The learner evaluates whether a data display represents quantities faithfully by examining axis scales, intervals, truncated axes, unequal categories, pictographs, and visual features such as area or three-dimensional effects that can exaggerate differences. They assess statistical claims in light of sample size, sampling method, omitted context, absolute versus relative change, and the difference between association and causation, recognizing that a graph or summary can be numerically accurate yet misleading. Formal statistical inference, confidence intervals, and advanced causal analysis are outside this scope.
A data display can use correct numbers but still create a misleading impression. To evaluate it, check:
A school surveys students about how often they recycle.
A bar graph shows the two numbers, but its vertical axis starts at instead of .
The number of students increased by
So, more students recycled.
The relative increase is
The percentage increase is , but the number increased from to only students.
Because the graph starts at , the first bar rises only from to , while the second rises from to . The second bar may look about four times as tall as the first:
However, the actual number of recyclers is only
times as large. The truncated axis exaggerates the visual difference.
The school says, “The campaign caused recycling to increase dramatically for all students.”
A better evaluation is:
A fair conclusion would be: “Among the surveyed students, the number who recycled increased from to 10.”
Click a topic below to practice the foundational skills you'll need, learn the steps, or master this skill
Earned ?