measures of variability
TRANSCRIPT
Lesson 6Measures of Variability
Measures of Variability
Consider two sets of distributions. Compute for their means.
a. 70, 76, 78, 81, 85, 88, 89 x = 81
b. 77, 78, 79, 80, 83, 85, 85 x = 81
Measures of Variability
IMPORTANT!!!The measure of averages alone is not
sufficient to describe the distribution.
Averages locate the center of the distribution but they do not tell us how are the observations are positioned in relation to the center.
Measures of Variability
Absolute Measures of Variability The range, the quartile deviation, the average
or mean deviation, the standard deviation, and the variance give us the distances of the scores from the measures of central tendency. They are categorized as the measures of absolute variability since their units are the same as those of the original units
Relative Measures of VariabilityIf two or more distributions of different units are
to be compared, it is more appropriate to use the measures of relative variability. These measures are the coefficient of variation and the standard scores.
Measures of VariabilityThe RangeThe range is the simplest and the easiest to
compute among the measures of variability because it depends only on the pair of extreme values
But it is also the most unstable since its value easily fluctuates with the change in either of the highest or lowest observations.
It is also the most unreliable because it does not give the dispersion or spread of the observations between the two extreme values.
Measures of Variability
NOTE:A more reliable measure should include all the values in the distribution to give us an adequate spread of all the scores from the average.
Measures of Variability
Exclusive Range It is the difference between the
highest and the lowest score in a distribution.
The exclusive range is used for ungrouped data.
Measures of Variability
Example: If the lowest score in a distribution is 24 and the highest is 39, what is the exclusive range?
Measures of VariabilityInclusive range. o It is the difference between the
lower class boundary of the lowest observation and the upper class boundary of the highest observation.
oGenerally, the inclusive range is used for grouped data.
Measures of Variability
Example:
5 – 9 lower scores 10 - 14 15 - 19
20 – 24 higher scores
Find the inclusive range.
Measures of Variability
Solution:
R = highest UCB - lowest LCB = 24.5 - 4.5R = 20 (inclusive range)
Measures of VariabilityQuartile DeviationReduces the effect of the extremely low
and high observations on the measure of dispersion.
The quartile deviation is used when the median is the preferred measure of the central tendency, that is, if there are scattered or extreme observations in the distribution.
It gives the spread of the observations around the median or the middle 50% of the cases in a distribution.
Measures of Variability
Interquartile Range - the spread of distribution used the third and first quartiles corresponding to the 75th and 25th percentiles, respectively.
Interquartile Range = Q3 Q1
Measures of VariabilitySemi-interquartile range (also known
as quartile deviation) of a set of data is defined
Semi-interquartile Range =
where Q1 and Q3 are the first and third
quartiles of the data respectively.
213 QQ
QD
Measures of VariabilityExample:
Using the given set of observation, determine the
Interquartile range ( IR )Quartile deviation ( QD )
Sets of Observations:A: 10, 8, 6, 4, 14, 11 , 16, 7
B: 9, 4, 8, 7, 9, 8, 10, 17
Measures of VariabilitySolution:
16 14 11 10 8 7 6 4Solving for Q3 and Q1
Interquartile Range = Q3 – Q1 = 13.25 – 6.25 = 7
Quartile Deviation =
75.6100
1875100
1
nP
25.2100
1825100
1 nP
25.133 Q
25.6Q
5.327
213
Measures of VariabilityFor grouped data, consider the example below.
Example :
The following data are frequency distributions of two sections in Statistics in a 100 item examination. Set A Set B
f f70-74 4 71-75 375-79 12 76-80 880-84 15 81-85 1385-89 13 86-90 790-94 6 91-95 595-99 5 96-100 4
Solving for Q1 or P25 and Q3 or P75
Measures of Variability
The formula for Q a quartile point is
Quartile Deviation
cfi
cfin
lPbi
ii
100
213 QQ
QD
Measures of Variability
Set A
LL – UL f <cf 70-74 4 4 75-79 12 16 80-84 15 31 85-89 13 44 90-94 6 50 95-99 5 55 c = 5 n = 55
Measures of Variability For Set A
For Set B
512
475.135.7425
p 513
3125.415.8475
p
93975.42
5625.78442.88
QD
58
3105.7525
p 57
24305.8575
p
956.42
875.79786.89
QD
5625.7825 p 442.8875 p
875.7925 p 786.8975 p
Measures of VariabilityThe Average or Mean Deviation
The average or mean deviation takes into account all the values in a given distribution.
The formula for mean deviation is
where is read as “the absolute value of x and x = score of value = mean n = number of cases
n
XXMD
XXX
X
Measures of Variability
Example 4. Given the two sets of distribution,
let us solve for the mean deviations.
Set A: 28, 29, 32, 37 and 39 Set B: 25, 32, 33, 40 and 45
Measures of VariabilitySet A
X 28 29
32 37 39
This means that on the average, the scores deviated from the mean by 4.
xx 3328 3329
3332
3337 3339
165x 20
n
XXMD
4
520 or
Measures of VariabilitySet B
X 25 32
33 40 45
This means that on the average, the scores deviated from the mean by 6.
xx
n
XXMD
3525 3532 3533
3540
3545
175 30
65
30 or
Measures of Variability
Sets Mean Mean Deviation
A 33 4
B 35 6
If the mean deviations for sets A, and B, are 4, and 6 respectively, it can be concluded that B has greater variability than A.
Measures of VariabilityGrouped Data
For the grouped data the mean deviation can be calculated using the following formula:
nxxf
MD i 1
where xi = class marksx = grouped data meann = total frequencyfi = frequency per class intervals
Measures of Variability
Example 5. Compute the mean deviation for the data in Example 3. Set A Set B
LL – UL f LL – UL f70-74 4 71-75 375-79 12 76-80 880-84 15 81-85 1385-89 13 86-90 790-94 6 91-95 595-99 5 96-100 4c = 5 n = 55 c = 5 n = 40
Measures of Variability(1)CI
LL - UL
(2)fi
(3)xi
(4)fixi
(5) (6)
70 – 74 4 72 288 11.82 47.28
75 – 79 12 77 924 6.82 81.84
80 – 84 15 82 1230 1.82 27.3
85 – 89 13 87 1131 3.18 41.34
90 – 94 6 92 552 8.18 49.08
95 – 99 5 97 485 13.18 65.9
c = 5 n = 55 4610 312.74
82.8355
4610
nxf
x ii 69.555
74.312
nxxf
MD ii
Measures of VariabilityVariance and Standard Deviation
Using the absolute value symbol, the influence of the negative signs is eliminated in the computation of the mean deviation or average deviation.
Another way would be to square the deviation from the mean. Adding these squared deviations and dividing the sum by the number of items, the variance, which is a basic measure of dispersion, is obtained.
Extracting the square root of the variance yields the standard deviation, the most commonly used measure of dispersion or variability.
Measures of Variability
1
2
n
xxi
1
1
2
1
n
XXS
n
i
Variance and Standard Deviation of Ungrouped Data
The formula for variance ungrouped data is
s2 =
The standard deviation for ungrouped data is the square root of the variance
Measures of Variability
If the standard deviation of the population is needed, S is changed
to σ. The formula for σ2 is given by
where N = size of the population
Measures of VariabilityExample 3. To illustrate the computation for standard deviation of the sample of ungrouped data, consider the distribution below.
Score Deviation Squared Deviations (X) from the Mean from the Mean
7 7 – 10 = -3 (-3)2 = 9
8 8 – 10 = -2 (-2)2 = 4 10 10 – 10 = 0 (0)2 = 0 12 12 – 10 = 2 (2)2 = 4 13 13 – 10 = 3 (3)2
= 9
XX 2XX
50 X 0 XX 262
XX
105
50 orX
Measures of Variability
S2 =
1
2
n
xxi
= 6.5
1
1
2
1
n
XXS
n
i 55.2S
An equivalent formula does away with computing for the mean and the deviations of the observations.
variance = S2
1
2
2
nnx
x
standard deviation =
1
2
2
nnX
XS
=
Measures of Variability
Example: Solve for the variance and standard deviation
X (X2)
10 102 = 100 9 92 = 81 7 72 = 49 5 52 = 25 4 42 = 16 2 22 = 4
Σx2 = 275
37 X
Measures of Variability
Solution:
1
2
2
nnx
x
166
372752
2
s
s2 =
5834.46
s2 =
06.3s
= 9.366
Measures of Variability
An Efficient Way of Computing the Sample Standard Deviation of Grouped Data
The short formula for standard deviation for grouped data is
22
nfd
ndf
cs
where c = class size f = class frequencyd = class deviationn = number of cases or total frequency
Measures of Variability
Example 4.To illustrate the preceding formula, take for example the grouped data which shows the wages per day
of the laborers in a certain construction.
Wages No. of
Laborers
(f)
d fd (d)2 f(d)2
80-84 8 2 16 4 32
75-79 12 1 12 1 12
70-74 16 0 0 0 0
65-69 13 -1 -13 1 13
60-64 9 -2 -18 4 3658 f 3' fd 93)( 2df
Measures of Variability
22)(
nfd
ndf
cs
2
583
58935
s
325.6s
Measures of Variability
Another formula that can be used to solve the variance and standard deviation of grouped data is the long method.
variance
1
2
2
n
xxfs
standard deviation = S =
1
2
nxxf
Measures of Variability
58 f 4161 fX 1208.2321)( 2XXf
Using the preceding data, let us compute the variance using this second formula.
Measures of Variability
nfX
X 74.7158
4161 orX
1
2
nXXf
s158
1208.2321
s
326.6s
Measures of Variability
Coefficient of Variation
The coefficient of variation is used to express the standard deviation as a percentage of the mean.
To compare two distributions with different means and standard deviation we have to compute for the coefficient of variation to express them in the same unit, that is, in terms of percentage.
To illustrate we use the formula
%)100(xsCV
whereCV stands for the coefficient of
variation; S, the standard deviation;and X is the mean
Measures of Variability
Example 8Department store A has a mean weekly
sales of 340 bags with a standard deviation of 12. Department store B has a mean weekly sales of 550 bags with a standard deviation of 15. In relative terms, which store has the greater variability in their weekly sales?
Measures of Variability C
C
Example 9:The mean score of a statistics test
of class A is 75 with a standard deviation of 13 while class B hasa mean score of 86 with a standard
deviation of 16.
Which class has a larger variation from the mean?
Measures of Variability C
C
%33.17)100(7513
%60.18)100(8616
Solution:
Class A:
Class B:
Interpretation:
Thus, Class B has a larger variation from the mean.This means that the students are more
homogenous in Class A than in Class B. This further implies that the teacher in Class B
has to exert more effort to meet the individual needs of the students.
END OF THE SHOW