Recent chats

How can I compare two graphs when their scales are different?

Quick answer

Read both axes before comparing shapes: note their ranges, interval sizes, units, and which categories or time points are shown. Then compare values or calculate changes using the same units and matching categories. Different axis scales can make trends look more or less dramatic, so similar-looking lines do not necessarily show equal changes—and lines that look different may reflect different scale choices.

LET’S WORK THROUGH IT

The question, the details, the next step.

Question

I have two graphs that seem to show the same pattern, but their vertical axes use different numbers. What should I check first?

Geez

Start with the axes, not the apparent shape of the lines or bars. For each graph, write down the vertical-axis minimum and maximum, the gap between labeled ticks, and the unit being measured. Also check the horizontal axis: are the same dates, groups, or other categories included, in the same order?

For example, a vertical axis running from 0 to 100 in steps of 20 gives a different visual frame from one running from 40 to 60 in steps of 5. Changing axis scaling can substantially change how a chart appears and can overemphasize or deemphasize a trend [S1]. Matching-looking slopes or bar heights alone do not show that two sets of values changed by the same amount.

Follow-up question

How do I make a fair comparison if one graph uses a narrow range and makes a small change look dramatic?

Geez

Read the plotted values and compare the actual difference, not just the amount of space the change occupies on the page. If the values are comparable, express both changes in the same units. For example, suppose Graph A goes from 50 to 60 units and Graph B goes from 48 to 58 units. Both rise by 10 units, even if one line looks steeper because its vertical axis covers a narrower range.

If the starting values differ, decide whether you need to compare absolute change or relative change. In this example, both absolute changes are 10 units, but the starting points differ. You could calculate each percentage change as the change divided by the starting value, then multiplied by 100, if relative change is what your question asks about. State which comparison you chose: the graph alone does not determine what counts as meaningful.

Follow-up question

What if the graphs use different units or show different categories?

Geez

First check whether the units can be expressed in a common form. If so, convert the values before comparing them and label the converted unit clearly. If they measure different kinds of things, do not treat the raw numbers as interchangeable; explain the difference or compare a shared measure only if one is provided.

Next, match the categories or time points. If one graph shows January through June and the other shows January through April, compare only the months both include, or note that the coverage differs. Watch for categories that are missing, reordered, grouped differently, or labeled in a way that changes what each value represents.

Follow-up question

Can you give me a short checklist I can use while looking at the graphs?

Geez

Use this sequence and jot down the answers before describing the pattern:

  1. Range: Record each axis’s minimum and maximum. Does either axis start above zero or cover a much narrower span?
  2. Intervals: Note the value represented by each tick gap. Are the steps equal, and are they the same on both graphs?
  3. Units: Identify what each number measures. Convert only when the units describe comparable quantities.
  4. Coverage: Match the dates or categories, and note anything missing, reordered, or grouped differently.
  5. Values and change: Compare corresponding values, then calculate the difference. If relevant, compare percentage changes separately.
  6. Conclusion: Describe what the numbers show and mention any scale or coverage difference that could affect the visual impression.

Copyable note: “Graph A uses [range] in [intervals] [unit]; Graph B uses [range] in [intervals] [unit]. Both show [shared categories]. From [start] to [end], A changes by [amount] and B by [amount]. The visual impression [does/does not] match the numerical comparison because [reason].”

A practical checklist

  • Record each axis range and tick interval before comparing graph shapes.

  • Write down the unit for each measure and convert only comparable quantities.

  • Check that both graphs cover the same categories or time points.

  • Compare corresponding values and calculate absolute changes; calculate relative changes only when useful.

  • Describe scale or coverage differences that may affect the visual impression.

A few more useful answers

Does a line that looks steeper always mean a larger increase?

No. The vertical and horizontal scales affect how steep a line appears. Compare the plotted values and the amount of change over the same interval.

Should I always compare percentage changes instead of raw differences?

No. Use raw differences when the amount changed is the relevant question. Percentage change can help compare changes from different starting values, but it answers a different question.

What if the graphs show different time periods?

Compare only the periods both graphs share, if those match-ups are meaningful, and state what is excluded. If there is no suitable overlap, avoid presenting them as a direct comparison.

Sources

References used for this guide.

  1. [S1] Data Analysis Charts: How Axis Scaling Affects Interpretation
    openstax.org · Checked

YOUR QUESTION. YOUR DETAILS.

Let’s work through your situation.

Get a tailored answer or help with your next task. Starts at $1, with a quote before you pay.

Ask your own question
← Back to Geez at Work