Measures of central tendency are used to determine the central values of a data set. Measurement commonly used for measures of central tendency are the mode, median and mean Mode (Ungrouped data) The mode of a set of data is the value that occurs the most number of times in the set. For the data set 10, 11, 12, 13, 14, 14, 15, 16, the mode is 14 because 14 is the value that occurs most frequently. A data set that has one value that occurs with the highest frequency is said to be unimodal. For the data set 5, 5, 5, 6, 7, 8, 9, 9, 9, 10, two values 5 and 9 both occur with the same highest frequency) which is three times. Hence the modes for this data set are 5 and 9. This data set is said to be bimodal. All sets of numerical data have a mean and a median but not all of them have modes. For example, the data set 1, 2, 4, 5, 6, 8, 9, 11 has no mode. Mode is a suitable and useful measure of central tendency in the shoe business. If the manager of a shoe shop wants to stock the shoes which are most saleable, he needs to know the mode so that sufficient orders of the desired size are made. Median (Ungrouped data) Median is the value at the centre of a data set after the data set is arranged in ascending or descending order. Find the median for each set of data shown below. (a) 75, 67, 48, 66, 89, 51, 70 (b) 75, 67, 48, 66, 189, 51, 70 (a) Arrange the numbers in ascending order, 48, 51, 66, 67, 70, 75, 89 3 numbers 3 numbers Since 67 is the middle number after the numbers are arranged in ascending order, 67 is the median. (b) Arrange the numbers in ascending order, 48, 51, 66, [67], 70, 75, 189
Therefore, 67 is the median for this set of numbers.
For data which has even number of observations, the median cannot be obtained directly because there are no values in the centre. For this case, we obtain the median by finding the average of two observations which are nearest to the centre.
Median (Grouped data)
For a grouped frequency distribution, the median is the (^)th observation. An estimate of the median may be obtained. (a) by calculation, (b) using a cumulative frequency curve.