Take 11 quiz scores in order: 3, 6, 7, 7, 8, 8, 8, 9, 9, 10, 10. The median, the 6th value, is 8. Leave it out and split the rest: the lower half 3, 6, 7, 7, 8 has median Q1=7, and the upper half 8, 9, 9, 10, 10 has median Q3=9. With the minimum 3 and maximum 10, the five-number summary is 3, 7, 8, 9, 10. Draw a box from 7 to 9 with a line at 8, and whiskers reaching out toward the lowest and highest scores.
Each piece of the plot holds about a quarter of the scores: the left whisker, the left part of the box, the right part and the right whisker. Their lengths show spread, not how many values they hold. Outliers are checked with fences: IQR=9−7=2, so the lower fence is 7−1.5(2)=4. The score 3 is below it, so it is drawn as a separate dot, and the left whisker stops at 6, the lowest score that is not an outlier.
A box plot hides detail. A data set with one hump and another with two separate humps can share the same five numbers and draw identical box plots. If the question is about shape, such as whether there are two peaks, use a histogram or a dot plot. Also, the line in the box is the median, not the mean; in skewed data the two can be far apart.
Think of it like this. A box plot is like recording five moments of a race: when the first runner, the quarter-way runner, the middle runner, the three-quarter runner and the last runner cross the line. Each gap holds a quarter of the field. A long gap means those runners were strung out, not that there were more of them.
Core idea. A box plot pictures min, Q1, median, Q3, and max as a box with whiskers, showing center and spread at a glance.