CHAPTER THREE

STATISTICAL
DESCRIPTION OF DATA
 3.1 Measures of Central Tendency
Introduction

• A measure of central tendency is a descriptive statistic that describes the

average, or typical value of a set of scores.

• It is also defined as a single value that is used to describe “center” of the

data

Typical value
(Center of data)

2
 Introduction

• Good properties of typical average

– Computation should be based on all the observed values.

– It should be simple to understand and easy to interpret.

– As little as affected by fluctuations of sampling.

– should not unduly be influenced by extreme values.

– it should be defined rigidly which means that it should have a definite value

• There are three common measures of central tendency

– Mean

– Median

– Mode

3
 The Summation Notation

• Also called Sigma notation

• Sigma is a Greek letter ∑ meaning “sum”

• Let X is a variable

n ending point/

X
Upper limit of
the summation
i
i 1
Summation
notation
Xi is the index of
summation, each
starting point/
term of the sum
Lower limit of
the summation
(index of the
summation)
4
 The Summation Notation..

• Properties of summation notation

n

X
i 1
i  X1  X 2    X n

XY
i 1
i i  X 1Y1  X 2Y2    X nYn

 i 1 2
X 2

i 1
 X 2
 X 2
  X 2
n

n n

 CX
i 1
i  C  X i  CX 1  CX 2    CX n
i 1

5
 The Mean

• Mean is the most commonly used measure of central tendency. There are
different types of mean
– Arithmetic mean,

– Weighted mean,

– Geometric mean (GM) and

– Harmonic mean (HM)

• If mentioned without an adjective (as mean), it generally refers to the

arithmetic mean.

6
 The Arithmetic Mean

• It is computed by adding all the values in the data set divided by the number of
observations in it.

• If we have the raw data, mean is given by the formula

n

X i
X i 1
n

• If we have frequency distribution (ungrouped) mean is given by the formula

n

fX i i
X i 1
n
• If we have frequency distribution (grouped) mean is given by the formula
n LCB/UCB is lower/upper class boundary
fm i i
LCBi  UCBi
X i 1
, where mi 
n 2
7
 The Arithmetic Mean …

• Example 1: The following data is the weight (in Kg) of eight youths:
32,37,41,39,36,43,48 and 36. Calculate the arithmetic mean of their weight.
(Ans:312/8=39 )

• Example 2: The ages of a random sample of patients in a given hospital in Ethiopia is

given below: (Ans: 16.075)

Age (xi) Number of patients (fi)

10 3
12 6
14 10
16 14
18 11
20 5
22 4
8
 The Arithmetic Mean …

• Example 3: Age in year of 20 women who attended health education at Jimma Health
center in 1986 is summarized in the table. What is the mean age of these women. (Ans:
670/20=33.5)

Time (in seconds) Number of students

23-26 3
27-30 4
31-34 3
35-38 5
39-42 5

9
 Properties of Arithmetic Mean …

• It can be computed for any set of numerical data, it always exists, and unique.

• It depends on all observations.

• The sum of deviations of the observations about the mean is zero i.e.

• It is greatly affected by extreme values.

• It lends itself to further statistical treatment, for instance, combinations of means.

• It is relatively reliable, i.e. it is not greatly affected by fluctuations in sampling.

• The sum of squares of deviations of all observations about the mean is the minimum

• If a constant is added to all observations, the new mean is old mean plus constant

• If all observations are multiplied by a constant, the new mean is the multiple of the constant and old
mean

• If wrong value is recorded and latter on it is discovered, the new corrected mean is

X corr X wrong
X corr  X wrong 
n
10
 Weighted Mean

• Weighted mean is calculated when certain values in a data set are more
important than the others.

• A weight wi is attached to each of the values xi to reflect this importance.

• The weighted mean is computed as

k

w x i i
Xw  i 1
k

w
i 1
i

• Example: CGPA of a students (each result is weighted by credit of a course) [Ans:

2.88]

11
 Geometric Mean

• It is defined as the arithmetic mean of the values taken on a log scale.

• It is also expressed as the nth root of the product of an observation.

• GM is an appropriate measure when values change exponentially and in case of

skewed distribution that can be made symmetrical by a log transformation.

• Note: The geometric mean is useful in finding the average of percentages,

ratios, indexes, or growth rates.

• One important disadvantage of GM is that it cannot be used if any of the values

are zero or negative.
12
 Geometric Mean…

Example 1:- The G.M of 4, 8 and 6 is.

Solution:

Example 2: The man gets three annual raises in his salary. At the end of first year,
he gets an increase of 4%, at the end of the second year, he gets an increase of 6%
and at the end of the third year, he gets an increase of 9% of his salary. What is the
average percentage increase in the three periods?

Solution:

13
 Properties of geometric mean
– Its calculations are not as such easy.
– It involves all observations during computation
– It may not be defined even it a single observation
is negative.
– If the value of one observation is zero its values
becomes zero.
 Harmonic Mean

• Another important mean is the harmonic mean, which is suitable measure of

central tendency when the data pertains to speed, rates and price.

• It is the reciprocal of the arithmetic mean of the observations.

• Let be n variant values in a set of observations, then simple

harmonic mean is given by:

• Note: SHM is used for equal distances, equal costs and equal rates.

15
 Harmonic Mean

Example 1: A motorist travels for three days at a rate (speed) of 480 km/day. On
the first day he travels 10 hours at a rate of 48 km/h, on the second day 12 hours at
a rate of 40 km/h, on the third day 15 hours at a rate of 32 km/h. What is the
average speed?
Solution: Since the distance covered by the motorist is equal
( ), so we use SHM.

so the required average speed = 38.92 km/hr

We can check this, by using the known formula for average speed in elementary
physics.
Check;
=
=
16
 Weighted harmonic mean (WHM)
• WHM is used for different distance, different cost and different
rate.

Example 1: A driver travel for 3 days. On the 1st day he drives for
10h at a speed of 48 km/h, on the 2nd day for 12h at 45 km/h and
on the 3rd day for 15h at 40 km/h. What is the average speed?
Solution: since the distance covered by the driver is not equal, so
we use WHM by taking the distance as weights (wi).
 Properties of harmonic mean
• It is based on all observation in a distribution.
• Used when a situations where small weight is
give for larger observation and larger weight
for smaller observation
• Difficult to calculate and understand
• Appropriate measure of central tendency in
situations where data is in ratio, speed or rate.
 Relation between AM, GM, and Hm

• If all the values in a data set are the same, then all the three means (arithmetic
mean, GM and HM) will be identical.

• As the variability in the data increases, the difference among these means also
increases.

• Arithmetic mean is always greater than the GM, which in turn is always greater
than the HM.
– AM > GM > HM

19
 Median

• If the sample data are arranged in increasing order, the median is

– if n is an odd number, median is middle value

• Example: systolic blood pressure of seven persons were given as 113, 124, 124,
132, 146, 151, and 170. what is the median systolic blood pressure? (Ans: 132)

– if n is an even number, midway between the two middle values

• Six men with high cholesterol participated in a study to investigate the effects of diet
on cholesterol level. At the beginning of the study, their cholesterol levels (mg/dL)
were as follows:366, 327, 274, 292, 274 and 230. what is the median cholesterol
level? (Ans:283)
20
 Median …

– If the data is in ungrouped frequency distribution, median is the class with largest
less than cumulative frequency smaller than or equal to half of the total observation
• Example: Forty five students were taken to field and evaluated their performance using 60m
pure speed test. The time is recorded in seconds, and the result is summarized in the table. What
is the median performance of these students. (Ans: 19 secs)

Time (in Number of Less than

seconds) students cumulative
frequency
15 4 4
16 9 13
18 8 21
19 14 35
20 10 45

21
 Median …

– If the data is in grouped frequency distribution, median is

• Example: fifty students were taken to field and evaluated their performance using 100 m
pure speed test. The time is recorded in seconds, and the result is summarized in the table.
What is the median performance of these students. (Ans: 20.81 secs)

Time (in seconds) Number of students

14-16 6
17-19 12
20-22 16
23-25 9
26-28 7
22
 Mode

• The most frequent observation (value) in a data

• An observation with the largest frequency

• There can be no mode Eg: 25, 27, 22, 18

• There can be only one mode-unimodal Eg: 25, 27, 22, 25,18

• There can be two mode-bimodal Eg: 25, 27, 22, 27, 25, 18, 20

• There can be more than two mode-multimodal Eg: 25, 27, 22, 27, 25, 18, 20, 19, 22, 17

• Mode grouped frequency distribution

• f1 = frequency of the modal class

• f0 = frequency of the class preceding the modal class
• f2 = frequency of the class next to the modal class

23
 Mode…

• The most frequent observation (value) in a data

– Example: Twenty five amateur cyclists were taken to field and their time is
recorded to complete a given distance. The time is recorded in seconds, and
the result is summarized in the table. What is the modal time to complete the
distance. (Ans: 29.5 secs)

Time (in seconds) Number of

Atheletes
15.5- 21.5 3
21.5-27.5 6
27.5-33.5 8
33.5-39.5 4
39.5-45.5 3
45.5-51.5 1
24
 Quantiles

• Quartiles are three points which divide an array into four parts in
such a way that each portion contains an equal number of
elements.
– First quartile (Q1) 25% of the observations lies below or equal to it

– Second quartile (Q2) 50 % of the observations lies below or equal to it and

– Third quartile (Q3) 75% of the observations lies below or equal to it

• The ith quartile for raw data is

i n  1
Qi 
4
• If there is an even number of data items, then we need to get the average
of the middle numbers.
25
 Quantiles

• Example: Find the median, lower quartile and upper quartile of the
following numbers.
a) 12, 5, 22, 30, 7, 36, 14, 42, 15, 53, 25

b) 12, 5, 22, 30, 7, 36, 14, 42, 15, 53, 25, 65

• Solution: first arrange data from smallest to largest

a)

b)

13 23.5 39
26
 Quantiles

• The ith quartile for grouped frequency distribution is

27
 Quantiles …

• Deciles are nine points which divide an array into 10 parts in such
a way that each part contains equal number of elements.
– The nine deciles are denoted by D1, D2, …, D9

– First decile (D1) 10% of the observations lies below or equal to it

– Second decile (D2) 20% of the observations lies below or equal to it etc

• The ith decile for grouped frequency distribution is

28
 Quantiles …

• Percentiles are 99 points which divide an array into 100 parts in

such a way that each part consists of equal number of elements.
– The ninty nine percentiles are denoted by P1, P2, …, P99

– First percentile (P1) 1% of the observations lies below or equal to it

– Second percentile (P2) 2% of the observations lies below or equal to it etc

• The ith percentile for grouped frequency distribution is

29
 Quantiles …

– Example:- The following frequency distribution is the score of 25 students.

Score Number
of
students Compute the following quantities
25-29 1 ● First quartile (Ans:44.92)
30-34 1
●Ninth decile (Ans:65.75)
35-39 1
●forty fifth percentile (Ans:51.38)
40-44 3
45-49 3 Remark:
50-54 6
Q1  P25
55-59 4
Q2  D5  P50  Median
60-64 3
65-69 2 Q3  P75
70-74 1 D1  P10 ; D2  P20 ;; D9  P90
30
3.2 Measures of Dispersion

31
 Introduction
– Central tendency measures do not reveal the variability present in the data.

– Dispersion is the scatteredness of the data series around it average.

– Dispersion is the extent to which values in a distribution differ from the

average of the distribution

– A measure of statistical dispersion is a nonnegative real number that is zero

if all the data are the same and increases as the data become more diverse.

• Why we need measures of dispersion?

– Determine the reliability of an average

– Serve as a basis for the control of the variability

– To compare the variability of two or more series and

– Facilitate the use of other statistical measures.

32
 Introduction…

• Properties of a good measures of dispersion

– It should be rigidly defined

– It should be easy to understand and to calculate

– It should be based on all observations of data

– It should be easily subjected to further mathematical treatment

– It should be least affected by sampling fluctuation

– It shouldn’t be unduly affected by extreme values

33
 Introduction…

• There are many types of dispersion measures

– Range /Relative Range (Coefficient of range)

– Inter Quartile Range/ coefficient of quartile deviation

– Mean Absolute Deviation /Coefficient of mean deviation

– Variance/Standard Deviation/ coefficient of variation

• Measures of dispersion cane be absolute or relative.

– When measurements are observed with different units, or have different
averages use relative measures of dispersion.

34
 Range (R)

• Range is the difference between two extreme values in a data

• Denoted by R

R = max − min

• Only two values are used in its calculation.

• It is influenced by an extreme value (non-robust).

• It is easy to compute and understand.

35
 Relative Range (RR)

• Relative range is the ratio of the difference and sum of the two
extreme values in a data

• Denoted by RR/CR

max  min
RR 
max  min

• Example: what is the range and relative range of the following

data: 4, 8, 1, 6, 6,2, 9, 3, 6, 9. (Ans: R=8, RR=0.8)

36
 Properties of range

• It is the simplest crude measure and can be easily

understood

• It takes into account only two values which causes it

to be a poor measure of dispersion

• Very sensitive to extreme observations

• The larger the sample size, the larger the range

 Inter Quartile Range

• Measures the range of the middle 50% of the values only

• Is defined as the difference between the upper and lower quartiles

• Interquartile range = upper quartile - lower quartile

= Q3 - Q1

• The semi-interquartile range (or SIR) is defined as the difference of

the first and third quartiles divided by two

SIR = (Q3 - Q1) / 2

• The SIR is often used with skewed data as it is insensitive to the extreme
scores

38
 Coefficient of Quartile Deviation
• The ratio of the difference to sum of the two extreme quartiles of a
data. Denoted by CQD
Q3  Q1
CQD 
Q3  Q1

• Example: The following data are recorded: 9, 7, 3, 7, 1, 2, 5, 4, 5,

10, 10, 2, 2, 2, 6, 7, 9, 8, 5, 6. What are the SIR and CQD for the
recorded data?

• Solution: put in ascending order: 1, 2, 2, 2, 2, 3, 4, 5, 5, 5, 6, 6, 7,

7, 7, 8, 9, 9, 10, 10. (Ans: SIR=2.5, CQD=0.5)

39
 Properties of IQR

• It is a simple and versatile measure

• It encloses the central 50% of the observations
• It is not based on all observations but only on two
specific values
• Since it excludes the lowest and highest 25% values, it
is not affected by extreme values
• Less sensitive to the size of the sample
 Mean Absolute Deviation (MAD)

• Measures the ‘average’ distance of each observation away from the mean of
the data
• Gives an equal weight to each observation
• Generally more sensitive than the range or interquartile range, since a
change in any value will affect it
• The Mean Absolute Deviation of a set of n numbers is
n

 x x i
MAD  i 1
n

– All values are used in the calculation.

– It is not unduly influenced by large or small values (robust)
– The absolute values are difficult to manipulate.

41
 Coefficient of Mean Deviation (CMD)
MAD
CMD 
x
– All values are used in the calculation.

– It is not unduly influenced by large or small values (robust)

– The absolute values are difficult to manipulate.

• Example: For the following data

52.5, 46.8, 38.8, 37.6, 32.3.

• Compute MAD and CMD?

• Solution: (Ans: MAD=6.44, CMD=0.16)

42
 Step 2 Step 3

Observation x xx xx

1 52.5 10.9 10.9

2 46.8 5.2 5.2

3 38.8 -2.8 2.8

4 37.6 -4 4

5 32.3 -9.3 9.3

Total 208 0 Step 4 32.2

Step 1
Mean 41.6 0 Step 5 6.44
 Properties of mean deviation

• MD removes one main objection of the earlier measures, that it

involves each value

• It is not affected much by extreme values

• Its main drawback is that algebraic negative signs of the

deviations are ignored which is mathematically unsound
 Variance

• Variance is the mean of squared deviation of observations from

their arithmetic mean

– All values are used in the calculation.

– It is not extremely influenced by outliers (non-robust).

– The units of variance are awkward: the square of the original

units.
• Therefore standard deviation is more natural since it recovers the original units.
45
• In general, the sample variance is computed
by:
 Standard Deviation

• One of the most useful measures of dispersion is the standard deviation.

• It is based on deviations from the mean of the data.

• The sample standard deviation is found by calculating the square root of

the variance.
s
 ( x  x )2 .
n 1
• To calculate standard deviation follow this step
1. Calculate the mean of the numbers

2. Find the deviations from the mean.

3. Square each deviation

4. Sum the squared deviations

5. Divide the sum in Step 4 by n – 1

6. Take the square root of the quotient in Step 5

47
Example 1: Compute the variance for the sample: 5, 14, 2, 2 and
17.
Solution:

Example 2: Suppose the data given below indicates time in

minute required for a laboratory experiment to compute a certain
laboratory test. Calculate the mean, variance and standard
deviation for the following data.
32 36 40 44 48 Total
2 5 8 4 1 20
64 180 320 176 48 788
2048 6480 12800 7744 2304 31376
 Properties of Variance
• The variance is always non-negative ( ).
• If every element of the data is multiplied by a
constant "c", then the new variance

• When a constant is added to all elements of the

data, then the variance does not change.
• The variance of a constant (c) measured in n
times is zero. i.e. (var(c) = 0).
 Coefficient of Variation

• The Coefficient of Variation (CV) for a data set defined as the ratio of the standard
deviation to the mean

• It shows the extent of variability in relation to mean of the population.

• It is a normalized measure of dispersion of a probability distribution or frequency

distribution.

s
CV  100%
x
– All values are used in the calculation.

– The actual value of the CV is independent of the unit in which the measurement has been
taken, so it is a dimensionless number.

– For comparison between data sets with different units or widely different means, one
should use the coefficient of variation instead of the standard deviation.
50
 Coefficient of Variation
Example: Last semester, the students of Biology and Chemistry Departments took
Stat 273 course. At the end of the semester, the following information was recorded.

Department Biology Chemistry

Mean score 79 64
Standard deviation 23 11
Compare the relative dispersions of the two departments’ scores using the
appropriate way.
Solution:

Biology Department Chemistry Department

11 23
CV  100  17.19% CV  100  29.11%
64 79
Since the CV of Biology Department students is greater than that of Chemistry
Department students, we can say that there is more dispersion in the distribution of
Biology students’ scores compared with that of Chemistry students.

51
 Standard Score

• If X is a measurement from a distribution with mean X and standard

deviation S, then its value in standard units is
X X
Z
S
• Z gives the deviations from the mean in units of standard deviation

• Z gives the number of standard deviation a particular observation lie

above or below the mean.

• It is used to compare two observations coming from different groups

52
 Standard Score

• Example: Two groups of people were trained to perform a certain task

and tested to find out which group is faster to learn the task. For the two
groups the following information was given:
Value Group one Group two
Mean 10.4 min 11.9 min
Stan.dev. 1.2 min 1.3 min

• Relatively speaking:

a) Which group is more consistent in its performance? (Ans: Group 2)

b) Suppose a person A from group one take 9.2 minutes while person B from Group
two take 9.3 minutes, who was faster in performing the task? Why? (Ans: person B
is faster)

53
 S1 1 .2
Coefficient of variation for group 1: CV   100 %   100 %  11 . 54 %
x1 10 . 4
S2 1 .3
Coefficient of variation for group 2: CV   100 %   100 %  10 . 92 %
x2 11 . 9

CV for group 2 < CV for group 1 group 2 is more consistent

x A  x1 9 . 2  10 . 4
Z-score of Person A: Z     1 . 00
S1 1 .2

xB  x2 9.3  11 .9
Z-score of Person B: Z  S 
1.3
 2.00
2

Z-score of Person B < Z-score of Person A  Person B is faster than

person A
 Skewness

• Skewness is the degree of asymmetry or departure from symmetry of a

distribution.

• A skewed frequency distribution is one that is not symmetrical.

• Skewness is concerned with the shape of the curve not size.

– If the frequency curve (smoothed frequency polygon) of a distribution has a longer
tail to the right of the central maximum than to the left, the distribution is said to be
skewed to the right or said to have positive skewness.

– If it has a longer tail to the left of the central maximum than to the right, it is said to
be skewed to the left or said to have negative skewness.

55
 Skewness

– For moderately skewed distribution, the following relation holds among the three
commonly used measures of central tendency

mean-mode=3(mean-median)

– Pearson an measures of skewness

mean  mod e
skewness  or
s tan dard deviation

m3  x  x  2
 x  x 3

skewness  , where m2  and m3 

m2  2
3
n 1 n 1
• If skewness > 0, then it is positively skeweed

• If skewness = 0, then it is positively symmetric

• If skewness < 0, then it is negatively skeweed

56
 Kurtosis

• Kurtosis is the degree of peakdness of a distribution, usually taken relative to a

normal distribution.
– A distribution having relatively high peak is called leptokurtic.

– If a curve representing a distribution is flat topped, it is called platykurtic.

– The normal distribution which is not very high peaked or flat topped is called
mesokurtic.

57
 Kurtosis

• Measures of kurtosis

m4   x  x 4
  x  x 2

kurtosis  2 , where m4  and m2 

m2 n 1 n 1
– If kurtosis > 3, then the curve is leptokurtic

– If kurtosis = 3, then the curve is mesokurtic

– If kurtosis <3, then the curve is platykurtic

58
 Skewness and Kurtosis (Example)

• Example: the following observations are score of 100 students.

Score 61 64 67 70 73
Number of students 5 18 42 27 8

Can we say the distribution is skeweed? What is the shape of the distribution?
(Ans: mean=67.45, m2=8.61, m3=-2.72, m4=201.39)

m3  2.72
skewness    0.11
m2  8.61
3 3
2 2
• The distribution is negatively skewed

m4 201.39
kurtosis  2   2.71
m2 8.612

• The shape is platykurtic

59
 x f f*x

f x  x 
2
 f x  x   f x  x  
3 4

61 5 305 208.0125 -1341.68 8653.84

64 18 1152 214.245 -739.145 2550.051
67 42 2814 8.505 -3.82725 1.722263
70 27 1890 175.5675 447.6971 1141.628
73 8 584 246.42 1367.631 7590.352
Total 100 6745 852.75 -269.325 19937.59
mean 67.45 m2 8.613636 m2 -2.72045 m4 201.3898

skewness 
m3  2.72 m4 201.39
  0.11 kurtosis  2   2.71
m2  2 8.61 2 m2 8.61
3 3
2