CMV6120

Mathematics

Unit 20 : Measures of Central Tendency and Dispersion

Learning Objectives

The students should be able to: determine the mean, median and mode from ungrouped data determine the mean, median and mode from grouped data determine the range, inter quartile range and standard deviation.

Activities
Teacher demonstration and student hand-on exercise. Use MS Excel spreadsheet, internal functions and data analysis to measure central tendency and dispersion.

Reference
Suen, S.N. (1998) Mathematics for Hong Kong 5A; Rev. Ed.; Canotta

Unit 20 Central tendency & dispersion

Page 1 of 13

CMV6120

Mathematics

Measures of Central Tendency and Dispersion

1. Measure of central tendency: mean, median and mode from grouped and ungrouped data
For a set of data, we determine a quantity used to summarise the whole set of data. This quantity is termed a measure of central tendency. The most commonly used measures are mean, medium and mode.

1.1 mean
For ungrouped data, Mean( x) = Example 1 Find the mean for the set data: 3, 7, 2, 1, 7 Solution
x= 3 + 7 + 2 +1 + 7 5

x1 + x2 + x3 + ... + xn n

= For grouped data,

Mean( x) = x1 f1 + x2 f 2 + x3 f 3 + ... + xn f n f1 + f 2 + f 3 + ... + f n

Example 2 a) b) Find the mean of the set of data: 25, 36, 42, 38, 36 Find the mean from the set of grouped data Class mark Frequency Solution a) mean = 10.5 19 30.5 6 50.5 3 70.5 2 90.5 1 110.5 2

Unit 20 Central tendency & dispersion

Page 2 of 13

CMV6120

Mathematics

b) x 10.5 30.5 50.5 70.5 90.5 110.5 sum mean = Example 3 The HK Consumer Price Index B from 1997 to 2003 was as following: 1996 99.7 1997 105.5 1998 108.5 1999 103.4 2000 99.4 2001 97.7 2002 94.7 2003 92.1 Calculate the average consumer price index B: a) For the first 4 years, (1997 2000). b) For the next 3 years, (2001 2003) c) For all 7 years d) Suppose the original data was lost, and only the 4- and 3-year averages in a) and b) were available. Would it still be possible to calculate the overall 7-year average? How? Solution a) From 1997 2000, n = 4. The average price index = (105.5 + 108.5 + 103.4 + 99.4) 4 = b) From 2001 2003, n = 3. The average price index = (97.7 + 94.7 + 92.1) 3 = c) From 1997 2003, n = 7. The average price index = (105.5 + 108.5 + 103.4 + 99.4 + 97.7 + 94.7 + 92.1) 7 = d) The average price index over 7 years = ( = 4+ 3 ) (4 + 3) f 19 6 3 2 1 2 33 xf 199.5

Unit 20 Central tendency & dispersion

Page 3 of 13

CMV6120

Mathematics

1.2 Median
For ungrouped data, Median = the middle datum, when n is odd. Median = the mean of the two middle data, when n is even.

e.g.1

For the set of data 2, 4, 7, 9, 21 middle datum median = 7

e.g.2

For the set of data 3, 5, 7, 7

middle of two data median = (5 + 7) 2 =6

For grouped data, Step 1: Step 2: Example 4 a) Find the median of 2, 3, 10, 12, 999. b) Find the median of 2, 3, 10, 12, 22, 123. c) The cumulative frequency polygon for maths marks of a class is given below, find the median mark. Solution a) Median = b) Median = c) Total frequency = 40 The rank of median = 40/2 =
frequency

Draw the cumulative frequency polygon. The median is the datum corresponding to the middle value of the cumulative frequency.

Cumulative frequency polygon for marks in maths

40 35 30 25 20 15 10 5

Unit 20 Central tendency & dispersion

Page 4 of 13

89 .5 79 .5 69 .5 59 .5 49 .5 39 .5 29 .5 19 .5 9. 5

99 .5

Marks

CMV6120

Mathematics

From the cumulative polygon, median = Example 5 The provisional figures on the population by age group in Hong Kong as at 9/2001 are tabulated below. Draw a cumulative frequency polygon and determine the median age for the population. Age group > 70 0 9 10 19 20 29 30 39 40 49 50 59 60 69 Population 676 885 1208 677 503 499 1000 1267 (000) Solution
Age group x x < 10 10 x < 20 20 x < 30 30 x < 40 40 x < 50 50 x < 60 60 x < 70 70 < x Population ('000) 676 885 1000 1267 1208 677 503 499 cumulative population (000 000) 0.676 1.561

Less than Cumulative frequency polygon for population

8.0 7.0 Population (000 000) 6.0 5.0 4.0 3.0 2.0 1.0

The rank of median = 6.715/2 = The median age =

30

20

70 + 70

40

10

Age group

50

60

1.3 mode
For ungrouped data, mode is the datum that has the highest frequency. For grouped data, modal class is the class that has the highest frequency. Example 6 a) Find the mode of the data: 1, 2, 2, 2, 3, 3, 9 b) Find the modal class Class Frequency Solution a) The mode is 10 - 14 2 15 - 19 8 20 - 24 7 25 - 29 3

Unit 20 Central tendency & dispersion

Page 5 of 13

CMV6120

Mathematics

b) The modal class is Example 7 The temperature in degree Celsius each day cover a three week period were follow: 17, 18, 20, 21, 19, 16, 15, 18, 20, 21, 21,,22, 21, 19, 20,19, 17,16,16,17. Compute the mean, median, and mode of these raw dates by using two-degree intervals starting with 15-16. Draw a cumulative frequency polygon. Solution
Temperature () 15 16 17 18 19 20 21 22 23 24 Sum Temperature cumulative () frquency < 14.5 < 16.5 < 18.5 < 20.5 < 22.5 < 24.5 Tally Frequency Class mark f x fx

Cumulative frequency polygon for temperature

24 20 Frequency 16 12 8 4 0

24 .5 22 .5 20 .5 18 .5 16 .5 14 .5
Temperature ()

The mean temperature = The modal class of temperature is The median temperature is

Remark Mean seems to be the most commonly used (and often misused) quantity for measuring central tendency. If the distribution of the data set shows a strong degree of skewness, then mean is not a reliable measure as it is strongly affected by the extreme values. In this case, medium may be a better choice. Mode is used when there is reason to choose the most commonly occurring data value as the representative for the whole data set.

Unit 20 Central tendency & dispersion

Page 6 of 13

CMV6120

Mathematics

2. Measure of dispersion: Range, Inter-quartile range and Standard deviation

Apart from using a measure of central tendency to summarise a set of data, we need a quantity to measure the degree of dispersion of the set of data (so that we can determine the reliability of the set of data). Range is a measure that is very simple to use but it provides relatively little information on dispersion. Quartile deviation is used in association with the median whereas standard deviation goes with the mean. 2.1 Range For ungrouped data, the range is the difference between the largest datum and the smallest datum. For grouped data, the range is the difference between the highest class boundary and the lowest boundary. Example 8 a) Find the range of the data: 1, 2, 2, 2, 3, 3, 9 b) Find the range of the grouped data Class Frequency Solution a) The range = b) The range = 10 - 14 2 15 - 19 8 20 - 24 7 25 - 29 3

2.2 Inter quartile range

Inter quartile range = Q3 Q1 where Q1, Q2, Q3 are called quartiles which divide the data (which have been ranked, i.e. arranged in order) into four equal parts. Moreover, Q2 is the median of the whole set of data, Q1 is the median of the lower half, Q3 is the median of the upper half. Quartile deviation, Q.D. = (Q3 Q1)

Unit 20 Central tendency & dispersion

Page 7 of 13

CMV6120

Mathematics

Example 9 Find the inter quartile range of a) 1, 2, 3, 5, 11, 12, 13. b) 1, 2, 3, 4, 11, 12, 13, 14. Solution a) inter-quartile range = b) inter-quartile range = Example 10 The following frequency distribution gives the life hours of a sample of 50 light bulbs: Life hours (000) Frequency 0.6 to under 0.7 0.7 to under 0.8 0.8 to under 0.9 0.9 to under 1.0 1.0 to under 1.1 1.1 to under 1.2 1.2 to under 1.3 2 4 6 14 13 7 4 Life hours Cumulative Up to (000) frequency 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3

Find the median and the inter-quartile range of the data. The rank of median = 50 = The median of life hours is
Less than cumulative frequency polygon for life hours of 50 sample light bulbs

hrs.
56 Frequency 48 40 32 24 16 8 0

The rank of upper quartile = 50 = = 38 , to the nearest integer The upper quartile Q3 is hrs The rank of lower quartile = 50 = = 13 , to the nearest integer The lower quartile Q1 is hrs. The inter-quartile range = Q3 Q1 = Quartile deviation = (Q3 Q1) =

1. 3 1. 2 1. 1 1. 0 0. 9 0. 8 0. 7 0. 6
Life hour (000)

Unit 20 Central tendency & dispersion

Page 8 of 13

CMV6120

Mathematics

Example 11 Find the range, inter-quartile range and quartile deviation for the data in example 4 and example 7 respectively.

2.3 Standard deviation

For ungrouped data x1, x2,,xn, with a mean
x

, the standard deviation () is

( x1 x ) 2 + ( x 2 x ) 2 + ( x 3 x ) 2 + ... + ( x n x ) 2 n

For grouped data with class marks x1, x2,,xn; corresponding frequencies f1,f2,,fn, and a mean the standard deviation () is
( x 1 x ) 2 f 1 + ( x 2 x ) 2 f 2 + ( x 3 x ) 2 f 3 + ... + ( x n x ) 2 f n f 1 + f 2 + f 3 + ... + f n

Example 12 Find the standard deviation for a) the ungrouped data 8, 9, 10, 10, 11 b) the grouped data x f Solution a) mean x =
i

17 2

22 4

27 7

32 8

37 7

42 4

47 2

xi = n

standard deviation =

( xi x ) 2
i

Calculator key-in method: Model 3600 Set Statistic mode MODE 3 Clear memory Key-in data mean s.d.
KAC 8 DATA 10 DATA 11 DATA SHIFT 1 SHIFT 2

Model 3900
MODE 2 KAC 8 DATA 9 DATA 10 DATA 10 DATA 11 DATA SHIFT 4 SHIFT 5

Model 506R
2ndF MODE 3 0 2ndF CA 8 DATA 9 DATA 10 DATA 10 DATA 11 DATA 2ndF 4 2ndF 6

9 DATA 10 DATA

Unit 20 Central tendency & dispersion

Page 9 of 13

CMV6120

Mathematics

b)

mean x =
i

xi = n

standard deviation =

( xi x ) 2
i

Calculator key-in method: Model 3600 Set Statistic mode MODE 3 Clear memory Key-in data
KAC 17 2 22 4 27 7 32 8 37 7 32 4 47 2 SHIFT SHIFT DATA DATA DATA DATA DATA DATA DATA 1 2

Model 3900
MODE 2 KAC 17 2 22 4 27 7 32 8 37 7 32 4 47 2 SHIFT SHIFT DATA DATA DATA DATA DATA DATA DATA 4 5

Model 506R
2ndF MODE 3 0 2ndF CA 17 2 DATA 22 4 DATA 27 7 DATA 32 8 DATA 37 7 DATA 42 4 DATA 47 2 DATA 2ndF 4 2ndF 6

mean s.d. Example 13

The life hours of 50 light bulbs has the following frequency distribution. Complete the table with class marks. Calculate the mean and standard deviation. Life hours (000) 0.6 to under 0.7 0.7 to under 0.8 0.8 to under 0.9 0.9 to under 1.0 1.0 to under 1.1 1.1 to under 1.2 1.2 to under 1.3 Solution
xi mean x = = i n

Class mark 0.65 0.75 0.85 0.95 1.05 1.15 1.25

Frequency 2 4 6 14 13 7 4

standard deviation =

( xi x ) 2
i

Unit 20 Central tendency & dispersion

Page 10 of 13

CMV6120

Mathematics

Example 14 The height of Basil team members at the 2002 FIFA World Cup is listed as following: Marcos 1.93 Cafu 1.76 Lucio 1.88 Roque Junior 1.86 Edmilson 1.85 Carlos 1.68 Richardino 1.76 Silva 1.85 Ronaldo 1.83 Rivaldo 1.86 Ronaldinho 1.80 Calculate the average height, and the standard deviation: Solution

Example 15 Find the mean and standard deviation for the data given below: Age group x 5 15 25 35 45 55 65 75 Solution Population ('000) 676 885 1000 1267 1208 677 503 499

Unit 20 Central tendency & dispersion

Page 11 of 13

CMV6120

Mathematics

Practice
1. The Hong Kong unemployment rate in the year of 4/2003 4/2004 was as following: 5/2003 8.3 6/2003 8.6 7/2003 8.7 8/2003 8.6 9/2003 8.3 10/2003 8.0 11/2003 7.5 12/2003 7.3 1/2004 7.3 2/2004 7.2 3/2004 7.2 4/2004 7.1 5/2004 7.0 Calculate the average, median, mode and the standard deviation of unemployment rate: a) For 5/2003 12/2003 b) For1/2004 5/2004 c) For all 13 months. Find the mean, median, mode of the following: 10, 13, 14, 14, 14, 15, 15,16, 17, 22 Which student has the highest average mark? Student English Chinese Mathematics 4. A 78 80 59 B 63 85 71 C 55 72 95

2.

3.

The frequency distribution of the lengths of 100 leaves from a certain species of plant is given below: length (mm) 20 24 25 29 30 34 35 39 40 44 45 49 50 54 Frequency 6 10 18 25 22 15 4

Unit 20 Central tendency & dispersion

Page 12 of 13

CMV6120

Mathematics

5.

The following table shows the distribution of heights of 50 students: Height (cm) 160 164 165 169 170 174 175 179 180 184 185 189 Frequency 8 12 14 7 6 3

Find the range and standard deviation of heights. 6. 7. 8. 9. The mean of one set of six numbers is 9 and the mean of a second set of eight numbers is 12.5. Calculate the mean of the combined set of fourteen numbers. The mean of the numbers a, b, c, d is 8 and the mean of the numbers a, b, c, d, e, f, g is 11. What is the mean of the numbers e, f, g? Find the mean and standard deviation of the 5 numbers in term of x: x5, x-3, x2, x+1, x+4. The mean of the five numbers 6, 9, 2, x, y is 5 and the standard deviation is values of x and y.
6

. Find the

Answer: 1.a) 8.16: 8.3; undefined; 0.52 2) 15: 4.5; 4 3) 72; 73; 74 8) x1; 3.16 9) (3, 5); (5, 3)

b) 7.16: 7.2; 7.2; 0.11 c) 7.78: 7.5; undefined; 0.65 4) 37.4 5) 30; 7.14 6) 11 7) 15

Unit 20 Central tendency & dispersion

Page 13 of 13