You are making a bar graph of class votes on favorite colors. The most votes any color received was 18. Which scale would work best for the vertical axis?
A0, 2, 4, 6, 8, 10 — keeps numbers small and easy to read
B0, 5, 10, 15, 20 — starts at 0, goes by equal steps of 5, reaches past 18
C1, 2, 3, 4, 5 — simple single-digit numbers
D0, 10, 30, 50 — round numbers that are easy to remember
A good scale starts at 0, uses equal steps, and extends just past the largest value. A scale of 0, 5, 10, 15, 20 satisfies all three: it starts at 0, goes by equal increments of 5, and reaches 20 which is above the maximum of 18. Option A only goes to 10, which can't show a bar of 18. Option C only goes to 5. Option D uses unequal jumps (10, then 20, then 20 again), which distorts the visual comparison between bars — a major error that makes the graph misleading.
Question 2 Multiple Choice
What is wrong with a bar graph whose scale goes 0, 5, 10, 15, 25?
AThe scale is wrong because it doesn't start at 1
BThe last step jumps by 10 instead of 5, making the intervals unequal and distorting the graph
CThe scale should only use even numbers to be accurate
DNothing — any increasing numbers work as a scale
The intervals on a scale must be equal. Going 0, 5, 10, 15 uses steps of 5, but then jumping to 25 uses a step of 10. Unequal steps distort the graph visually: a bar reaching 25 would look only slightly taller than a bar reaching 15, even though it's 10 units higher. The graph's power is that height visually represents quantity — unequal steps break that relationship and make the graph misleading.
Question 3 True / False
A bar graph would be a good choice for showing how a plant's height changes each day over two weeks.
TTrue
FFalse
Answer: False
Bar graphs are designed for categorical data — distinct, named groups that don't have a natural order or continuity (favorite colors, types of pets, preferred subjects). Plant height over time is continuous data with a meaningful sequential order — each day follows the previous one. This type of data is better represented with a line graph, which shows change over time and allows you to see trends and patterns. Using a bar graph for time-series data suggests the days are independent categories rather than points on a continuum.
Question 4 True / False
All bars in a bar graph should be drawn the same width.
TTrue
FFalse
Answer: True
Consistent bar width is a visual design rule that ensures fairness and clarity. If bars have different widths, readers might interpret the area (width × height) rather than just the height as the data value, which would distort comparisons. Equal widths mean the only variable that carries information is height, which directly represents the count. Consistent width is part of what makes a bar graph readable and honest.
Question 5 Short Answer
When creating a bar graph, why does it matter where your scale starts and how its intervals are spaced?
Think about your answer, then reveal below.
Model answer: The scale must start at 0 so that bar height accurately represents the actual count — a bar for 10 should be twice as tall as a bar for 5. The intervals must be equal so the visual spacing between values is consistent. If the scale doesn't start at 0 or uses unequal steps, bars that look twice as tall don't actually represent twice as much data, which makes the graph misleading.
This question goes to the heart of why bar graphs work: their power is that height directly encodes quantity, allowing instant visual comparison. That only works if the scale is linear (equal steps) and starts at zero (so relative heights are proportional to relative quantities). Graphs with non-zero baselines or unequal intervals are a common source of data misrepresentation — even in professional contexts.