Download as pdf or txt
Download as pdf or txt
You are on page 1of 23

MEASURES OF VARIABILITY

To compute the ff:


standard deviation from ungrouped and grouped data;
Variance
Coefficient of Variation; and
Quartile Deviation
Variation

Mean
Data Set 1: 47 48 52 55 58 52
Data Set 2: 46 50 51 53 60 52
Data Set 3: 49 50 52 53 56 52
Data Set 4: 50 51 53 54 55 52
Dot Plot
Variability

Measures of Central Tendency are NOT enough to


compare data sets
Information as to the degree of spread of the data or
the variation of the data relative to the mean is an
important concept in statistics
“Variability” , “Dispersion”, or “Spread” of data
Measures of Variability

Range
Mean Absolute Deviation (MAD)
Variance or Mean Square
Standard Deviation
Range

Is the difference between the highest observation


and the lowest observation in a given set of data
R = XN – X1
It is easy to compute
It highly unstable
Range for Ungrouped

XN X1 R
Set 1: 47 48 52 55 58 58 47 11
Set 2: 46 50 51 53 60 60 46 14
Set 3: 49 50 52 53 56 56 49 7
Set 4: 50 51 53 54 55 55 50 5
Range for Group

Class Interval f
95-99 4
90-94 8
85-89 12
80-84 7
75-79 6
70-74 2
65-69 1

Using the class boundaries, R = 99.5 – 64.5 = 35


Using the apparent limits, R = 99 – 65+1 = 35
Other Measures of Variability based
on the Idea of Range

10-90 Percentile Range


PR = P90 - P10

Interquartile Range
QD= Q3 - Q1
2
Compute for QD
Class f <cf % 𝑁 − 𝐹$
Interval 𝑄# = 𝐿𝐿 + 𝑐
𝑓
95-99 4 40
2
90-94 8 36 𝑄" = 89.5 + 5 = 90.75
8
85-89 12 28
1
80-84 7 16 𝑄! = 79.5 + 5 = 80.21
7
75-79 6 9
70-74 2 3
65-69 1 1
c=5 N=40
Solution:
QD = 90.75 – 80.21 = 5.27
2
Mean Absolute Deviation (MAD)

It considers the deviation of the scores from the


arithmetic mean.
𝑑 = 𝑋 − 𝑋%
A deviation score is either zero, positive or negative
depending on whether the score is equal to the
mean, greater than the mean, or less than the mean,
respectively.
It is easy to interpret because it is the average
distance of the scores from the mean.
Mean Absolute Deviation (MAD)

% 3
∑(𝑋 − 𝑋)
𝑀𝐴𝐷 =
𝑁
∑ Summation
X Score
𝑋̅ Mean of the Score
N Total Number of Cases
Compute for MAD
MAD for Grouped Data
Class Interval f X 4
X-𝑿 4|
|X-𝑿 4|
f |X-𝑿
95-99 4 97 11.625 11.625 46.500
90-94 8 92
85-89 12 87
80-84 7 82
75-79 6 77
70-74 2 72
65-69 1 67
4 = 85.375
𝑿 N=40 SUM 238

MAD = 238/40 = 5.95

40 scores deviate from the mean of 85.375 by an average of


5.95 units
Variance

2 is a parameter while S2 is a statistic


2 = (X - )2

N
Compute for the Variance
(score (deviation score) (squared deviation)
) X d=X-X d2
49 2.5 6.25
48
42
45
48
45
47
48
=46.5 0 38

%
∑(𝑋 − 𝜇)% 38
𝜎 = = = 4.75
𝑁 8
Compute for Standard Deviation

Is interpreted as the average distance of the scores


from the mean.
2 = 4.75, = 2.18
It is more meaningful than variance because it is in
the same unit as the original scores. It is useful in
inferential statistics
Differentiate Two Sample Variance

Biased estimator
Unbiased estimator
Variance and SD for Grouped Data

Interval f X fX fX2
60-62 5 61 305 18,605
63-65 18 64
66-68 42 67
69-71 27 70
72-74 8 73
n=100 ∑𝑓𝑋 =6,745 ∑𝑓𝑥% =455,803

"(∑%&!) ) ∑%* ! ,-- .//,1-2 )(3,4./! )


𝑆! = = = 8.53
+! ,--!
Variance and SD for Grouped Data

S2 = 100(455,803) – (6,7452) = 8.53


1002

S.D. = 8.53 = 2.92

Interpretation: The Scores deviate from the mean by


an amount of 2.92 units, on the average.
Coded Deviation Method

Compute the Coded Deviation Formula


MEASURE OF RELATIVE VARIATION

Coefficient of Variation (c.v)

C.V. = S.D. 100 %


𝑋.̅
Compute for C.V.

Teaching 23 27 19 25 29 21
Experience
(Yrs)
Net Take 8,000 6,000 10,500 7,000 12,000 9,750
Home Pay
(PhP)

Data1 (YRS): 𝑋? = 24 𝑦𝑒𝑎𝑟𝑠 and S.D. = 3.742 years


Data2 (PhP): 𝑋? = 𝑃8, 875 and S.D. = P2,267.984

H.I. K.LM3
Data1 (C.V.) = J?
100% = 3M
(100%) = 15.59%
H.I. 33NL.OPM
Data2 (C.V.) = J?
100% = PPLQ
(100%) = 25.55%

You might also like