )	Squared Deviation ( $(x_{i} - μ)^{2}$ )
$8$	$8 - 10 = - 2$	$(- 2)^{2} = 4$
$9$	$9 - 10 = - 1$	$(- 1)^{2} = 1$
$10$	$10 - 10 = 0$	$0^{2} = 0$
$11$	$11 - 10 = 1$	$1^{2} = 1$
$12$	$12 - 10 = 2$	$2^{2} = 4$

Step 4: Find the Variance ( $σ^{2}$ )

Now we add up the 'Squared Deviation' column and divide by the number of scores ( $N = 5$ ).

σ^{2} = \frac{4 + 1 + 0 + 1 + 4}{5} = \frac{10}{5} = 2

Step 5: Find the Standard Deviation ( $σ$ )

Finally, we take the square root of the variance.

σ = \sqrt{2} \approx 1.414

So, the standard deviation of the quiz scores is approximately $1.414$ .

Let's Try Another Example

Practice makes perfect. Let's work through another example with slightly more challenging numbers. This time, we'll look at the number of minutes five friends spent on homework one night.

Example 2

Calculate the standard deviation for the data set: ${30, 45, 50, 35, 60}$ minutes.

Step 1: Find the Mean ( $μ$ )

μ = \frac{30 + 45 + 50 + 35 + 60}{5} = \frac{220}{5} = 44

The mean is $44$ minutes.

Step 2 & 3: Calculate and Square the Deviations

For $30$ : $30 - 44 = - 14$ , and $(- 14)^{2} = 196$
For $45$ : $45 - 44 = 1$ , and $1^{2} = 1$
For $50$ : $50 - 44 = 6$ , and $6^{2} = 36$
For $35$ : $35 - 44 = - 9$ , and $(- 9)^{2} = 81$
For $60$ : $60 - 44 = 16$ , and $16^{2} = 256$

Step 4: Find the Variance ( $σ^{2}$ )

Sum the squared deviations and divide by $N = 5$ .

σ^{2} = \frac{196 + 1 + 36 + 81 + 256}{5} = \frac{570}{5} = 114

The variance is $114$ minutes-squared.

Step 5: Find the Standard Deviation ( $σ$ )

Take the square root of the variance.

σ = \sqrt{114} \approx 10.677

The standard deviation for the homework times is about $10.68$ minutes. This tells us that the time spent on homework was quite variable among this group of friends.

What Does a High or Low Standard Deviation Mean?

The number you calculate for standard deviation is a powerful descriptor of your data set.

A low standard deviation indicates that the data points tend to be very close to the mean. This implies consistency and predictability. For example, a company that manufactures smartphone screens wants a very low standard deviation in screen thickness.
A high standard deviation indicates that the data points are spread out over a wide range of values. This implies variability. For example, the daily sales at an ice cream shop might have a high standard deviation, with low sales on cold days and very high sales on hot days.

Example 3

Let's compare two classes that took the same test. Both classes have an average score of $85$ , but their scores are different.

Class A Scores: ${82, 84, 85, 86, 88}$
Class B Scores: ${70, 80, 85, 90, 100}$

For Class A:

Mean $μ = 85$ .
Squared deviations are: $(82 - 85)^{2} = 9$ , $(84 - 85)^{2} = 1$ , $(85 - 85)^{2} = 0$ , $(86 - 85)^{2} = 1$ , $(88 - 85)^{2} = 9$ .
Variance $σ^{2} = \frac{9 + 1 + 0 + 1 + 9}{5} = \frac{20}{5} = 4$ .
Standard Deviation $σ = \sqrt{4} = 2$ .

For Class B:

Mean $μ = 85$ .
Squared deviations are: $(70 - 85)^{2} = 225$ , $(80 - 85)^{2} = 25$ , $(85 - 85)^{2} = 0$ , $(90 - 85)^{2} = 25$ , $(100 - 85)^{2} = 225$ .
Variance $σ^{2} = \frac{225 + 25 + 0 + 25 + 225}{5} = \frac{500}{5} = 100$ .
Standard Deviation $σ = \sqrt{100} = 10$ .

Conclusion: Class A has a standard deviation of $2$ , while Class B has a standard deviation of $10$ . Even though both classes had the same average, the scores in Class B were much more spread out. This single number clearly shows that the students' performance in Class A was far more consistent.

Key formulas for standard deviation by Algebra911.

What's the Difference Between Population and Sample Standard Deviation?

This is a slightly more advanced but very important concept. So far, we've been calculating the population standard deviation, which assumes our data set includes every single member of the group we're interested in.

A Population is the entire group. For example, all students in a specific school.
A Sample is a smaller part of that group. For example, 30 students selected from that school.

Most of the time in real life, it's impossible to get data for an entire population. Instead, we take a sample and use it to estimate the standard deviation of the whole population. When we do this, we need to use a slightly different formula called the sample standard deviation.

The only difference is that instead of dividing by the number of data points $N$ , we divide by $n - 1$ (where $n$ is the sample size).

Population Standard Deviation ( $σ$ ):

σ = \sqrt{\frac{\sum (x_{i} - μ)^{2}}{N}}

Sample Standard Deviation ( $s$ ):

s = \sqrt{\frac{\sum (x_{i} - \bar{x})^{2}}{n - 1}}

Notice the different symbols: $μ$ is the population mean, while $\bar{x}$ is the sample mean. $σ$ is the population SD, while $s$ is the sample SD.

Why $n - 1$ ? This is known as Bessel's correction. In simple terms, a sample is usually less spread out than the full population it came from. Dividing by a slightly smaller number ( $n - 1$ instead of $n$ ) makes the final result a little bit bigger. This inflation gives us a more accurate and unbiased estimate of the true population's standard deviation.

For most problems in your class, you'll be told whether you're working with a population or a sample. If it's not specified, it's common to treat the data as a full population.

What Are Some Common Mistakes?

When calculating standard deviation, it's easy to make small errors. Here are a few common pitfalls to watch out for:

Forgetting the Final Square Root: A very common mistake is to do all the work to find the variance ( $σ^{2}$ ) and then forget to take the square root to find the standard deviation ( $σ$ ). Always double-check that you've completed this final step!
Mixing Up Formulas: Be careful to use the correct formula. If you have data for the entire population, divide by $N$ . If you have a sample of a larger population, divide by $n - 1$ .
Calculation Errors: Squaring negative numbers can be tricky. Remember that a negative number squared is always positive (e.g., $(- 4)^{2} = 16$ , not $- 16$ ). Use a calculator and be careful with your order of operations.
Misinterpreting the Result: Don't assume that a "high" standard deviation is bad or a "low" one is good. It always depends on the context. High variability is great if you're analyzing a diverse investment portfolio, but it's terrible if you're manufacturing precision engine parts.

Quick Reference Summary

Here's a quick summary of the key ideas about standard deviation.

What it is: A number that measures how spread out data points are from their mean.
Low SD means: The data is clustered tightly together. It is consistent and predictable.
High SD means: The data is spread far apart. It is variable and less consistent.
The 5 Steps to Calculate (Population SD):
1. Calculate the mean ( $μ$ ).
2. Find the deviation for each point ( $x_{i} - μ$ ).
3. Square each deviation ( $(x_{i} - μ)^{2}$ ).
4. Average the squared deviations to get the variance ( $σ^{2}$ ).
5. Take the square root of the variance to get the standard deviation ( $σ$ ).
Key Formulas:
- Population SD: $σ = \sqrt{\frac{\sum (x_{i} - μ)^{2}}{N}}$
- Sample SD: $s = \sqrt{\frac{\sum (x_{i} - \bar{x})^{2}}{n - 1}}$

Frequently Asked Questions

Can standard deviation be a negative number?

No, standard deviation can never be negative. This is because it is calculated by taking the square root of the variance, which is the average of squared numbers. Since squared numbers are always non-negative, their average is also non-negative, and so is its square root.

What is the standard deviation of a single number or a set of identical numbers?

The standard deviation is zero. If you have only one number or all the numbers in a set are the same, there is no spread or variability. All data points are equal to the mean, so the deviations are all zero.

What's the difference between variance and standard deviation?

Standard deviation is the square root of the variance. The main advantage of standard deviation is that it is expressed in the same units as the original data (e.g., points, inches, dollars), making it much easier to interpret than the squared units of variance.

Why do we divide by n-1 when calculating the standard deviation for a sample?

We divide by $n - 1$ (Bessel's correction) to get a better, unbiased estimate of the true population standard deviation. A sample tends to have slightly less variability than the full population, so dividing by a smaller number makes our result a bit larger and more accurate.

Is a bigger standard deviation always a bad thing?

Not at all. It completely depends on the context. In manufacturing, you want a low standard deviation for product consistency. But in investing, a higher standard deviation might indicate higher risk but also the potential for higher returns.

How does an outlier affect the standard deviation?

An outlier, which is a data point far away from the others, will significantly increase the standard deviation. This is because the deviation for that point is large, and when you square it, it becomes even larger, heavily influencing the final result.

My calculator has two standard deviation buttons (σ and s). Which one do I use?

Use the $σ$ button (often labeled $σ_{n}$ or $σ_{x}$ ) when your data represents the entire population. Use the $s$ button (often labeled $s_{n} - 1$ or $s_{x}$ ) when your data is a sample of a larger population. The $s$ button uses the $n - 1$ formula.