Lesson 4 of 8 · 16 min
Column charts, histograms and choosing a display
Build frequency tables, column charts for discrete data and histograms for grouped data, then select and justify the right display for any numerical data set.
What you will be able to do
- Organise discrete data in a frequency table and display it in a column chart.
- Group continuous data into equal class intervals such as 20–<30, and construct a histogram.
- Identify the modal class of a histogram.
- Select and justify an appropriate display for a numerical data set.
A dot plot of 400 students' travel times would be a blur of dots, and a stem-and-leaf plot would need rows with dozens of leaves. For larger data sets, you count how many values fall in each group and display the counts.
For discrete data with only a few different values, count each value. A traffic survey on Brisbane's Gateway Bridge recorded the number of people in each of 40 cars.
| Number of people | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| Frequency | 18 | 12 | 5 | 4 | 1 |
A column chart of discrete data. Each column sits over one value, with gaps between them.
Note
Why the gaps?
A car can carry 2 or 3 people but not 2.5, so the values are separate points on the number line. The columns stand over those points with gaps between them, just like a column chart of categories.
Continuous data, or discrete data with many different values, is grouped into class intervals of equal width. The class 20–<30 means "from 20 up to but not including 30", so every value belongs to exactly one class. Five to ten classes usually show the shape well.
Here are the travel times of 40 students, grouped into classes 10 minutes wide.
| Travel time (minutes) | Frequency |
|---|---|
| 0–<10 | 6 |
| 10–<20 | 13 |
| 20–<30 | 10 |
| 30–<40 | 6 |
| 40–<50 | 3 |
| 50–<60 | 2 |
| Total | 40 |
Each column covers a class 10 minutes wide, and its height is the number of students in that class. The columns touch because time is continuous.
Definition
Histogram
A histogram displays grouped numerical data on a continuous scale. Each column covers one class interval, the columns touch, and the height of each column is the frequency of its class. The class with the tallest column is the modal class: for the travel times it is 10–<20 minutes.
A student's travel time is exactly 30 minutes. In the grouped frequency table, which class does it belong to?
Worked example
Group data and construct a histogram
The heights, in centimetres, of 30 Year 11 students are:
165, 172, 158, 181, 170, 168, 176, 163, 189, 171, 160, 175, 184, 167, 179, 152, 173, 168, 186, 171, 161, 192, 175, 166, 180, 174, 169, 183, 164, 178
Organise the heights into a grouped frequency table with classes 150–<160, 160–<170, and so on. Then construct a histogram and state the modal class.
The finished histogram.
Step 1: Choose the classes
The smallest height is 152 cm and the largest is 192 cm, so five classes of width 10 cm cover them: 150–<160 up to 190–<200.
Five to ten equal classes usually show the shape well.
Step 2: Count the values in each class
Work through the list, tallying each height into its class.
Height (cm) 150–<160 160–<170 170–<180 180–<190 190–<200 Total Frequency 2 10 11 6 1 30 The total must equal the number of students, which checks that no value was missed.
Step 3: Draw the histogram
Put height on the horizontal axis with a continuous scale and frequency on the vertical axis from zero. Draw one column per class, touching its neighbours, with height equal to the frequency. Label both axes and add a title.
Step 4: State the modal class
The tallest column is 170–<180 cm, with 11 students.
Try it yourself first, then check the working one step at a time.
Answer
Frequencies 2, 10, 11, 6 and 1 for the classes 150–<160 to 190–<200 cm, drawn as the histogram shown. The modal class is 170–<180 cm.
Method
Choose equal, non-overlapping classes that cover the smallest to the largest value.
Tally the values into the classes and check that the total is .
Draw touching columns on a continuous scale, with labelled axes.
The tallest column is the modal class.
Watch out for
Putting a value on a class boundary in the wrong class: 170 belongs to 170–<180.
Leaving gaps between the columns.
You have now met four displays for numerical data. The syllabus asks you to select, construct and justify one, so the choice needs reasons. Two questions decide it: what type of data is it, and how many values are there?
| Data | Good display | Why |
|---|---|---|
| Categorical | Column or bar chart; pie chart | Categories are separate groups; a pie chart shows parts of a whole |
| Numerical, small set, few different values | Dot plot; column chart | Shows the frequency of every value |
| Numerical, small to medium set, two-digit values | Stem-and-leaf plot | Shows the shape and keeps every value |
| Numerical, large set or many different values | Histogram | Grouping into equal classes shows the shape |
Worked example
Select and justify a display for four data sets
For each data set, select an appropriate display and justify your choice.
- (a) The number of pets owned by each of 25 students (from 0 to 4).
- (b) The reaction times, in milliseconds, of 300 drivers in a driving-simulator study.
- (c) The scores, out of 50, of 24 students on a test.
- (d) The favourite streaming service of 150 students.
Step 1: (a) Number of pets
Numerical, discrete, with only five possible values and 25 students. A dot plot (or a column chart) shows how many students have each number of pets.
Step 2: (b) Reaction times
Numerical, continuous, with 300 values that are nearly all different. A histogram with equal classes (for example 50 ms wide) shows the shape; a dot plot or stem-and-leaf plot would be far too crowded.
Step 3: (c) Test scores
Numerical, two-digit values, and a medium-sized set of 24. A stem-and-leaf plot shows the shape and keeps every score, so the teacher can still find the median and quartiles.
Step 4: (d) Streaming service
Categorical, nominal. A column or bar chart (or a pie chart, if there are only a few services) suits categories. None of the numerical displays applies.
Classify the variable first. It rules out half the displays straight away.
Try it yourself first, then check the working one step at a time.
Answer
(a) Dot plot or column chart. (b) Histogram. (c) Stem-and-leaf plot. (d) Column or bar chart, or a pie chart. Each choice is justified by the type of data and the size of the data set.
Method
Classify the variable.
Consider how many values there are, and how many different values.
Name the display and justify it with the type of data, the size of the data set and what the display shows.
Watch out for
Choosing a histogram for categorical data.
Justifying with "it's easy to read" instead of the type and size of the data.
The number of siblings of each of 25 students is recorded. The values range from 0 to 5. Which display is most suitable?
Common mistake
Histogram errors that cost marks
- Gaps between the columns. A histogram's scale is continuous, so neighbouring classes touch.
- Unequal class widths. Keep every class the same width so the heights compare fairly.
- Ambiguous classes like 10–20 and 20–30. Which one gets 20? Write 10–<20 and 20–<30.
- A histogram of categorical data. Categories need a column or bar chart with gaps.
Exam tip
Select and justify
A full answer names the display and gives reasons from the data: the type of variable, the size of the data set, and what the display will show. For example: "A histogram, because reaction time is continuous and there are 300 values; grouping them into equal classes shows the shape of the distribution."
Summary
Discrete data with few values: a frequency table and a column chart, with separate columns over each value.
Continuous data, or data with many different values: group into equal, non-overlapping classes such as 20–<30 and draw a histogram.
In a histogram the columns touch, and the class with the tallest column is the modal class.
Choose a display from the type of data and the size of the data set, and justify your choice with both.
Finished the lesson?
Try the 2 quick checks above and the lesson completes itself, or mark it complete now.