Reading data displays.
Each chart type is built for a different question. The real skill is matching the chart to the question — and spotting the misleading ones — not memorizing each shape.
Every newspaper, sports broadcast, and quarterly earnings report leans on charts. Some of them are honest summaries of the data. Others are designed to mislead, accidentally or otherwise. The difference between a clear-eyed reader and a confused one is mostly the ability to look at a chart and ask: what question is this picture actually answering?
Four chart types cover the vast majority of what we see in the wild: bar charts compare categories side by side, histograms show how a single numeric variable is distributed, box plots pack the five-number summary (min, Q1, median, Q3, max) into one compact shape, and pie charts show parts of a whole. Each one is built for a particular question, and a careful reader knows which is which on sight.
By the end of this lesson, we will match any of these four charts to the question it best answers, read off central values and spread directly from the picture, and spot the most common misleading-chart traps (the pie chart that totals more than 100%; the bar chart with a stretched y-axis that makes a tiny gap look huge).
Four charts, four questions.
Histogram. Looks like a bar chart with no gaps. The x-axis is numeric, sliced into bins. Each bar's height is the count of values that fall inside that bin. Question answered: how is this numeric variable distributed? Histograms reveal shape — bell-curved, skewed, bimodal — that summary statistics alone cannot.
Box plot (box-and-whisker). A horizontal box from Q1 to Q3, a vertical line inside the box marking the median, and "whiskers" running out to the min and max. Question answered: where is the center, how spread out is the data, and is it symmetric or skewed? Five numbers, one shape.
Pie chart. Slices of a circle, each slice's angle proportional to a category's share. The slices must total 100%. Question answered: what fraction of the whole does each category represent? Best for 2-5 slices; with more slices, switch to a bar chart.
Four chart types, each answering a different question. Bar charts compare categories (gaps between bars); histograms show numeric distributions (no gaps); box plots pack the five-number summary into one compact shape; pie charts show parts of a whole.
A data display is a visual summary of a data set. The shape of the display — bars, dots, slices, boxes — is chosen to match the question being asked. Picking the wrong shape misleads, even when the numbers are right.
Words you'll see on every chart
- Category (bar chart) A discrete label on the x-axis: "Monday," "Coffee," "Canada." Bar charts arrange one bar per category, with visible gaps between bars to signal that the x-axis is not numeric.
- Bin (histogram) An interval on a numeric x-axis. Values in the data set are tallied into the bin they fall inside, and bin counts become bar heights. Bins touch each other because the underlying axis is continuous.
- Five-number summary Min, Q1 (25th percentile), median (50th percentile), Q3 (75th percentile), max. These five numbers are what a box plot displays in a single picture.
- Quartile (Q1, Q3) The 25th and 75th percentile values. Q1 separates the lowest quarter of the data from the rest; Q3 separates the highest quarter. The middle 50% of the data sits inside the box of a box plot, between Q1 and Q3.
- Skew A distribution that is not symmetric. A right-skewed (positive) distribution has a longer tail to the right; left-skewed has a longer tail to the left. Skew shows up as a lopsided shape in a histogram or an off-center median inside a box plot's box.
Same data, four different pictures.
"A commuter records her drive time in minutes every weekday for two weeks: 22, 28, 25, 31, 26, 23, 27, 24, 29, 26. Which chart type best answers each of these four questions, and what would the chart show?"
Question 1: "How does Monday's commute compare to Friday's?"
This is a comparison between categories (days of the week). Use a bar chart with one bar per day. The bars have visible gaps because the x-axis (days) is categorical, not numeric.
→ bar chartQuestion 2: "What does the distribution of commute times look like?"
This asks about shape: is the data clustered, spread, bell-shaped, skewed? Use a histogram with bins (say, 20-22, 22-24, 24-26, 26-28, 28-30, 30-32). Each bar's height is the count of commutes that fell in that bin. Bars touch because minutes are continuous.
→ histogramQuestion 3: "Where's the center and spread?"
For a quick center-and-spread summary, use a box plot. Sorted data: 22, 23, 24, 25, 26, 26, 27, 28, 29, 31. Median = (26+26)/2 = 26. Q1 = 24, Q3 = 28, min = 22, max = 31. The box runs from 24 to 28 with the median line at 26, and whiskers reach to 22 and 31. One picture, five numbers.
→ box plotQuestion 4: "What fraction of commutes were under 25 minutes?"
Three commutes (22, 23, 24) were under 25 minutes, out of 10 total. That's 30% — "parts of a whole." A pie chart with two slices (under 25 minutes = 30%, 25 minutes or more = 70%) communicates that fraction at a glance. With only two slices, however, a single sentence ("30% were under 25 minutes") arguably does the same job with less ink.
→ pie chart (or just a sentence)The takeaway.
The same data set produced four different pictures, each one answering a different question. The skill being tested isn't "how do I draw a bar chart?" — spreadsheets do that — but "which chart type matches this question?" That choice happens before the software ever opens.
Three problems. Match the chart to the question.
A high school principal wants to display enrollment by grade level (freshman, sophomore, junior, senior). Which chart type fits best? Type one word.
A teacher wants to display the distribution of scores on a 50-question test: how many students scored in the 0-10 range, the 10-20 range, etc. Which chart type fits best?
A researcher wants one picture showing the median, quartiles, and extremes of monthly rainfall over 30 years. Which chart type does that in a single shape?
Three fast questions before you move on.
Q1. What is the main difference between a bar chart and a histogram?
Q2. In a box plot, the line inside the box represents...
Q3. A pie chart shows four categories with slices labeled 35%, 25%, 30%, and 15%. What is wrong with this chart?
The histogram leads to the bell curve.
Of the four chart types in this lesson, the histogram is the one that sets up the rest of the topic. When a histogram's bars trace out a smooth, symmetric, bell-shaped curve, the underlying data is well-modeled by a normal distribution. That observation is what Lesson 5 picks up: it formalizes the bell shape, names its two parameters (the mean and the standard deviation we computed in Lessons 2 and 3), and introduces the empirical rule that powers ALEKS review Q5.
Next: Lesson 5 covers the normal distribution and the empirical rule (68-95-99.7). Lesson 6 then closes the chapter by showing how that same bell curve produces the margin of error on every poll headline.
Continue to Lesson 05Different angle? Need another rep? These are optional — tap any that look helpful.
Charts Are Like Pasta — Crash Course Statistics #5
Bar charts, pie charts, pictographs, and histograms with the categorical-vs-quantitative distinction — the basis of the 'bars touching vs bars with gaps' rule. Strongest single-video coverage of multiple display types.
Plots, Outliers, and Justin Timberlake — Crash Course Statistics #6
Companion episode covering box plots, line charts, and scatter plots — fills the gaps left by #5 in the data-display landscape.