A z-score is a standardized value that shows how far a data point is from the mean in terms of standard deviations.
Understanding z-scores in Research
In social science research, data can vary widely across studies, populations, and scales. To compare scores fairly and understand how unusual a data point is, researchers often use z-scores. A z-score transforms a raw score into a standard score, telling us how many standard deviations a value is above or below the mean.
This entry explains what a z-score is, how it’s calculated, and why it’s important in research. It also provides real-world examples from across the social sciences to show how z-scores are used in data analysis and interpretation.
What Is a z-score?
A Standardized Score
A z-score (also called a standard score) is a way to express where a particular score falls in a distribution. It tells you how many standard deviations a value is from the mean of a data set.
- A z-score of 0 means the score is exactly at the mean.
- A positive z-score means the score is above the mean.
- A negative z-score means the score is below the mean.
This makes it easier to understand whether a score is typical or unusual, regardless of the original unit of measurement.
The Formula
To calculate a z-score:
z = (X − M) ÷ SD
Where:
- X = the raw score
- M = the mean of the sample or population
- SD = the standard deviation
Example
Imagine a student scores 85 on a test. The mean score for the class is 75, and the standard deviation is 5.
z = (85 − 75) ÷ 5 = 2
This means the student scored 2 standard deviations above the mean.
Why z-scores Matter in Research
Allows Comparison Across Different Scales
Researchers often collect data using different scales. A z-score helps compare scores from different sources by putting them on the same standardized scale.
Example:
- A z-score of 1.5 on a reading test and a z-score of 1.5 on a math test indicate that a student performed equally well relative to peers on both.
Identifies Outliers
Z-scores help spot outliers, or values that are much higher or lower than the rest. A score with a z of +3 or −3 is often considered an outlier.
Spotting outliers is important because they can:
- Distort the mean
- Affect correlation and regression analysis
- Reveal interesting cases that need further study
Supports Statistical Testing
In inferential statistics, z-scores are used to calculate p-values, confidence intervals, and test statistics. They are a key part of many tests, including the z-test, which compares a sample to a population with a known standard deviation.
Helps Interpret Normal Distributions
If data follow a normal distribution, z-scores can help determine how common or rare a value is.
- About 68% of values fall between z = −1 and z = +1.
- About 95% fall between z = −2 and z = +2.
How z-scores Are Used in Social Science Research
Psychology
A psychologist might use z-scores to evaluate how a client’s anxiety score compares to a standardized sample.
Example: A z-score of +2.3 means the person’s anxiety is well above average.
Education
An education researcher may compare student performance across schools. Since grading systems differ, z-scores make it possible to compare students fairly.
Sociology
A sociologist analyzing income inequality may convert incomes to z-scores to measure how far each household is from the national average.
Political Science
In survey research, z-scores can show how extreme a respondent’s opinion is compared to the general population.
Criminal Justice
In crime trend studies, z-scores can be used to show how a city’s crime rate compares to national or state averages.
Interpreting z-scores
General Ranges
- z ≈ 0: Average
- z between ±1: Typical or common range
- z beyond ±2: Unusual or noteworthy
- z beyond ±3: Extreme or potential outlier
Positive and Negative Scores
Positive z-score: The value is above the mean.
Negative z-score: The value is below the mean.
Raw Score Conversion
You can also convert z-scores back to raw scores using:
X = (z × SD) + M
Best Practices for Using z-scores
Check for Normality
Z-scores are most meaningful when data follow a normal curve. If the data are skewed, z-scores may be misleading.
Use with Continuous Variables
Z-scores should be used with continuous variables like test scores, heights, or income—not with categorical variables.
Report Clearly
When using z-scores in reports or articles:
- State the mean and standard deviation used
- Explain what the z-score means in context
- Use graphs or standard score charts if helpful
Limitations of z-scores
- Not meaningful for non-normal distributions
- Can be affected by outliers
- Less helpful with small samples
Conclusion
The z-score is a vital tool in social science research for standardizing data, identifying unusual cases, and comparing values across different scales. It shows how far a score is from the mean in terms of standard deviations and plays a major role in both descriptive statistics and inferential statistics.
Whether you’re evaluating test results, modeling social behaviors, or analyzing trends, understanding z-scores helps improve clarity, accuracy, and comparability in your research.
Glossary Return to Doc's Research Glossary
Last Modified: 04/03/2025