Comparing Distributions

When statisticians want to understand data, they rarely look at just one group in isolation. Comparing distributions means examining two or more sets of data side by side to understand how they differ or how they are alike. For example, a researcher might compare the test scores of students taught by two different methods, or a sports analyst might compare the heights of basketball players versus soccer players. To do this rigorously, we examine three key characteristics of each distribution: its center (a typical value), its spread (how much the data varies), and its shape (the pattern of the data, including symmetry and outliers). Comparing distributions is one of the most powerful tools in exploratory data analysis. It allows us to ask meaningful questions: Does one group tend to score higher than another? Is one group more consistent while another is highly variable? Are there unusual observations pulling the data in one direction? The answers can reveal patterns that inform real decisions in medicine, education, sports, economics, and public policy. Comparing distributions is not just about computing numbers — it requires careful interpretation and clear communication. The most common visual tools for comparing distributions are side-by-side boxplots, back-to-back stem-and-leaf plots, and overlapping histograms or dotplots. Each tool highlights different aspects of the data. Alongside these visuals, numerical summaries — including the mean, median, standard deviation, interquartile range (IQR), and five-number summary — provide precise language for describing and comparing distributions. A complete comparison always addresses center, spread, and shape, and is written in context.

Key takeaways
  • Compare the Centers
  • Compare the Spreads
  • Compare the Shapes
  • Write a Complete Comparison in Context

Compare the Centers

The center of a distribution describes a typical value. The two main measures are the median and the mean. The median is the middle value when data is ordered, and it is resistant to outliers. The mean is the arithmetic average. When comparing two groups, state which group has the higher center and by how much. For example: 'Group A has a median of 78 and Group B has a median of 65, so Group A tends to score about 13 points higher.' Always refer to the context — what do those numbers actually represent?

Compare the Spreads

Spread describes how variable or consistent the data is. The two main measures are the standard deviation (s) and the interquartile range (IQR). The IQR = Q3 − Q1, the range of the middle 50% of data, and is resistant to outliers. The standard deviation measures average distance from the mean. A larger spread means the data is more variable; a smaller spread means data points cluster close to the center. For example: 'Group A has an IQR of 12 while Group B has an IQR of 28, meaning Group B's scores are much more spread out.' Larger spread can indicate inconsistency or a more diverse group.

Compare the Shapes

Shape describes the overall pattern of the distribution. Key features include: symmetry (does the data look balanced around the center?), skewness (is there a long tail to the left or right?), number of modes (unimodal, bimodal?), and outliers (are there data points far from the rest?). A right-skewed (positively skewed) distribution has a tail extending to higher values; a left-skewed (negatively skewed) distribution has a tail toward lower values. When a distribution is skewed, the mean is pulled toward the tail while the median stays closer to the bulk of the data. Always note any outliers, as they can strongly influence the mean and standard deviation.

Write a Complete Comparison in Context

A strong statistical comparison ties all three elements together in sentences that reference the actual context. Avoid vague statements like 'Group A is bigger.' Instead write: 'Group A (median = 78, IQR = 12) has a higher and more consistent set of scores than Group B (median = 65, IQR = 28). Both distributions appear roughly symmetric with no extreme outliers, but Group B shows greater variability, suggesting more diverse performance among its students.' This kind of paragraph-form answer is expected on AP Statistics free-response questions.

The Key Equation

IQR=Q3−Q1\text{IQR} = Q_3 - Q_1
  • Q_1 — First Quartile (same as data): The median of the lower half of the data; 25% of values fall below this point.
  • Q_3 — Third Quartile (same as data): The median of the upper half of the data; 75% of values fall below this point.
  • \text{IQR} — Interquartile Range (same as data): Measures the spread of the middle 50% of the data. Resistant to outliers, making it preferred over range when outliers are present.

Worked Example

Problem. Two classes took the same math exam (out of 100 points). Class A scores: 55, 62, 68, 70, 72, 74, 78, 80, 82, 88. Class B scores: 40, 55, 60, 70, 72, 74, 80, 85, 90, 98. Compare the distributions of the two classes by computing and comparing the median, IQR, and describing shape.

MedianA=MedianB=73,IQRA=12,IQRB=25\text{Median}_A = \text{Median}_B = 73, \quad \text{IQR}_A = 12, \quad \text{IQR}_B = 25

Answer. Median_A = Median_B = 73; IQR_A = 12; IQR_B = 25. Class B is more variable despite having the same center.

Figure
Diagram
Side-by-Side Boxplots: Class A vs Class B405565758595100Score (out of 100)A55Q1=68Med=73Q3=8088B40Q1=60Med=73Q3=8598
Origins · 1 minJohn Tukey and how this was discovered
JT
John Tukey
1915–2000 · American

John Tukey was a pioneering statistician who invented the boxplot (also called a box-and-whisker plot) in 1970 as part of his groundbreaking work on Exploratory Data Analysis (EDA). He developed systematic methods for visually summarizing data distributions, making it far easier to compare multiple groups. His 1977 book 'Exploratory Data Analysis' transformed how statisticians examine and present data, emphasizing visual tools alongside formal calculations. Tukey also coined the word 'bit' (short for binary digit) and is credited with early use of the term 'software' in a mathematical context — he had an extraordinary influence far beyond statistics.

  • Measures of Center — Understanding mean and median is essential before comparing the centers of two distributions.
  • Measures of Spread — IQR and standard deviation are the primary tools used to compare variability across groups.
  • Boxplots and Five-Number Summary — Side-by-side boxplots are the most common visualization tool for comparing distributions.
  • Shape of Distributions — Identifying skewness, symmetry, and outliers is a core component of comparing distributions.
  • Outliers — Outliers can dramatically shift the mean and standard deviation, affecting how distributions compare.