Evaluate the presentation of data

Introduction to Evaluating Data Presentation

In your Thinking Skills exam, especially in Paper 2, you won't just read text; you will often be given data in the form of charts, graphs, and tables. While numbers seem like "cold, hard facts," the way they are presented can be very misleading. Sometimes a graph is drawn to make a small increase look like a massive jump, or a table might leave out important context.

Evaluating the presentation of data means looking closely at how information is shown to see if it is clear, fair, and accurate. It’s about being a "data detective" and spotting tricks that might lead someone to the wrong conclusion.

***

1. Common Tricks in Visual Presentation

Graphs and charts are powerful because our eyes process images faster than numbers. However, this also makes it easy for a presenter to "cheat" our visual perception. Don't worry if you aren't a math expert; you just need to look for these specific features:

The "Vanishing" Zero (Truncated Axes)

The most common way to mislead is to start the vertical axis (the \(y\)-axis) at a number other than \(0\). This is called a truncated axis. By starting the scale at a higher number, small differences between bars or points look much larger than they really are.

Example: Imagine a graph showing a rise in crime from \(100\) cases to \(105\) cases. If the axis starts at \(95\), the bar for \(105\) will look twice as tall as the bar for \(100\). This makes a \(5\%\) increase look like a \(100\%\) increase!

Inconsistent Scales

Check if the gaps between numbers on the axes are equal. If an axis jumps from \(10\) to \(20\), then suddenly from \(20\) to \(100\), the "shape" of the data is distorted. A steady increase might look like it is slowing down, or vice-versa.

Comparing "Apples to Oranges"

Sometimes two graphs are placed side-by-side to suggest a relationship, but they use completely different scales or units. Always check the labels and units. If one graph measures "total cost in \( \$\)" and the other measures "percentage of profit (\( \% \))", they cannot be compared directly without careful calculation.

Quick Review: When looking at a graph, always ask: "Does the vertical axis start at \(0\)? Are the intervals equal? What are the units?"

***

2. Evaluating Numerical Presentation

Even when data is in a simple table, the way the numbers are grouped or described can be a bit "sneaky."

Percentages vs. Absolute Numbers

This is a favorite topic for examiners. Percentages can hide the actual size of a change. For example, a " \(100\% \) increase" sounds terrifying, but if it means the number of accidents went from \(1\) to \(2\), the absolute number is still very small. Conversely, a small percentage of a very large number (like \(0.5\%\) of a country's population) is actually a huge amount of people.

Selective Data (Cherry-Picking)

Data presentation is often misleading because of what it leaves out. If a company shows a graph of rising profits over three months, they might be "cherry-picking" a lucky streak while ignoring the fact that profits have been falling for the last three years.

Did you know? This is often called "suppressed evidence." If the data only shows a specific time frame or a specific group, the presentation might not tell the whole story.

Averages: Mean, Median, or Mode?

In Thinking Skills, we use the term average carefully. A "mean" average can be skewed by one or two extremely high or low numbers (outliers). If a presenter uses an average to describe a "typical" situation, ask yourself if a few extreme cases are making the data look better or worse than it really is.

***

3. Logical Relationships in Data

The syllabus requires you to identify logical relationships and patterns. When you evaluate the presentation, look for these three things:

• Correlation vs. Causation: Just because two lines on a graph go up at the same time (correlation) does not mean one caused the other (causation). The presentation might try to trick you into thinking there is a link when there isn't.
• Trends: Is there a clear direction (upward or downward)? If a presentation claims a "trend" based on only two or three data points, the evidence is weak.
• Consistency: Does the data in the table actually support the claim being made in the text? Sometimes the labels on a chart don't quite match the argument the author is making.

***

4. Step-by-Step: How to Evaluate Data in the Exam

When you encounter a source with data in Paper 2, follow these steps to assess its quality:

Step 1: Check the Labels. Look at the title, the axis labels, and the legend (key). Do you know exactly what is being measured? Is it \( \$\), \( \% \), or raw counts?

Step 2: Check the Scale. Does the \(y\)-axis start at \(0\)? If not, the differences are exaggerated. Are the increments (the gaps between numbers) consistent?

Step 3: Look for Context. Is there a date range? Is it too short to show a real trend? Is there data missing that you would need to make a fair judgment?

Step 4: Check the Wording. Does the author use "loaded" language like "skyrocketing" to describe a tiny increase shown in the data?

Common Mistake to Avoid: Don't just describe the data (e.g., "The line goes up"). To evaluate, you must explain why the presentation might be misleading or unhelpful (e.g., "The line appears to rise steeply, but this is because the axis starts at \(500\) rather than \(0\), making the increase seem more significant than it is").

***

Key Takeaways

1. Visuals can lie: Always check for truncated axes and inconsistent scales that exaggerate changes.
2. Percentages need context: \(50\%\) of a small number is still small; \(1\%\) of a huge number is still huge.
3. Look for gaps: Selective data presentation (cherry-picking) can hide the real trend.
4. Distinguish correlation from causation: Don't assume that because two data sets look similar on a graph, one is causing the other.
5. Use units and labels: Labels and units are the most important clues for spotting if a comparison is fair.