Quick Answer
To compare distributions, describe their center, spread, shape, and any unusual features such as outliers, gaps, or clusters. Use the mean when a distribution is roughly symmetric with no strong outliers, and use the median when it is skewed or contains outliers. Every comparison should include actual numbers — not just "Group A did better" — and explain what those numbers mean.
How Do You Compare Two Distributions?
When you have two or more data sets, you almost never want to look at them in isolation — you want to know which group scored higher, which is more consistent, and whether the two distributions are actually similar. Comparing distributions means answering that question across four dimensions: center (where is the typical value?), spread (how variable is the data?), shape (is it symmetric or skewed?), and outliers (are there any unusual values, gaps, or clusters?).
A complete comparison always references real numbers. "Group B did better" is not a comparison. "Group B's median (62) is 17 points higher than Group A's (45)" is.
Compare the Center
Find the mean and/or median of each data set, state which one is higher, by how much, and explain what that means in context.
Use the mean when the distribution is roughly symmetric with no strong outliers — it uses every value in the data set.
Use the median when the distribution is skewed or contains outliers. The median is resistant to extreme values, so a single unusually high or low number won't distort it.
Compare the Spread
Spread describes how variable or consistent the data is. Three common measures:
• Range = maximum − minimum. Easy to calculate, but affected heavily by outliers.
• IQR = Q3 − Q1. Measures the spread of the middle 50% of the data and is resistant to outliers.
• MAD (Mean Absolute Deviation) = the average distance of each data value from the mean. Because MAD is based on the mean, it is still affected by outliers — just usually less dramatically than the range is.
A smaller IQR or MAD means the data is more tightly clustered around the center (more consistent). A larger one means more variability.
Compare the Shape
Describe each distribution as symmetric, left-skewed, or right-skewed. Skew direction refers to the direction of the longer tail, not the taller side of the graph.
• Symmetric: left and right sides mirror each other; mean ≈ median.
• Right-skewed: a longer tail on the right; mean > median.
• Left-skewed: a longer tail on the left; mean < median.
Check for Outliers, Gaps, and Clusters
An outlier is a value that falls below Q1 − 1.5(IQR) or above Q3 + 1.5(IQR). A cluster is a group of values bunched close together; a gap is a stretch of the number line with no data at all.
Outliers matter because they can pull the mean and inflate the range while leaving the median and IQR almost unchanged — which is exactly why median and IQR are preferred for skewed data or data with outliers.
Same Center, Different Spread
This is the single most important idea in this article: two distributions can share the exact same center and still look nothing alike.
Example: Set A has a median of 50 and an IQR of 20. Set B has a median of 50 — the same typical value — but an IQR of only 6. Set B's data is packed tightly around 50; Set A's is spread widely above and below it. If you only compared medians, you'd wrongly conclude the two sets are "the same." They aren't. Same center does not mean same distribution — you always have to check spread too.
Same Center, Different Spread — Dot Plot Comparison
Set A (Median = 50, IQR = 20 — widely spread)
Set B (Median = 50, IQR = 6 — tightly clustered)
Set A
Median = 50 · IQR = 20
Dots spread widely from 30 to 70
Set B
Median = 50 · IQR = 6
Dots packed tightly around 50
Comparing Box Plots
Compare the median (center), the box length or IQR (spread of the middle 50%), and the whiskers (overall spread and possible outliers). A median sitting off-center in the box, or one whisker noticeably longer than the other, may suggest a skewed shape — a box plot alone doesn't prove skewness, since it only summarizes five numbers, not every data point.
Box Plot Comparison — Group A vs. Group B
Group A (Median = 45, IQR = 26)
Group B (Median = 62, IQR = 18)
Group A
Median = 45 · IQR = 26 · Range = 52
Median centered in box → roughly symmetric
Group B
Median = 62 · IQR = 18 · Range = 55
Median near Q3 → suggests left-skew
Comparing Histograms
Compare where the bars are tallest (the modal interval, not automatically "the center"), where most of the data is concentrated, the overall spread of the bars, and the shape. Describe any gaps (empty intervals) or unusual clusters. Two histograms can cover the exact same range of values while being concentrated in completely different parts of that range.
Histogram Comparison — Symmetric vs. Left-Skewed
Distribution A
Roughly symmetric
Distribution B
Left-skewed (tail on the left)
Distribution A — Symmetric
Peak at 10–14; counts taper evenly on both sides (2, 5 | 11 | 5, 2). Mean ≈ median.
Distribution B — Left-Skewed
Counts increase toward the right; tail stretches left. Mean < median.
Comparing Dot Plots
Look at where the dots cluster, how far the extreme values are from the rest of the data, and whether the medians match. Two dot plots can have identical medians while one is tightly bunched and the other is spread from end to end — another version of "same center, different spread."
Quick Summary
• Always compare both center and spread — never just one.
• For roughly symmetric data without strong outliers, the mean and MAD can be useful. For skewed data or data with outliers, the median and IQR are usually more appropriate.
• Same mean or same median never guarantees the same distribution.
• A histogram's tallest bar is a modal interval, not automatically "the center."
• Box plot shape observations suggest skew; they don't prove it.
• Always state your comparison with real numbers, in context.
Worked Examples
Example 1
Example 1 — Two Box Plots: Group A: Min=18, Q1=32, Median=45, Q3=58, Max=70. Group B: Min=25, Q1=50, Median=62, Q3=68, Max=80.
Solution
- Center: Median A = 45; Median B = 62. Group B's typical score is 17 points higher.
- Spread: IQR A = 58 − 32 = 26. IQR B = 68 − 50 = 18. Group A has more spread in the middle 50%.
- Shape: In Group A, the median sits exactly halfway between Q1 and Q3 (13 and 13) — roughly symmetric. In Group B, the median (62) is closer to Q3 (68) than to Q1 (50) — distance 6 vs. 12 — which suggests a left-skew, and the longer lower whisker (Q1−Min = 25 vs. Q3−Max = 12) supports that.
- Outliers: No outliers are shown in either box plot.
Answer: Group B has a higher median (62 vs. 45) and less spread in the middle 50% (IQR 18 vs. 26). Group A appears roughly symmetric; Group B's shape suggests a left-skew.
Example 2
Example 2 — Two Histograms: Daily high temperatures (°F) over 30 days. City A: 60–64: 3, 65–69: 6, 70–74: 10, 75–79: 7, 80–84: 4 (n=30). City B: 60–64: 2, 65–69: 3, 70–74: 5, 75–79: 9, 80–84: 11 (n=30).
Solution
- Concentration: City A — 65–79 holds 6+10+7 = 23 of 30 days (76.7%) — concentrated in the middle of the range. City B — 75–84 holds 9+11 = 20 of 30 days (66.7%) — concentrated at the high end.
- Modal interval: City A peaks at 70–74; City B peaks at 80–84.
- Shape: City A's counts taper off fairly evenly on both sides of its peak (3, 6 | 10 | 7, 4) — roughly symmetric. City B's counts increase toward the highest interval with a tail stretching down to 60–64 — suggests a left-skew.
- Spread: Both cover the same overall interval range (60–84°F), but the values are distributed very differently within it.
Answer: City B is concentrated more heavily at the warmer end of the distribution, with 20 of 30 days (66.7%) between 75°F and 84°F. City A is concentrated more toward the middle, with a modal interval of 70–74°F and a roughly symmetric shape. Both distributions cover the same overall interval range, but City B is left-skewed while City A is roughly symmetric.
Example 3
Example 3 — Two Dot Plots, Same Median: Problems solved per day over 9 days. Student A: 4, 5, 5, 6, 6, 6, 7, 7, 8. Student B: 1, 2, 4, 6, 6, 6, 8, 10, 11.
Solution
- Center: Both ordered lists have 6 as the 5th (middle) value — median = 6 for both.
- Spread: Range A = 8 − 4 = 4. Range B = 11 − 1 = 10. IQR A: Q1 = 5, Q3 = 7, IQR = 2. IQR B: Q1 = 3, Q3 = 9, IQR = 6.
- Shape/clusters: Student A's values cluster tightly around 6. Student B also clusters at 6, but with two extreme values (1 and 11) stretching the data much wider.
Answer: Same median (6) for both students, but Student A (range 4, IQR 2) is far more consistent than Student B (range 10, IQR 6) — a direct example of same center, different spread.
Example 4
Example 4 — Same Mean, Very Different Distributions: Set P: 18, 20, 22, 24, 26, 28, 30, 32 (n=8). Set Q: 10, 15, 18, 20, 22, 25, 28, 62 (n=8).
Solution
- Center: Sum P = 200, mean = 25; median = (24+26)/2 = 25 — mean equals median. Sum Q = 200, mean = 25; median = (20+22)/2 = 21 — mean is noticeably higher than the median.
- Spread: IQR P: Q1 = 21, Q3 = 29, IQR = 8. Range P = 14. IQR Q: Q1 = 16.5, Q3 = 26.5, IQR = 10. Range Q = 52.
- Outliers: Upper fence for Q = 26.5 + 1.5(10) = 41.5. Since 62 > 41.5, it is a statistical outlier.
- Shape: Set P is symmetric (mean = median). Set Q is right-skewed (mean 25 > median 21), pulled upward by the outlier at 62.
Answer: Both sets share the same mean (25), but Set P is symmetric with a compact spread (IQR 8, range 14), while Set Q is right-skewed with a much wider spread (IQR 10, range 52) driven by an outlier at 62. Same mean does not mean same distribution.
Common Mistakes
Comparing only the means. Two data sets can share a mean and still have very different spreads or shapes — always check spread too.
Ignoring spread entirely. A comparison that only mentions center is incomplete.
Calling a distribution "better" without context. Whether a higher or lower value is good depends on what's being measured — state what the numbers represent.
Giving a conclusion with no numbers. "Group A is more consistent" means nothing without an IQR, range, or MAD to back it up.
Confusing the direction of skew. Right-skewed means the tail points right (mean > median); left-skewed means the tail points left (mean < median) — it's not about which side has taller bars.
Assuming a smaller range means a smaller IQR. Range only looks at the two extreme values; a data set can have a large range but a small IQR if most values are clustered in the middle.
Assuming the same mean or median means the same distribution. Matching centers can hide very different spreads and shapes.
Practice Problems
Set A: Min=15, Q1=28, Median=40, Q3=52, Max=65. Set B: Min=20, Q1=45, Median=58, Q3=66, Max=75. Compare the center and spread of both sets.
Hint: Find the IQR (Q3 − Q1) for each, then compare medians.
Data Set M: 12, 15, 18, 20, 23, 25, 28, 30. Data Set N: 5, 8, 12, 20, 25, 30, 35, 40. Compare the median and IQR of each.
Hint: Order the data (already done here), then find Q1 and Q3 for each half.
Histogram X (n=25): 0–4: 2, 5–9: 4, 10–14: 11, 15–19: 6, 20–24: 2. Histogram Y (n=25): 0–4: 9, 5–9: 8, 10–14: 5, 15–19: 2, 20–24: 1. Which is roughly symmetric? Which is skewed, and in which direction?
Hint: Find the modal interval for each, then check how the counts taper off on either side.
Group R: 30, 32, 34, 36, 38. Group S: 10, 20, 34, 40, 66. A classmate says these two groups have "identical" distributions because they share the same mean and median. Explain why that's incomplete.
Hint: Both groups do share a mean and median of 34 — but check the IQR or range before agreeing they're identical.
Box Plot C: Median=55, IQR=10. Box Plot D: Median=55, IQR=28. Write a full 2–3 sentence comparison.
Hint: Same median means same center — the real difference is in the spread.
Practice Problem Answers
Frequently Asked Questions
Do two distributions with the same mean have the same shape?
No. Two data sets can share a mean while one is symmetric and the other is skewed with an outlier (see Example 4). Always check the median, spread, and shape before assuming they're alike.
What is the difference between range and IQR?
Range is the maximum minus the minimum, so a single extreme value can inflate it. IQR is Q3 − Q1, the spread of the middle 50% of the data, and is resistant to outliers.
When should I use the median instead of the mean?
Use the median when the distribution is skewed or contains outliers — it isn't pulled around by extreme values the way the mean is.
What does a wider box in a box plot mean?
A wider box means a larger IQR — the middle 50% of the data is more spread out. It does not by itself tell you how many data points there are.
Can two distributions have the same median but different spreads?
Yes — this is one of the most common situations in statistics (see Example 3 and Practice Problem 5). Same median never guarantees the same spread.
How do outliers affect the mean and range?
An outlier can pull the mean toward itself and dramatically increase the range, while the median and IQR usually stay nearly unchanged — which is why median/IQR are preferred when outliers are present.
Is comparing distributions the same as SOCS?
Some textbooks use the mnemonic SOCS for Shape, Outliers, Center, and Spread. StudyBAP teaches the same core ideas using the comparison sequence center → spread → shape → outliers.
Related StudyBAP Lessons
Algebra 1 · Unit 9 · Chapter 1
Measures of Center
Algebra 1 · Unit 9 · Chapter 2
Measures of Spread
Algebra 1 · Unit 9 · Chapter 3
Dot Plots & Histograms
Algebra 1 · Unit 9 · Chapter 4
Box Plots & Five-Number Summary
Algebra 1 · Unit 9 · Chapter 5
Comparing Distributions
Algebra 1 · Unit 9 · Unit Review
Unit 9 Review
Algebra 1 · Unit 9 · Unit Cover
Unit 9 Cover
Still have a question?
Submit your Algebra question here. The best questions are answered every week on StudyBAP.
Ask a Question