Visualizing Data Effectively
Choosing the Right Graph
Graphing goals
A good graph should reveal the distribution, highlight comparisons, and make patterns easy to interpret. The choice depends on whether the variable is categorical or numerical and whether the goal is comparison, trend, or relationship.
Common plots and uses
| Graph | Best for | What it shows |
|---|---|---|
| Bar chart | Categorical data | Counts or proportions by category |
| Histogram | Single numerical variable | Shape, center, spread, and skewness |
| Box plot | Comparing numerical groups | Median, quartiles, IQR, outliers |
| Scatter plot | Two numerical variables | Association, direction, strength, clusters |
Visual design tip
Start axes at zero for bar charts unless there is a clear reason not to; otherwise differences can look exaggerated.
Reading a distribution
Interpreting a histogram
-
1
Step 1: Identify the main shape and number of peaks.
-
2
Step 2: Note the center and range of values.
-
3
Step 3: Look for skewness and unusually high or low values.
-
4
Step 4: Consider whether bins are too wide or too narrow.
-
5
Step 5: State the interpretation in context.
Which plot is most appropriate for showing the relationship between two numerical variables?
Scatter plots are designed for two numerical variables.
Correct answer: Scatter plot
What feature of a box plot identifies the middle 50% of the data?
The box spans the first and third quartiles, so its length is the IQR.
Correct answer: The interquartile range, shown by the box from to .