Comparing data setsFinding the median from a frequency table

Data sets can be compared by looking at their similarities and differences. This can be done by first calculating an average and a measure of spread for reach.

Part ofApplications of MathsStatistics

Finding the median from a frequency table

Here are Kieran’s results in a frequency table

Number of tracks on albumFrequency
91
104
113
123
130
141
Number of tracks on album9
Frequency1
Number of tracks on album10
Frequency4
Number of tracks on album11
Frequency3
Number of tracks on album12
Frequency3
Number of tracks on album13
Frequency0
Number of tracks on album14
Frequency1

From the table, we can find the total number of albums

\(1 + 4 + 3 + 3 + 0 + 1 = 12\, albums\)

As the results are in a table, they are already ordered for us. In this case, as there are \(12\) results, the median is between the \(6th\) and \(7th\) result.

We look at the frequency column and determine when the cumulative frequency passes the \(6th\) and \(7th\) results.

(There is an explanation of Cumulative Frequency on a previous page in this section)

The first row has cumulative frequency \(1\)

The second row has cumulative frequency \(5\)

The third row has cumulative frequency \(8\)

The third row passes the \(6th\) and \(7th\) results, so the median must be in there.

Median is \(11\) tracks

In general, if there are \(n\) results, the median will be result \((n+1) \div 2\)

For example: for five numbers, the median is the \((5+1) \div 2 = 3rd\) result.

For six numbers, the median is the result between \(3rd\) and \(4th\) due to the fact that \((6 + 1) ÷ 2 = 3.5\).

Question

Find the median number of tracks on Suzie's albums.

Number of tracks on albumFrequency
63
70
82
91
103
114
122
Number of tracks on album6
Frequency3
Number of tracks on album7
Frequency0
Number of tracks on album8
Frequency2
Number of tracks on album9
Frequency1
Number of tracks on album10
Frequency3
Number of tracks on album11
Frequency4
Number of tracks on album12
Frequency2

Remember, when you are working out the median:

  • Put the results in numerical order (in a frequency table this will already be done)
  • Count the total amount of results and add one
  • Divide this by 2 to find the the position of the middle result
  • Find the middle result in the numerically ordered list or frequency table
  • You will then have the median of the set of results