Describing Distributions
Describe the shape, center, spread, and outliers of a distribution using SOCS.
Try the Interactive Version!
Learn step-by-step with practice exercises built right in.
Describing Distributions
SOCS Framework
Always describe a distribution using S–O–C–S:
- Shape
- Outliers
- Center
- Spread
Shape
Symmetry:
- Symmetric: left and right halves mirror each other (mean ≈ median)
- Skewed left (negatively skewed): tail extends left, peak right (mean < median)
- Skewed right (positively skewed): tail extends right, peak left (mean > median)
Modality:
- Unimodal: one peak
- Bimodal: two peaks (often two subpopulations)
- Multimodal: more than two peaks
- Uniform: roughly equal height across range
Peakedness:
- Roughly normal (bell curve)
- Flatter than normal (platykurtic)
- Sharper than normal (leptokurtic)
Outliers
Definition: observations unusually far from the rest
- Identify using boxplot (beyond whiskers using 1.5·IQR rule)
- Or context: "100 hours of TV watching" when most watch <20
Impact:
- Pull mean toward outlier (mean not resistant)
- Median unaffected (median is resistant)
- Increase standard deviation and range
Investigation: is it a genuine measurement, data entry error, or unusual case?
Center
Mean (\(\bar{x}\)): arithmetic average
- \(\bar{x} = \frac{\sum x_i}{n}\)
- Pulled by outliers (not resistant)
Median: middle value when ordered
- 50th percentile
- Resistant to outliers
- Preferred for skewed distributions
Mode: most frequent value
- Used for categorical or discrete data
Rule of thumb:
- Symmetric distribution: mean ≈ median
- Skewed distribution: prefer median
Spread
Range: max − min
- Simplest measure
- Affected by outliers
- Not resistant
Interquartile Range (IQR): \(Q3 - Q1\)
- Middle 50% of data
- Resistant to outliers
- Preferred for skewed distributions
Variance (\(s^2\)): average squared deviation from mean
- \(s^2 = \frac{\sum(x_i - \bar{x})^2}{n-1}\) (sample variance, divide by n−1)
Standard Deviation (\(s\)): square root of variance
- \(s = \sqrt{s^2}\)
- Same units as data
- Measures typical distance from mean
- Estimated: in roughly normal data, about 68% within 1s of mean
Worked Example
Data: Heights (inches) of 10 students: 62, 64, 65, 66, 67, 68, 69, 71, 72, 75
SOCS Description:
- Shape: roughly unimodal and symmetric (slight right skew due to 75)
- Outliers: boxplot Q1 ≈ 65.5, Q3 ≈ 70.5, IQR = 5; fences at 65.5 − 7.5 = 58 and 70.5 + 7.5 = 78. No outliers.
- Center: mean = \(\frac{62+64+...+75}{10} = 67.9\) inches; median = \(\frac{67+68}{2} = 67.5\) inches (very close, confirming near symmetry)
- Spread: range = 75 − 62 = 13 inches; IQR = 5 inches; \(s \approx 3.7\) inches
Comparison Language
When comparing two distributions:
Shape: "Distribution A is symmetric while Distribution B is right-skewed."
Center: "The median for Group X is approximately _____ inches, compared to _____ inches for Group Y, so Group X tends to be taller."
Spread: "Group X has an IQR of _____, while Group Y has IQR of _____, so Group Y is more variable."
Outliers: "Distribution A has one outlier at _____, while Distribution B has no outliers."
Common Mistakes
- Saying "mean = 50" when you haven't calculated it
- Forgetting to identify shape when asked to describe
- Confusing resistant vs. non-resistant (median is resistant; mean is not)
- Using mean and median interchangeably in skewed data
AP Exam Tip
On free response, examiners want to see you use SOCS explicitly. Write:
- "Shape: ..."
- "Outliers: ..."
- "Center: ..."
- "Spread: ..."
Use appropriate statistics for the shape (median/IQR for skewed; mean/std dev for symmetric).
📚 Practice Problems
1Problem 1easy
❓ Question:
A distribution shows most values clustered near 50, with a long tail extending to the right toward 100. Describe the shape and identify where the mean is relative to the median.
💡 Show Solution
This distribution is skewed to the right (positively skewed). When data is skewed right, the mean is pulled toward the tail (toward the higher values), so the mean > median. The long tail of extreme high values increases the average more than it affects the median. This is common in real-world data like incomes or test scores with a ceiling effect.
2Problem 2medium
❓ Question:
Two datasets have the same median (both 70) but different shapes: Dataset A is symmetric, while Dataset B is skewed left. Without seeing the distributions, explain what this tells you about their means.
💡 Show Solution
Dataset A (symmetric): Mean ≈ Median ≈ 70, because symmetry means the values balance equally on both sides.
Dataset B (skewed left): Mean < Median. The median is 70, but the mean is pulled toward the left tail by the extreme low values. The mean will be less than 70.
Why? In a left-skewed distribution, the tail extends toward lower values. These extreme low outliers pull the mean down more than they affect the median (which is just the middle position). This is common in age-at-death data or grade distributions where there's a floor but not a ceiling.
3Problem 3hard
❓ Question:
A histogram shows test scores for a large class with two distinct peaks: one at 70 and another at 85. Interpret this distribution and suggest what might explain it. How would you describe center and spread?
💡 Show Solution
Distribution characteristics:
This is a bimodal distribution with two peaks (modes) at 70 and 85. The presence of two modes suggests two distinct groups within the class, not a single homogeneous population.
Possible explanations:
- Different preparation levels (some students studied more thoroughly than others)
- One mode represents students who barely passed; the other represents strong performers
- The class might contain different ability levels mixed together
Center & Spread:
- A single measure of center (like the mean) would be misleading—it would fall around 77-78, not representing either peak well
- Better approach: Report both modes separately, or note that the distribution is bimodal
- Spread: The distance between the two peaks (15 points) is noteworthy; report the range and/or standard deviation, noting the gap suggests distinct subgroups
Lesson: Always look for multiple peaks; they suggest mixture of populations.
⚠️ Common Mistakes: Describing Distributions
Avoid these 3 frequent errors
Practice with Flashcards
Rate this topic's cards with spaced repetition. Cards join your deck when you finish a topic's lesson and take its exit quiz.
Browse All Topics
Explore more AP Statistics topics