Bar Charts and Line Plots
Bar charts compare category values using bar lengths, while line plots in elementary statistics place repeated marks above values on a number line. In brief: a bar reaching 8 can represent eight responses in one category, and four marks above the value 3 in a line plot represent four observations equal to 3.
The term line plot is sometimes used differently, so this guide uses it for a dot plot or stacked-mark plot, not a connected time-series line graph. Both displays summarize data, but they organize the horizontal positions differently. Clear labels, scales and counting rules make the distinction easy to read.
What does a bar chart show?
A bar chart assigns a bar to each category, with its length representing a numerical quantity such as a count, percentage or total. The categories may be types of transport, groups of respondents or named periods. The value axis and title identify what the bar lengths measure.
Bars can be vertical or horizontal. Horizontal bars are often convenient when category names are long, while vertical bars are familiar for short labels. Orientation does not change the data, provided the scale and category mapping remain clear.
Bars ordinarily have equal widths and visible gaps because they represent separate categories. Their lengths, not their decorative areas or volumes, carry the values. A three-dimensional styling effect can make those lengths harder to compare, so a simple flat display is often clearer.
What is a line plot or dot plot?
A line plot places a mark above each observed numerical value on a number line. Repeated values produce a stack of marks. If one mark represents one observation, the stack height is the frequency of that value.
Unlike a category bar chart, the horizontal positions have numerical spacing. The distance from 1 to 2 should match the distance from 2 to 3 on an ordinary linear number line. Leaving a value with no observations as an empty position preserves that numerical structure.
Some classroom displays use X marks instead of dots. The symbol choice does not change the interpretation, but a key should state whether each mark represents one observation or a larger count. This article uses one X per observation in its text-based representation.
How do you construct a bar chart from a table?
Suppose a hypothetical survey records preferred travel to school: walking 8, cycling 5, bus 11 and car 6. Place the four categories on one axis and label the other axis number of students. Use a value scale beginning at zero and reaching at least 11.
Draw equal-width bars to heights 8, 5, 11 and 6, with consistent gaps. The bus category is largest, and the total is 30 students. A scale step of one or two students can work, provided the intermediate height 11 remains readable.
The text display below uses one X per student to show the same comparisons in a simple form. It is a compact representation of bar lengths, not a conventional axis drawing. The accompanying numerical column lets the reader verify each value without depending on visual counting alone.
- Identify the categories and the count being displayed.
- Choose a zero-based value scale reaching at least 11.
- Draw equal-width bars with heights 8, 5, 11 and 6.
- Add a title and axis labels, then compare every bar with the source table.
| Travel mode | Count | Text bar |
|---|---|---|
| Walking | 8 | XXXXXXXX |
| Cycling | 5 | XXXXX |
| Bus | 11 | XXXXXXXXXXX |
| Car | 6 | XXXXXX |
How do you construct a line plot from numerical data?
Consider the hypothetical observations 1, 2, 2, 3, 3, 3, 4, 4, 5. Draw a number line covering 1 through 5 with equal intervals. Put one mark above 1, two above 2, three above 3, two above 4 and one above 5.
The plot contains nine marks altogether, matching the nine observations. Its tallest stack is at 3, so 3 is the most frequent value. The smallest observed value is 1 and the largest is 5, giving a range of 4.
The table below records the stack associated with each number-line position. In a drawn plot, these marks would be stacked vertically over their numerical locations. Keeping all values from 1 through 5 in the reference makes the spacing and counts unambiguous.
| Number-line value | Frequency | Marks at that value |
|---|---|---|
| 1 | 1 | X |
| 2 | 2 | XX |
| 3 | 3 | XXX |
| 4 | 2 | XX |
| 5 | 1 | X |
How do bar charts, line plots and line graphs differ?
A bar chart emphasizes comparisons among categories, and rearranging unordered categories may change the presentation without changing their meaning. A line plot preserves a numerical axis and shows the distribution of observations along it. Its gaps and clusters have numerical significance.
A line graph connects successive points, usually over time or another ordered variable. The connections emphasize change between observations. It is different from a line plot of stacked observations, despite the similar name.
Choose the display according to the question. Preferred transport suits a category bar chart, small numerical measurement data suit a line plot, and temperature recorded at successive times may suit a line graph. A display should express the variable’s structure rather than being chosen only because it is visually familiar.
What comparisons can you make from these displays?
From the travel bar chart, bus exceeds cycling by 11 − 5 = 6 students. Walking and cycling together account for 13 of the 30 responses. The chart supports these count comparisons directly, but does not explain why students chose those modes.
From the line plot, six of nine observations are at most 3 because the first three frequencies total 1 + 2 + 3 = 6. Three observations exceed 3. The plot also shows a concentration near the middle value without needing a separate frequency table.
A visual pattern should be described in terms of the actual data and scale. Do not claim that a difference is important in a broader population solely because one bar is taller. The sample size, sampling process and practical context would be needed for stronger conclusions.
Which scale and labeling choices make a display clear?
Use a value range that includes all the observations and a tick interval that can be read without excessive crowding. A bar chart’s zero baseline lets lengths represent full values fairly. If the chart shows percentages, label the axis as percentages rather than counts.
For a line plot, preserve equal numerical spacing and include intermediate positions even when their frequency is zero. If measurements occur in quarters or halves, label the number line with the corresponding fractional intervals. A stack above 1.5 must be positioned halfway between 1 and 2 on a linear scale.
Provide a key when one mark represents more than one observation. A symbol representing two students cannot be counted as one student merely because it appears once. Half symbols or partial marks need an explicit convention, and using one mark per observation avoids that complexity for small datasets.
Why are these displays useful and which mistakes are common?
Both displays turn organized counts into visual comparisons. Their background is frequency representation: repeated observations are summarized by a bar length or stack height. The numerical axis in a line plot additionally preserves the order and spacing of the measured values.
Common mistakes include connecting category bars with an unjustified line, treating a line plot as a time trend and omitting zero-frequency numerical positions. Another is choosing a nonzero baseline that makes modest bar differences look much larger than their actual values.
Check the total count, every category label and the meaning of each mark before interpreting the picture. Reconstructing the source frequencies from the display is a strong practical check. A readable chart should allow another person to recover the main values and understand exactly what the observations represent.