Standard deviation is a number that tells you how spread out a set of scores is around the average. In psychology, it appears constantly: in research papers, clinical assessments, and standardized tests like IQ scores. If the standard deviation is small, most people scored close to the average. If it’s large, scores were scattered widely. Think of it as measuring how far the typical person lands from the group’s mean.
What Standard Deviation Actually Tells You
The mean gives you the center of a dataset. Standard deviation gives you the shape around that center. Conceptually, it answers a simple question: how far from the average is the average person? A small standard deviation means participants responded similarly. A large one means their responses varied a lot.
Imagine you give an anxiety questionnaire to 200 college students. The mean score is 50, and the standard deviation is 5. That tells you most students scored fairly close to 50. Now imagine a different sample where the mean is also 50 but the standard deviation is 20. Same average, completely different picture. In the second group, some students reported very low anxiety and others reported very high anxiety. The mean alone would make these two groups look identical. The standard deviation reveals they’re not.
The 68-95-99.7 Rule
Many psychological traits, including IQ, personality scores, and reaction times, follow a bell-shaped curve called a normal distribution. When data is normally distributed, standard deviation carves the curve into predictable slices:
- 68% of scores fall within one standard deviation of the mean.
- 95% of scores fall within two standard deviations.
- 99.7% of scores fall within three standard deviations.
This pattern is sometimes called the empirical rule, and it makes standard deviation far more powerful than just a number on a page. It lets you instantly estimate how unusual any given score is.
IQ Scores: The Classic Example
IQ tests are built around a mean of 100 and a standard deviation of 15. That single fact lets you map the entire distribution. One standard deviation above the mean is an IQ of 115, and one below is 85. About 68% of the population falls in that 85-to-115 range.
Go two standard deviations out, and you reach IQ 70 on the low end and IQ 130 on the high end. Roughly 95% of people score within this band. Only about 2.14% score above 130, and another 2.14% score below 70. At three standard deviations (IQ 145 or above, IQ 55 or below), you’re looking at just 0.13% of the population on each side.
This is why standard deviation matters so much in clinical psychology. An IQ of 70, exactly two standard deviations below the mean, is one of the criteria used to identify intellectual disability. An IQ of 130, two standard deviations above, is a common threshold for giftedness. The cutoff isn’t arbitrary. It’s defined by where someone falls relative to the standard deviation of the population.
Clinical Thresholds and Diagnosis
Psychologists routinely use standard deviation to decide whether a test score is “normal” or clinically meaningful. A common benchmark is two standard deviations from the mean. If your score on a neuropsychological test falls two or more standard deviations below average, that’s typically considered impaired. If a therapy patient’s symptoms drop to within two standard deviations of a healthy population’s mean, that shift is often considered clinically significant recovery.
This two-standard-deviation convention appears across psychological assessment. It provides a consistent, mathematically grounded way to distinguish typical variation from something that warrants attention, whether that’s a learning disability, cognitive decline after brain injury, or improvement during treatment.
Z-Scores: Standard Deviation as a Unit
One of the most common uses of standard deviation in psychology is converting raw scores into z-scores. The formula is straightforward: take a person’s score, subtract the group mean, and divide by the standard deviation. The result tells you how many standard deviations that person sits above or below average.
A z-score of 0 means you scored exactly at the mean. A z-score of +1.5 means you scored one and a half standard deviations above average. A z-score of -2 means two standard deviations below. This conversion puts different tests on the same scale. You can directly compare someone’s performance on a memory test (scored 0 to 50) with their performance on an attention test (scored 0 to 200) by converting both to z-scores. Without standard deviation, that comparison wouldn’t be possible.
Measuring How Big an Effect Really Is
When psychologists run experiments, statistical significance alone doesn’t tell the full story. A therapy might produce a statistically significant improvement, but is the improvement meaningful? Standard deviation helps answer this through a measure called effect size.
The most common version in psychology, Cohen’s d, is calculated by dividing the difference between two group means by their pooled standard deviation. The result tells you how large the difference is in standardized units. Cohen’s guidelines, drawn from decades of behavioral research, classify a d of 0.2 as a small effect, 0.5 as medium, and 0.8 as large. A d of 0.8 means the two groups differ by eight-tenths of a standard deviation, a substantial gap. Without standard deviation as the yardstick, researchers would have no standardized way to communicate the practical size of their findings.
Standard Deviation vs. Standard Error
These two terms look similar but answer different questions. Standard deviation describes the spread of individual scores in your sample. Standard error describes how precisely you’ve estimated the group mean. Standard deviation tells you about people. Standard error tells you about your measurement of the average.
In a psychology paper, you’ll see both. When authors report descriptive statistics (the average anxiety score of participants, for instance), they pair the mean with a standard deviation so readers can picture how varied the group was. When authors report inferential statistics (testing whether two groups differ), they use standard error to build confidence intervals around the mean. Confusing the two is one of the most common mistakes in interpreting research, so it’s worth keeping the distinction clear: standard deviation is about individual variability, standard error is about the precision of the average.
How It Appears in Research Papers
In APA format, which is the standard writing style for psychology, standard deviation is abbreviated as SD and does not need to be defined for the reader. It’s typically reported alongside the mean (abbreviated M) in parentheses, like this: participants scored higher on the well-being scale (M = 72.3, SD = 8.1). For survey and questionnaire data, both the mean and standard deviation are reported to one decimal place. For other types of data, two decimal places are standard.
When you see these numbers in a study, you can immediately start interpreting them. A reported SD of 8.1 with a mean of 72.3 tells you that roughly two-thirds of participants scored between about 64 and 80. If another study reports the same mean but an SD of 3.0, that group was much more homogeneous, with most participants clustered between 69 and 75. The SD gives you a mental picture of the sample that the mean alone never could.