1.6 Describing the Distribution of a Quantitative Variable

SOCS: The Four Things to Describe

For any quantitative distribution, describe Shape, Outliers, Center, and Spread (SOCS) in context. Shape describes the overall pattern (symmetric, skewed, unimodal, etc.), outliers are unusual values, center is a typical value (mean/median), and spread describes variability (range, IQR, standard deviation). Always relate your description back to the variable and units given in the problem.

Describing Shape

Shape can be symmetric, skewed right (long tail toward high values), or skewed left (long tail toward low values). Also note if it's unimodal, bimodal, or uniform. Shape helps decide which center/spread measures are appropriate: symmetric data pairs with mean & standard deviation, while skewed data or outliers pair better with median & IQR since these are resistant to extreme values.

Comparing Distributions

When comparing two or more distributions, never describe them separately — use comparative language like higher, lower, more variable, or similarly shaped. Compare all four SOCS components: e.g., 'Distribution A is skewed right while B is roughly symmetric; A has a higher median and larger IQR than B.' Context and units must appear in every comparison statement.

Key terms

SOCS

Shape, Outliers, Center, Spread — the four things to describe in a quantitative distribution.

Skewed right

A distribution with a long tail extending toward higher values.

Skewed left

A distribution with a long tail extending toward lower values.

Resistant statistic

A measure (like median or IQR) that is not strongly affected by outliers or skewness.

Comparing distributions

Must use comparative language (higher, lower, more variable) and address shape, outliers, center, and spread in context.

Try an AP-style question

Which four features should you describe for a quantitative distribution?

1. Mean, median, mode, and range
2. Slope, intercept, r, and residuals
3. Shape, outliers, center, and spread
4. Categories, counts, percents, and totals