What should you check before comparing two bars?
GMAC says Graphics Interpretation may use several visual forms, including scatterplots, x/y graphs, bar or pie charts, and statistical distribution curves. On the exam, answers are selected from dropdown menus. This Upsilar practice workshop uses two fictional bar charts; its questions do not reproduce the exam interface.
Read the title, time period, unit, and axis origin. A bar at 48 thousand represents 48,000 units, not 48. A truncated chart can make a small difference look dramatic; the numerical values should determine your answer. Here, the bars start at zero, and the same data appear in a table so you can work without the chart.
Then ask whether the categories are comparable. Quarterly sales cannot be directly compared with an annual cumulative total. A taller bar tells you something about the quantity shown, not necessarily profitability or the number of customers.
Does an increase of eight units always correspond to the same percentage?
The chart shows fictional quarterly sales for a company, in thousands of units. It represents levels, not growth rates directly.
| Reference | Quarter | Units sold (thousands) |
|---|---|---|
| 1 | Q1 | 40 |
| 2 | Q2 | 48 |
| 3 | Q3 | 44 |
| 4 | Q4 | 55 |
From Q1 to Q2, sales increased by what percent? Was the percentage increase from Q3 to Q4 larger?
Choose an answer.
The solution, step by step
20%; Yes.
The method in action
- From Q1 to Q2, the increase is 8 from a starting value of 40, so 8/40 = 20%. The difference in bar heights is not enough; divide by the starting quarter’s value.
- From Q3 to Q4, the increase is 11 from 44, or 25%. It is greater both as a percentage and in absolute volume.
- A 25% increase from Q1 to Q2 would use the wrong base; 8% confuses the absolute difference with a rate. Both “No” choices contradict 25% > 20%.
Which category represents one quarter of the total?
A second, independent dataset: 120 fictional deliveries across four areas. The numbers describe mutually exclusive categories that cover all deliveries in this example.
| Reference | Area | Number of deliveries |
|---|---|---|
| 1 | North | 30 |
| 2 | South | 45 |
| 3 | East | 15 |
| 4 | West | 30 |
What fraction of all deliveries went to the East area? Which two areas each accounted for 25 percent of all deliveries?
Enter your answer before viewing the review.
The solution, step by step
East: 1/8. North and West each accounted for 25%.
The method in action
- The total is 30 + 45 + 15 + 30 = 120. East accounts for 15/120 = 1/8, or 12.5%. The denominator is the total across all four areas.
- North and West each account for 30/120 = 1/4. South accounts for 45/120 = 3/8, not one quarter.
- Comparing East only with South would give 1/3, a ratio between two areas that does not answer the question about the whole.
The pitfall
A chart cannot tell you what it does not measure
Sales rise from Q3 to Q4, but the chart provides neither prices nor costs. It cannot show that profit increased. Likewise, an area receiving more deliveries does not necessarily have more customers: several deliveries may go to the same customer.
Stay within the available evidence. If a relationship depends on a missing variable, do not fill it in with a plausible story. That would be an inference error, even if all your calculations are correct.
How can you vary this workshop?
For each chart, write one question about a value, one about a difference, and one about a proportion. Identify how the calculation changes. Then find a claim that cannot be established from the data and name the missing information.
This routine builds active reading: every answer should connect to a unit, a comparison base, and the values used. It also prepares you to read other visual forms, without suggesting that two charts cover every exam format.
Key takeaways
- Read the unit, time period, and axis origin.
- Distinguish an absolute difference from a relative change.
- Do not infer profit or causation without corresponding data.