Variance and standard deviation are measures of how spread out data values are around the . They help describe whether a dataset is tightly clustered or widely scattered. This matters in science, economics, testing, and engineering because two datasets can have the same average but very different variability.
Understanding spread gives a fuller picture than the mean alone.
Variance measures the average squared distance from the , while standard deviation is the square root of variance. Because standard deviation is in the same units as the original data, it is often easier to interpret. In a normal distribution, standard deviation also helps describe how much of the data lies near the mean.
These ideas are used to compare consistency, detect unusual values, and model uncertainty.
Understanding Standard Deviation
To calculate standard deviation by hand, begin by finding the mean. Then find how far each value is from that mean. Some differences are positive and some are negative, so adding the raw differences would always cancel to zero.
Squaring each difference solves this problem and gives larger gaps more weight. Add the squared differences, divide by the correct number of values, then take the square root.
The final square root brings the answer back into the original unit. If the data are heights in centimetres, the standard deviation is measured in centimetres, not square centimetres.
The divisor depends on what the data represent. Use the population method only when every member of the group has been measured. A school recording the heights of every student in one class has population data for that class.
More often, students work with a sample taken from a much larger group. In that case, divide by one less than the sample size. This adjustment is called Bessel's correction.
A sample mean is estimated from the same data, which makes the distances from that mean slightly too small on average. Dividing by one less corrects some of that underestimation when using a sample to learn about a population.
Standard deviation is useful when consistency matters. A factory may compare the masses of cereal boxes. Two production lines could have the same average mass, yet one line produces boxes close to the target while the other produces boxes that vary a lot.
The second line creates more underfilled or overfilled boxes. In sport, an athlete with a steady set of race times can have a lower standard deviation than an athlete with the same average time but more uneven performances. Weather forecasts, exam scores, medical measurements, investment returns, and experimental results all use spread to describe how reliable or variable results are.
Be careful with unusual values. Because distances are squared, one extreme result can raise the standard deviation greatly. This is sometimes important, such as when an extreme measurement signals a fault.
It can also make a dataset look less typical than most of its values really are. Inspect a graph or list of the data before trusting one summary number. The common rules about one, two, and three standard deviations work best for data shaped roughly like a symmetric bell.
Skewed data, such as incomes or waiting times, may not follow those percentages closely. For such data, the median and interquartile range can give a clearer picture alongside standard deviation.
Key Facts
- Population variance:
- Sample variance:
- Population standard deviation:
- Sample standard deviation:
- A larger standard deviation means data are more spread out from the mean.
- For a normal distribution, about 68% of values lie within standard deviation, 95% within , and 99.7% within .
Vocabulary
- Mean
- The mean is the average value found by adding all data values and dividing by the number of values.
- Variance
- Variance is the average squared distance of data values from the mean.
- Standard deviation
- Standard deviation is the and measures spread in the original units of the data.
- Population
- A population is the complete set of values or individuals being studied.
- Sample
- A sample is a smaller group taken from a population and used to estimate population characteristics.
Common Mistakes to Avoid
- Using instead of for sample variance, which gives a biased estimate of population spread when working from sample data.
- Forgetting to square the deviations before averaging, which is wrong because positive and negative deviations would cancel out.
- Confusing variance with standard deviation, which is wrong because variance is in squared units while standard deviation is in the original units.
- Assuming a larger mean always means a larger standard deviation, which is wrong because center and spread describe different features of a dataset.
Practice Questions
- 1 Find the population variance and population standard deviation of the data set .
- 2 A sample of quiz scores is . Find the sample mean, sample variance, and sample standard deviation.
- 3 Two classes have the same average test score of . Class A has a standard deviation of and Class B has a standard deviation of . Explain which class has scores that are more consistent and why.