Reading a histogram: frequency, bins, and distribution shape

A histogram condenses a data set by grouping values into neighboring intervals. Unlike a bar graph of named categories, the bars represent ranges on a numerical scale and usually touch. A question may ask for a frequency, a range, the most common interval, or a description of distribution shape.

Read the axes before interpreting the bars

The horizontal axis shows the bins, such as test scores from 60 to under 70 and 70 to under 80. The vertical axis shows the number of observations, percentage, or another stated quantity. Check whether a boundary belongs to the left or right interval; a score exactly equal to 70 may be placed in the 70–79 bin depending on the chart’s labels.

Find frequency and range

The height of a bar gives the frequency for its bin when all bins have equal width and the axis is labeled frequency. If a graph supplies unequal bin widths, bar area or density may represent frequency instead, so read the axis label and instructions. To estimate the range, identify the lowest and highest intervals containing data; the histogram may not reveal each exact minimum or maximum.

Describe the shape

A cluster is a region with many observations; a gap has few or none. A roughly balanced distribution has similar tails on each side of its center. A right-skewed distribution has a longer tail toward higher values; a left-skewed one has a longer tail toward lower values. A separated bar may suggest an unusual value, but the graph alone does not explain why it occurred.

Common interpretation errors

• Calling a histogram bar a single value rather than an interval.
• Comparing bar heights without checking whether bin widths differ.
• Assuming the tallest bin equals the mean; it identifies the modal interval.
• Claiming exact individual values when only grouped data are shown.
• Ignoring a broken or nonzero vertical-axis scale.

Key takeaway

Read the intervals and axis scale first. Then use bar heights and distribution shape to answer the specific question, while remembering that grouped bars do not show every observation.

Read the bins before describing the shape

A histogram groups numerical observations into intervals called bins. The horizontal axis shows the value intervals and the vertical axis usually shows frequency or relative frequency. A bar height represents how many observations fall in its interval; the bars touch because the numerical scale is continuous. Read the interval boundaries before making a claim. If a bar covers scores from 60 up to but not including 70, its height does not mean every score in that interval was 60 or 70.

Suppose a histogram of commute times has bin heights 4, 9, 15, 8, and 3. The tallest bar shows the interval with the most observations, or the modal bin. The total number of observations is 39 if bar heights are frequencies and the bins cover all cases. If the vertical scale is relative frequency, heights may instead sum to 1 or 100%. Check the axis label rather than assuming what height measures.

Describe symmetry, skew, and gaps cautiously

A distribution is roughly symmetric when its left and right sides have similar shapes around a center. A right-skewed distribution has a longer tail toward larger values; a few high observations pull the tail right. A left-skewed distribution has a longer tail toward smaller values. The skew direction is named for the tail, not for where most bars are concentrated. In a right-skewed graph, many observations can sit at lower values while the tail stretches toward high values.

A gap is an interval with no observed values, while an isolated bar may indicate an outlier or a separate subgroup. Neither conclusion should be made without checking the scale and bin width. Different bin widths can make the same data look smoother, lumpier, or more separated. When comparing histograms, make sure axes and bin definitions are comparable.

Do not infer exact values from grouped data

A histogram summarizes values within intervals; it generally does not reveal every individual observation. If a bin covers 20 to 30 and has height 12, you know there are 12 values in that interval, not their exact values. You may estimate a center from the picture, but exact median or mean calculations require the underlying data or additional stated assumptions. The tallest bin identifies a modal interval, not necessarily one exact mode.

Also distinguish frequency from density. When all bins have equal width, comparing bar heights usually compares counts. With unequal bin widths, some histograms scale height so that bar area, rather than height alone, represents frequency. Read the graph’s notes. A question that gives unequal intervals may be testing this distinction.

• Read both axes and units first.
• Add bar frequencies only when the axis shows counts and all observations are included.
• Name skew by the direction of the longer tail.
• Treat bins as ranges; do not claim to know exact observations inside them.
• Compare bin widths and scales before comparing two histograms.

Compare distributions using consistent intervals

The shape of a histogram depends partly on bin width and starting point. If two graphs use different intervals, bar heights may not be directly comparable even when they contain the same data. Before comparing groups, check that bin widths, axis scales, and frequency definitions match. A different bin choice can reveal or hide local clusters without changing any observation.

A cumulative frequency graph or table answers how many observations are at or below a boundary. To estimate the median, locate half the total count and find the interval where cumulative frequency reaches that position. Grouped data generally identify a median interval rather than its exact value; do not claim precision the bins do not provide.

Apply it to the evidence or sentence

A histogram groups quantitative values into intervals, or bins, and uses adjacent bars because the scale is continuous. Read the horizontal axis to identify each interval and the vertical axis to identify its frequency or relative frequency. If bins have equal widths, taller bars represent more observations; with unequal widths, compare bar area or the density scale as specified. A bar covering 20–30 may include one endpoint and exclude the other according to the bin convention, so do not infer exact values from the graph unless boundaries are shown. Histograms describe distribution shape, center, spread, and possible gaps, but they do not preserve individual data values.