Measures of central tendency

By the end of this lecture we

should be able to
1. Compute & distinguished b/w the uses of measures of central
tendency: arithmetic mean, median, mode. Harmonic mean,
geometric mean
2. Compute & list uses for measures of dispersion: range,
variance, standard deviation. Mean deviation, I/Q range etc
3. Compute mean & standard deviation for grouped data.
4. Understand the distinction b/w population mean & sample
means.

2
 Mean (Average)
 Value about which observation tends to cluster

 Computed by addition of values (magnitude) of all

observation in the sample and divide the sum by the
number of observations.

3
Example
 Let us compute the mean (or average) of this
sample:

   110  118  110  122  110  150  120

= n 6

 In the above example, there are some new

mathematical notations.

4
Example (cont.)
   110  118  110  122  110  150  120
= n 6

 First, a symbol that denotes the mean of the sample.

5
Example (cont.)
   110  118  110  122  110  150  120
= n 6

 The second part of the equation shows how this

quantity is computed or the formula

6
Example (cont.)
   110  118  110  122  110  150  120
= n 6

 The sigma symbol summation (  ) tells us to sum all

the individual Xs.

7
Example 2.1 (cont.)
   110  118  110  122  110  150  120
= n 6

 Lastly, we must divide by ‘n’,

that is: the number of observations.

8
 Median
 Median is the middle value of the measurement when they
are arranged in ascending or descending order.
 Eg. Age of 5 students.
Unarranged – 1. 5. 3. 2. 4
Ascending order – 1, 2, 3, 4, 5
Median in this case is 3
 if the middle observations are 2, then median is average of
central 2 values.
E.g..1, 2, 3, 4, 5, 6
Median = 3+4 / 2 = 3.5 years

9
Mode
 observation occurring most frequently.
 30, 32, 35, 37, 41, 45, 41
 mode = 41 years
 a data may have more then one mode
 30, 32, 35, 37, 41, 45, 41, 32
 mode = 32 & 41

10
 Advantages and Disadvantages of each
measurement
Advantages Disadvantages

Mean 1. computation is easy 1. give impossible figure in discrete

2. result is firm, established 2. affected by extreme values
3. uses all the data 3. needs all the data for calculation.
4. Can be used in statistical work
Median 1. not affected by extreme values 1. cannot be used in further statistical
2. exact value can be computed work.
3. of immense importance in 2. it needs order to data.
experimental studies 3. does not comprise of all data

Mode 1. not affected by extreme 1. there may be more then one mode
values 2. does not represent all data
2. histograms can be obtained 3. not used in further statistical work.
3. easy to find & understand
11
  Is arithmetic mean alone enough to interpret the
data??
 NO, mean alone is misleading
 What about the following 2 data sets

data set 1 = 8, 8, 9, 10, 11, 12, 12

data set 2 = 1, 4, 8, 10, 12, 16, 19

12
Measurement of Dispersion
 Range
 Difference in value b/w maximum & minimum value.
 Range = X max – X min
 Quick to compute
 Not very useful. Does not take into consideration bulk
of values, only extreme values

13
Measurement of Variation
1. The Standard Deviation is a measure, which
describes how much individual measurements differ,
on the average, from the mean.
 A large standard deviation shows that there is a wide
scatter of measured values around the mean
 Small standard deviation shows that the individual
values are concentrated around the mean with little
variation among them.
 Standard deviation: it is the square root of the variance

14
STANDARD DEVIATION (calculation)

 ( Xi  X )
2
SD 
n 1

Data: 1,2,3,4,5

15
 Frequency table for calculating S.D.

xi x Xi - x (Xi – x)2 S (Xi – x)2/n-1

1 1-3 (-2) 2=4
2 2-3 (-1) 2 =1
3 3 3-3 (0) 2 =0
4 4-3 (1) 2 =1

5 5-3 (2) 2 =4
15 15/5=3 0 10 S2 = 10/4=2.5

Total S = √2.5 = 1.6

16
 Computing the sample variance

Important:
All sample variances are computed this way!
We always take the mean;
subtract each score from the mean;
square the result;
sum the squares;
and divide by the (sample size-1)
17
 Mean and S.D. from group data
 Formula for Mean
=S f Xi / n

Formula for variance

(X i  x )2
S2 = S f
n 1

18
 Example 1
1. Calculate the mean and standard deviation of the following
data set by making frequency table. (number of children in
different families)
2,2,5,3,0,1,3,2,3,4,0,3,4,5,7,3,2,4,1,0,5,8,6,5,4,2,4,4,7,6.
1. Also calculate X – S and X + S
2. Also calculate X – 2 S and X + 2 S
3. Also calculate X – 3 S and X + 3 S
 What percentage of measurements fall within each interval?

19
0,0,0,1,1,2,2,2,2,2,3,3,3,3,3,4,4,4,4,4,4,5,5,5,5,6,6,7,7,8
X = # of f = frequency (f * x) (X – )2 f (X – )2
children
- 0.72
0 3 0 (0 – 3.5)2= 12.25 12.25 * 3 = 36.75
1 2 2 (1 – 3.5)2 = 6.25 6.25 * 2 = 12.5
2 1.39 5 10 (2 – 3.5)2 = 2.25 2.25 * 5 = 11.25

3 5 15 (3 – 3.5)2 = 0.25 0.25 * 5 = 1.25

4 6 24 (4 – 3.5)2 = 0.25 0.25 * 6 = 1.5
5 4 20 (5 – 3.5)2 = 2.25 2.25 * 4 = 9.0

6
5.61 2 12 (6 – 3.5)2 = 6.25 6.25 * 2 = 12.5
7 2 14 (7 – 3.5)2 = 12.25 12.25 * 2 = 24.5
8 7.72 1 8 (8 – 3.5)2 = 20.25 20.25 * 1 = 20.25
Total = 30 105 129.5
20
Mean = 105/30 = 3.5
Mean and S.D. from group data (discreet data)

 Formula for Mean

=S f Xi / n
= 105 / 30 = 3.5
 Formula for variance
S2 = S f (Xi - )2 / n-1
= 129.5 / 30 – 1
= 129.5 / 29 = 4.46
Take square root of 4.46 to get S.D.
S.D. = 2.11
21
 Part 2
1. Also calculate X – S and X + S and What
percentage of measurements fall within this
interval?
X – S = 3.5 – 2.11 = 1.39
X + S = 3.5 + 2.11 = 5.61
Between 1.39 and 5.61 lies 20 values out of 30
So (20 / 30)* 100 = 66.6%
2. Also calculate X – 2 S and X + 2 S
3. Also calculate X – 3 S and X + 3 S

22
To calculate X – 2 S and X + 2 S

 X – (2 S) = 3.5 – (2*2.11) = 3.5 – 4.22 = - 0.72

 X + (2 S) = 3.5 + (2*2.11) = 3.5 + 4.22 = 7.72

Between - 0.72 (or zero) and 7.72 lies 29 values out of 30

So (29 / 30)* 100 = 96.6% of the total values lie between
these 2 values

23
Outliers
 In a normal distribution any measurement with a
Z-score of greater than 3 or less than -3 is called
an outlier.

24
Tutorial
 Find the mean, median, mode, range, variance, and
standard deviation for the data? 8, 5, 3, 5, 2, 1.
 Calculate the mean and standard deviation of the
following data set by making frequency table.
2,2,5,3,0,1,3,2,3,4,1,3,4,5,3,3,2,4,1,0,5,4,5,4,2,4,4,3,6.
 Also calculate X – S and X + S
 Also calculate X – 2 S and X + 2 S

 What percentage of measurements fall within each

interval?

25
 Home work 3 Number of babies
born per day
(days)

 Data presented in the following 0 6

table represent the number of 1 2
babies born per day in a maternity
2 9
home.
3 4
 Calculate the mean number of 4 5
babies born per day 5 3
 Calculate the range and standard 6 1
deviation of the number of babies
born per day in the month? Total 30
 Also calculate the values between
X±S 26
27
X = # of f = frequency (f * x) (X – )2 f (X – )2
babies
0 6 0 (0 – 2.43)2= 5.9 5.9 * 6 = 35.4

1 2 2 (1 – 2.43)2 = 2.04 2.04 * 2 = 4.08

2 9 18 (2 –2.43)2 = 0.18 0.18 * 9 = 1.62

3 4 12 (3 – 2.43)2 = 0.32 0.32 * 4 = 1.28

4 5 20 (4 – 2.43)2 = 2.46 2.46 * 5 = 12.3

5 3 15 (5 – 2.43)2 = 7.3 7.3 * 3 = 21.9

6 1 6 (6 – 2.43)2 = 12.74 12.74 * 1 = 12.74

30 days 73
MEAN  S f X / n
Mean = 73/30 = 2.43 89.32 28
Mean and S.D. from group data
 Formula for Mean
=S f Xi / n
= 73 / 30 = 2.43
 Formula for variance
S2 = S f (Xi - )2 / n-1
= 89.32 / 30 – 1
= 89.32 / 29 = 3.08
Take square root of 3.08 to get S.D.
S.D. = 1.75

29