Standard deviation measures how spread out a dataset is from its mean. This cheat sheet gives students a clear by-hand workflow for small datasets, where every calculation can be shown in a table. It is useful for checking calculator results, understanding variability, and comparing data sets in statistics class.
The main steps are to find the mean, subtract the mean from each value, square each deviation, add the squared deviations, divide by the correct count, and take the square root. For a population, use . For a sample, use , which uses because samples estimate population spread.
Key Facts
- The mean of a population is , and the mean of a sample is .
- A deviation is the distance from a value to the mean, written as for a population or for a sample.
- The sum of deviations from the mean is always , except for small rounding errors.
- Population variance is .
- Sample variance is .
- Population standard deviation is .
- Sample standard deviation is .
- Standard deviation is in the original units of the data, while variance is in squared units.
Vocabulary
- Mean
- The mean is the average value found by dividing the sum of the data values by the number of values.
- Deviation
- A deviation is the difference between a data value and the mean, such as .
- Variance
- Variance is the average of the squared deviations from the mean, using for a population or for a sample.
- Standard Deviation
- Standard deviation is the square root of variance and describes the typical distance of data values from the mean.
- Population
- A population is the entire group of data values being studied, with size .
- Sample
- A sample is a smaller group chosen from a population, with size , used to estimate information about the population.
Common Mistakes to Avoid
- Dividing by for a sample, which is wrong because sample standard deviation uses to better estimate population spread.
- Forgetting to square the deviations, which is wrong because positive and negative deviations would cancel out before measuring spread.
- Stopping at variance, which is wrong if the question asks for standard deviation because you must take or .
- Rounding too early, which can make the final standard deviation inaccurate because small errors build up across several steps.
- Mixing population and sample notation, which is wrong because and describe a population while and describe a sample.
Practice Questions
- 1 Find the population standard deviation for the dataset .
- 2 Find the sample standard deviation for the dataset .
- 3 A dataset has and . Find the sample variance and sample standard deviation.
- 4 Two classes have the same mean test score, but Class A has a larger standard deviation than Class B. Explain what this means about the scores in the two classes.
Understanding Standard Deviation by Hand (Step-by-Step)
Squaring the deviations does more than remove negative signs. It gives larger gaps much more weight. A score that is four units from the mean contributes sixteen squared units, while a score that is two units away contributes only four.
This makes standard deviation sensitive to unusual values. One very high or very low measurement can change it a lot.
That sensitivity is useful when an outlier represents a real event, such as an unusually long commute or an extreme test score. It can be misleading when the unusual value came from a recording mistake.
The choice between population and sample depends on where the data came from. A population calculation is appropriate when every member of the group of interest is included. For example, a teacher can treat scores from every student in one class as a population when describing that class.
A sample calculation is appropriate when only part of a larger group was measured. Surveying thirty students to estimate the sleep habits of an entire school creates a sample.
Dividing by one less than the sample size makes the estimated spread a little larger. This correction helps offset the fact that a sample usually looks less spread out than the full population it represents.
Standard deviation has meaning only when it is connected to the center and shape of the data. A small value means the observations tend to stay near the mean. A large value means they are more scattered.
It does not tell which direction values are scattered, because values above and below the mean contribute equally after squaring. It does not identify the typical value by itself either. Two groups can have the same mean but very different consistency.
In sports, two players may have the same average points per game. The player with the smaller standard deviation scores more consistently from game to game.
For data that form a roughly bell shaped pattern, standard deviation gives a useful guide to what is usual. Many observations lie within one standard deviation of the mean. Nearly all lie within two or three standard deviations when the pattern is close to bell shaped.
This rule should not be forced onto every dataset. Data such as house prices, incomes, or social media follower counts are often strongly skewed.
A few huge values can pull the mean upward and inflate the standard deviation. In such cases, the median and interquartile range may describe a typical person or item more fairly.
When calculating by hand, keep more decimal places than you plan to report. Rounding the mean too early changes every deviation, so the later squared values can drift from the correct result. A table helps reveal mistakes.
The signed deviations should balance to zero before squaring, apart from tiny rounding differences. Every squared deviation must be zero or positive.
The final standard deviation cannot be negative and should use the same unit as the original measurements, such as seconds, centimeters, or dollars. Label whether the result is for a sample or population, since the same list of values produces two slightly different answers.