6 Charts and Data Visualization

Match the chart to the question

Choose a chart by asking what relationship the reader should notice. Common goals include comparing categories, showing change over time, displaying a distribution, showing parts of a whole, and examining the relationship between two numerical variables. A chart is usually clearer when it emphasizes one main message rather than trying to show several at once.

Choose among common chart types

Use a bar chart or column chart to compare values across categories, such as sales by product. A horizontal bar chart can work well when category names are long. For sales of several products in each region, a clustered bar or column chart can compare a few series across categories, but too many series can make the chart crowded.

Use a line chart to show change over ordered time, such as monthly temperature or revenue, and arrange time chronologically. Columns can be useful for amounts accumulated within separate periods.

To show parts of a whole, use a stacked bar or column chart. A pie chart is appropriate only when there are few categories and the values make up one meaningful whole, such as departments’ shares of a total. A 100% stacked chart can compare proportions across groups.

Use a histogram or box-and-whisker plot to show the spread of numerical values, such as a distribution of test scores. A box-and-whisker plot can also help compare score distributions across classes. Use a scatter plot to examine the relationship between two numerical variables, such as study hours and exam scores; each point represents a pair of values.

Prepare and build the chart

Before charting, check that the data has clear column headings, one observation or category per row, and values in the correct format. Select only the headings and data needed for the comparison. In a spreadsheet, use Insert to choose a chart type or preview recommended charts. Treat recommendations as starting points, then check whether the chart fits the question and data.

For example, to compare monthly sales for three products, select the month labels and the three product columns, then insert a line chart. If the months appear as separate lines and the products appear along the horizontal axis, use the chart’s data or row-and-column controls to correct the arrangement.

Format for clarity and accessibility

Give the chart a concise, informative title that states what is being compared and, when useful, the time period. Label the axes and include units, such as revenue in dollars or temperature in degrees Celsius.

Use a legend when needed, but label lines or series directly when that is clearer. Do not rely on color alone to distinguish categories; direct labels or different symbols can also help readers. Use readable text and restrained colors, and include only gridlines or data labels that help readers understand the values.

Keep scales honest and easy to read. Bar and column charts used for comparisons generally need a zero baseline. If an axis is truncated, make that clear because truncation can exaggerate differences. Avoid decorative effects such as 3-D perspective when they make values harder to compare.

Interpret and check the chart

Before drawing a conclusion, read the title, axes, units, scale, and legend. Then look for the pattern relevant to the question, such as the largest or smallest category, a rise or decline, a cluster, a gap, or an unusual value. A scatter plot may show that two variables tend to increase together, but that pattern alone does not prove that one causes the other.

Check that the plotted data matches the source table, that categories are in the intended order, and that missing or filtered values have not changed the story. A useful chart makes its main finding visible without hiding important context.