Common Statistical Symbols Reference
A lookup table covering the symbols you will encounter most often in research statistics. Symbols are shown in their proper mathematical typeset. Italicized symbols in APA papers are marked with a note.
Descriptive Statistics
| Symbol | Name | Meaning / When Used |
|---|---|---|
| $$M$$ | Sample mean | The arithmetic average of a sample. APA uses M (italic). Some textbooks write \(\bar{X}\). |
| $$\mu$$ | Population mean | The true mean of the entire population. Greek letters denote population parameters. |
| $$SD$$ | Sample standard deviation | Average distance of scores from the sample mean. APA uses SD (italic). Some texts write s. |
| $$\sigma$$ | Population standard deviation | The true standard deviation of the population. \(\sigma^2\) is the population variance. |
| $$SE$$ | Standard error (of the mean) | The standard deviation of the sampling distribution. \(SE = SD / \sqrt{n}\). Used to build confidence intervals. |
| $$n$$ | Sample size (subgroup) | Number of participants in a subgroup or condition. Lowercase n. |
| $$N$$ | Total sample size | Total number of participants across all groups. Uppercase N. |
| $$\Sigma$$ | Summation | "Sum of." \(\Sigma X\) means add up all values of X. Appears in nearly every statistics formula. |
Inferential Test Statistics
| Symbol | Name | Meaning / When Used |
|---|---|---|
| $$t$$ | t statistic | Test statistic for t-tests. Measures how many standard errors the sample mean is from the null. Reported as t(df) = value. |
| $$F$$ | F statistic | Test statistic for ANOVA and regression. Ratio of between-group variance to within-group variance. Reported as F(df1, df2) = value. |
| $$\chi^2$$ | Chi-square | Test statistic for chi-square tests (independence, goodness of fit). Compares observed to expected frequencies. Reported as \(\chi^2\)(df, N = value) = value. |
| $$df$$ | Degrees of freedom | Number of values free to vary. Depends on the test: t-test df = N − 2; ANOVA has dfbetween and dfwithin. |
| $$p$$ | p-value | Probability of obtaining results at least as extreme as observed, assuming the null hypothesis is true. p < .05 is the conventional significance threshold. |
| $$CI$$ | Confidence interval | Range of plausible values for a parameter. A 95% CI means: if we repeated the study many times, 95% of intervals would contain the true value. |
Hypothesis Testing & Error
| Symbol | Name | Meaning / When Used |
|---|---|---|
| $$H_0$$ | Null hypothesis | States there is no effect or no difference. The hypothesis you are trying to reject. E.g., \(H_0\!: \mu_1 = \mu_2\). |
| $$H_1$$ | Alternative hypothesis | States there is an effect or a difference. Also written \(H_a\). E.g., \(H_1\!: \mu_1 \neq \mu_2\). |
| $$\alpha$$ | Alpha (significance level) | The threshold for rejecting \(H_0\). Conventionally set at .05. Also the probability of a Type I error (false positive). |
| $$\beta$$ | Beta (Type II error rate) | Probability of failing to reject \(H_0\) when it is false (false negative). Statistical power = \(1 - \beta\). Not the same as regression \(\beta\). |
Effect Sizes & Relationships
| Symbol | Name | Meaning / When Used |
|---|---|---|
| $$d$$ | Cohen's d | Standardized mean difference. The most common effect size for t-tests. Small = 0.2, medium = 0.5, large = 0.8. |
| $$r$$ | Pearson correlation coefficient | Measures the strength and direction of a linear relationship between two continuous variables. Ranges from −1 to +1. |
| $$R^2$$ | Coefficient of determination | Proportion of variance in the DV explained by the model. Ranges from 0 to 1. Used in regression. |
| $$\eta^2$$ | Eta-squared | Proportion of total variance explained by the IV. Effect size for ANOVA. Small = .01, medium = .06, large = .14. |
Quick Memory Aids
- Roman letters = sample statistics (M, SD, s, r, n). Greek letters = population parameters (\(\mu\), \(\sigma\), \(\rho\), \(\beta\)).
- Lowercase n = subgroup size. Uppercase N = total sample.
- \(\alpha\) is about Type I error (false alarm). \(\beta\) is about Type II error (miss). Power = \(1 - \beta\).
- p-value: how surprised you should be if the null is true. Small p = surprising = reject \(H_0\).
- r vs. R²: r is direction + strength; R² is variance explained. \(R^2 = r^2\) in simple regression.