Maths

Common mistakes in cumulative frequency and histograms

4 mistakes learners make with cumulative frequency and histograms, each one named and explained.

All common mistakes

Both graphs are read off the axis they were built from: the total frequency on one, and frequency density on the other.

Frequency plotted where frequency density belongs

What it looks like

The class widths are unequal and you draw the bar heights as the frequencies.

Why it happens

Every bar chart so far has used frequency for height, and the bars still look right.

A worked example

Question. A histogram has classes of different widths. What goes on the vertical axis?

A common answer. frequency

The answer. frequency density, which is frequency divided by class width

Why. With unequal widths it is the area of each bar that represents the frequency, not its height.

Put the two side by side

Bar chart: height is the frequency.
Histogram with unequal widths: area is the frequency.

Check the widths first. If they differ, divide before you plot.

Quartiles read from the top of the axis instead of the total frequency

What it looks like

The axis runs to 100 but there are 80 people, and you read the median across from 50.

Why it happens

The axis label is the number in front of you, and it is usually a round number.

A worked example

Question. A cumulative frequency graph has 80 people, on an axis drawn to 100. Where do you read the median?

A common answer. across from 50

The answer. across from 40

Why. The median is halfway through the data, so halve the total frequency, not the axis.

Put the two side by side

The axis is just how far the paper goes.
The total frequency is how many there are.

Find the top of the curve, not the top of the axis. Halve that.

Quartiles taken from an unordered list

What it looks like

You count to the middle of the numbers as they were written down and call it the median.

Why it happens

Counting to the middle is the method, and the list already looks like a list.

A worked example

Question. To find the median of a list of numbers, what has to happen first?

A common answer. nothing, count to the middle

The answer. put the list in order

Why. The median is the middle value by size, so the order they were written in does not matter.

Put the two side by side

Counting to the middle of any list gives a number.
Counting to the middle of a sorted list gives the median.

Sort first, every time. Then count.

The midpoint plotted on a cumulative frequency graph

What it looks like

You plot each running total against the middle of its class.

Why it happens

Midpoints are what you plot for an estimated mean and for a frequency polygon, so the habit carries over.

A worked example

Question. On a cumulative frequency graph, what do you plot the running total against?

A common answer. the midpoint of the class

The answer. the upper boundary of the class

Why. The running total counts everything up to the end of that class, so it belongs at the end.

Put the two side by side

Frequency polygon: plot the midpoint.
Cumulative frequency: plot the upper boundary.

Ask what the total includes. It includes everything up to the top of the class.

WAJD spots these patterns in your child's answers and names the one behind their wrong answers, instead of just marking them wrong.

More maths topics