Choosing the Right Graph
In 2021, the Australian Bureau of Meteorology published Sydney's daily temperature record for 130 years as a single line graph — 47,450 data points that no pie chart could ever show.
Printable Worksheets
Print or save as PDF — or build a custom worksheet from any module's questions.
You have data showing the temperature of a cooling liquid every minute for ten minutes, and data showing the favourite science topics of your classmates.
Would you use the same type of graph for both sets of data? Why or why not?
Picture a thermometer in the school garden, checked every hour for a week. Monday hits 12°C at 7 am, rises to 28°C by 2 pm, drops back to 14°C by 8 pm. Plot those points on a grid and connect them — you instantly see the daily rise and fall. That shape is only visible because of the type of graph you chose. Line graphs are ideal when you want to show how something changes continuously over time or another numerical scale. The independent variable — usually time — goes on the horizontal axis. The dependent variable goes on the vertical axis. Points are plotted at each measurement and connected by lines to reveal the overall trend.
Use line graphs for temperature changes throughout a day, plant growth over several weeks, or reaction rates as chemicals are consumed. The key requirement is that both axes represent numerical scales. If your horizontal axis is made of categories (like 'red,' 'blue,' 'green') rather than numbers, a line graph is the wrong choice.
A student measures the temperature of a cooling liquid every minute for ten minutes. A line graph with 'Time (minutes)' on the horizontal axis and 'Temperature (degrees Celsius)' on the vertical axis shows a smooth downward curve. This makes the cooling pattern instantly visible — something a table of numbers alone cannot do as effectively.
The Bureau of Meteorology publishes line graphs for temperature, rainfall, and pressure trends across Australia. These graphs let farmers, pilots, and emergency services see patterns at a glance. A table of hourly temperatures would contain the same data, but the line graph makes the trend obvious.
Some students think a line graph is just dots connected by lines, so you can use it for any data. This is wrong. Connecting points with lines implies a continuous relationship between the values. If your categories have no natural order — like favourite colours or types of pets — the connecting lines are meaningless and misleading.
Know
- Line graphs are used to show trends over time or continuous data.
- Bar graphs are used to compare discrete categories or groups.
Understand
- The choice of graph affects how easily patterns and relationships can be seen.
- Inappropriate graph types can mislead or obscure the true story in the data.
Can Do
- Select an appropriate graph type for a given dataset.
- Justify graph choices using the nature of the data.
Wrong: You can use any graph type for any data; it is just personal preference.
Right: Graph types are designed for specific data structures. Using the wrong type can hide patterns or mislead the reader.
Wrong: Pie charts are good for comparing many categories.
Right: Pie charts work best with only a few categories that make up a whole. With many categories, they become difficult to read and compare.
Wrong: Using a line graph for categorical data.
Right: Line graphs imply continuity between points. Connecting categories like 'red', 'blue' and 'green' with a line suggests a relationship that does not exist.
Wrong: Starting the vertical axis above zero in a bar graph.
Right: This exaggerates differences between bars. Always start bar graph axes at zero unless there is a very good reason and the break is clearly marked.
Match each dataset to the most appropriate graph type.
Bar graphs are used to compare values across different categories. Each bar represents a group, and the height of the bar shows the value for that group. Use bar graphs when your independent variable is categorical — types of animals, brands of batteries, or months of the year — rather than a continuous numerical scale.
There are two critical rules for drawing bar graphs. First, all bars should be the same width and separated by equal gaps. Second, the vertical axis must start at zero. Starting above zero exaggerates small differences and can mislead readers into thinking one category is dramatically different when the actual gap is tiny.
A student surveys their class to find favourite science topics. The categories are Biology, Chemistry, Physics, and Earth Science. A bar graph with these four categories on the horizontal axis and 'Number of Students' on the vertical axis makes it easy to see which topic is most popular. A line graph would be wrong here because the categories have no natural numerical order.
The Australian Bureau of Statistics uses bar graphs extensively to compare states, industries, and age groups. Their style guide strictly requires that all bar graphs start at zero on the vertical axis. This rule exists because misleading bar graphs have been used in advertising and politics to make small differences look enormous.
Some students think you can start the vertical axis anywhere to 'make the graph fit the page.' This is a serious error. A bar graph that starts at 80 instead of 0 can make a value of 85 look five times larger than a value of 81, when the real difference is less than 5 percent. Starting at zero is not a suggestion — it is a requirement for honest data presentation.
Pie charts show how a whole is divided into parts. They are useful for displaying percentages or proportions, such as the composition of air or the distribution of energy sources. Choose pie charts only when you have a small number of categories — ideally fewer than six. Too many slices make the chart unreadable.
Scatter plots display individual data points on a grid with two numerical axes. They are excellent for revealing relationships between two variables, spotting correlations, and identifying outliers. If the points form a rough line, the variables are correlated. If they form a cloud, there is probably no relationship.
A pie chart showing that air is approximately 78% nitrogen, 21% oxygen, and 1% other gases is clear and effective. A scatter plot showing hours of sleep against test scores for thirty students might reveal that students who sleep 7–9 hours tend to score higher, while those who sleep less than 6 hours score lower — a pattern that would be invisible in a table.
The Australian Bureau of Statistics uses pie charts in its demographic reports to show population composition by age group or ancestry. For economic data, they prefer scatter plots and line graphs because these reveal trends and relationships that pie charts cannot display.
Some students think any graph looks 'scientific' so the choice does not matter. This is completely wrong. The wrong graph can hide the truth just as effectively as wrong data. A pie chart of weekly temperature would be meaningless because temperature is not a part of a whole. A bar graph of two correlated variables would hide the relationship. The graph type is part of the scientific argument.
Before choosing a graph, ask what you want the reader to see. A trend over time? A comparison between categories? A proportion of a whole? A relationship between two variables? The answer determines your graph type. The wrong choice does not just look odd — it can actively hide the truth or create false impressions.
Beyond choosing the right type, every graph needs clear labels. Axes must show the quantity and unit. The title must describe what the graph displays. If multiple data sets are shown, use a legend. A beautiful graph with unlabelled axes is useless because no one can interpret it.
A researcher wants to show that sea ice is declining. A line graph of ice extent over thirty years makes the downward trend obvious. A bar graph of the same data would work but would waste space. A pie chart would be completely wrong because ice extent is not a part of a whole. The right graph makes the message clear; the wrong graph buries it.
Both CSIRO and the ABS publish style guides for graph choice that their scientists must follow. These guides specify which graph types are appropriate for which data, minimum font sizes, colour palettes for accessibility, and mandatory axis labels. Professional science is as much about clear communication as it is about correct data.
Many students believe all graphs are interchangeable and the choice is just personal preference. This is one of the most dangerous ideas in data literacy. Graphs are tools, and like any tool, each has a specific job. Using a pie chart for time-series data or a line graph for categorical data is like using a hammer to paint a wall — it does not just fail, it makes a mess.
True or false? Tap as fast as you can. Build a streak.
Line graphs are best for showing trends over time.
Bar graphs should start at zero on the vertical axis.
Pie charts are good for comparing values between unrelated categories.
Scatter plots reveal relationships between two numerical variables.
Any graph type works for any dataset.
A graph with unlabelled axes is still useful if it looks professional.
Speed Round · 6 questions
True or false? Tap as fast as you can. Build a streak.
A line graph is the best choice for showing how temperature changes over time.
A pie chart is ideal for comparing the test scores of twenty different students.
Bar graphs should generally start the vertical axis at zero to avoid misleading comparisons.
A scatter plot is used to show proportions of a whole.
Categorical data should be displayed using a bar graph rather than a line graph.
Using the wrong graph type can never mislead the reader because all graphs show the same data.
How are you completing this lesson?
At the start of the lesson you were asked: "Why won't you ever see a pie chart of weekly temperature?" Before this lesson, you might not have known exactly why — it just might have seemed odd.
Now that you know the rules for choosing graph types, can you fully explain it? Think about whether weekly temperature is continuous or categorical, and what a pie chart is actually designed to show.
Choose the appropriate graph type for each dataset and explain why the other type would be unsuitable.
Quick Check · 5 questions
Check Your Understanding · 3 questions
1. A scientist measures the pH of a solution every 30 seconds during a reaction. What graph should they use and why?
2. Why is it misleading to use a pie chart to compare the heights of ten different students?
3. Describe a situation where a scatter plot would be more useful than a bar graph.
Show Your Working · 3 questions
SA1. Compare line graphs and bar graphs, explaining the type of data each is designed to display and giving a scientific example for each.
SA2. A student presents categorical data as a line graph. Explain why this is inappropriate and describe what misleading impression it might create.
Hint: Think about what a connecting line implies about the data points.
SA3. Describe the features of a well-constructed graph, regardless of the type chosen.
Quick Check
1. C — A line graph shows change over time. Plant height measured weekly is continuous data.
2. B — A pie chart shows parts of a whole like air composition.
3. B — Starting at zero prevents misleading comparisons of bar heights.
4. D — Scatter plots reveal relationships between two numerical variables.
5. B — The wrong graph can mislead the reader or hide important patterns.
Show Your Working Model Answers
SA1 (4 marks): Line graphs show trends over time or continuous data [1], e.g. temperature changes during a reaction [1]. Bar graphs compare discrete categories or groups [1], e.g. favourite science topics of classmates [1].
SA2 (4 marks): Categorical data has distinct groups with no numerical continuity between them [1]. A line graph implies continuity by connecting points with lines [1]. This suggests a trend or relationship that does not exist between categories [1]. The reader may be misled into thinking the categories follow a sequence or pattern [1].
SA3 (3 marks): A well-constructed graph has a clear title [1], labelled axes with quantities and units [1], a suitable scale, and uses the correct graph type for the data [1].
Line graph
Shows trends over time or continuous data
Bar graph
Compares values across discrete categories
Pie chart
Shows proportions of a whole
Scatter plot
Reveals relationships between two variables
Continuous data
Can take any value within a range
Categorical data
Falls into distinct groups or categories
Put what you have learned to the test! Jump through the questions in game form.
Play GameYour Badges
0 of 6Mark lesson as complete
Tick when you've finished Learn, Practice and the game. Earns +85 XP and +25 coins.
Work through this topic 1-on-1 with an experienced HSC tutor.
Book a free session →