Single-variable data: diagrams and histograms

Single-variable data: diagrams and histograms

Spec 2.1 · Edexcel A-Level Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • In a histogram the area of each bar is the frequency; frequency density =frequencyclass width=\dfrac{\text{frequency}}{\text{class width}}.
  • A frequency polygon joins points plotted at the class midpoints.
  • Cumulative frequency is plotted at the upper class boundary; read the median at n2\frac n2 and the quartiles at n4\frac n4 and 3n4\frac{3n}{4}.
  • Outliers lie below Q1−1.5×IQRQ_1-1.5\times\text{IQR} or above Q3+1.5×IQRQ_3+1.5\times\text{IQR}.
  • On a probability histogram the area of a bar is the relative frequency.

Histograms and frequency density

The area of each bar is the frequency, so the vertical axis is frequency density.

A histogram is for continuous grouped data, where class width can vary. There are no gaps between bars.

Rearranging, frequency == frequency density ×\times class width. If one bar’s height and frequency are known, the scale for all other bars follows from the ratio of areas. To estimate the frequency in part of a class, assume the data are spread evenly.

width 2width 4width 2f = 6f = 16f = 4ABCDEFGHIJ
Illustrative: bar area = frequency. Widths 2, 4, 2 with frequency densities 3, 4, 2 give frequencies 6, 16 and 4.
  • Frequency densityfrequencyclass width\dfrac{\text{frequency}}{\text{class width}}
  • Frequencydensity ×\times width

Worked example

A class 10≤t<2010\le t<20 has frequency 24. Find its frequency density, then the frequency of a class of width 5 and frequency density 3.

A bar has width 4 and height 2.5 on a frequency density axis. What is its frequency?

Frequency polygons

Plot the frequency at the midpoint of each class and join the points with straight lines.

A frequency polygon shows the shape of the distribution and is useful for comparing two data sets on the same axes. The midpoint of the class 20≤x<3020\le x<30 is 2525.

  1. 1

    Find midpoints

    Add the class boundaries and halve.

  2. 2

    Plot

    Frequency against midpoint.

  3. 3

    Join

    Join the points with straight lines.

Drawing a frequency polygon

At which xx-value is the point for the class 40≤x<6040\le x<60 plotted on a frequency polygon?

Cumulative frequency diagrams

Plot running totals at the upper class boundaries, then read values at n2\frac n2, n4\frac n4 and 3n4\frac{3n}{4}.

Cumulative frequency is the running total, plotted at the upper class boundary. The median is at n2\frac n2, Q1Q_1 at n4\frac n4 and Q3Q_3 at 3n4\frac{3n}{4}. The interquartile range is Q3−Q1Q_3-Q_1.

Within a class use linear interpolation: L+position−Fbelowf×wL+\frac{\text{position}-F_{\text{below}}}{f}\times w, with LL the lower boundary, ff the class frequency and ww its width.

0510152025301020304050Upper class boundaryCumulative frequency
Illustrative values for n = 30: read the median at the 15th value, Q₁ at the 7.5th and Q₃ at the 22.5th.
  • EstimateL+position−Fbelowf×wL+\dfrac{\text{position}-F_{\text{below}}}{f}\times w

Worked example

For n=100n=100 plants, 8 are under 10 cm and 30 are under 20 cm. Estimate Q1Q_1.

For n=80n=80, at which cumulative frequency do you read off the upper quartile?

Box and whisker plots and outliers

A box plot shows the minimum, quartiles, median and maximum, with outliers plotted separately.

A box and whisker plot has a box covering the interquartile range. Values beyond the fences are outliers: below Q1−1.5×IQRQ_1-1.5\times\text{IQR} or above Q3+1.5×IQRQ_3+1.5\times\text{IQR}. Whiskers stop at the most extreme values that are not outliers.

To compare two box plots, compare a measure of average (median) and a measure of spread (IQR or range), in context.

051015202530354045503.5Q1 = 17Q3 = 2639.545

Worked example

Marks have Q1=17Q_1=17 and Q3=26Q_3=26. Is a mark of 45 an outlier?

Which pair of measures should you use to compare two box plots?

Connection to probability

On a relative frequency histogram the total area is 1 and each bar’s area is a probability estimate.

Dividing each frequency by the total gives relative frequency. Dividing by class width as well gives a histogram with total area 1, where the area of a bar estimates the probability of that class.

As classes get narrower and the data set larger, the outline approaches a smooth probability density function. Probability is then the area under the curve.

−4−3−2−112340.10.20.30.40.5xyprobability density
As classes narrow, the histogram outline approaches a smooth curve. The area under it is a probability; here the shaded area is about 0.68.
  • Area of a barrelative frequency
  • Total area11

Worked example

A class contains 60 of 200 eggs. What is the area of its bar on a probability histogram?

What is the total area of a relative frequency density histogram?

Try an exam question

The quartiles of a data set are Q1=20Q_1=20 and Q3=32Q_3=32. Use the 1.5×IQR1.5\times\text{IQR} rule to decide whether a value of 55 is an outlier.

[3 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

How many calls lasted between 2 and 6 minutes?

1 mark, marked against the mark scheme

Revision notes