Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Numerical Descriptive Measures: A. Measures of Central Tendency

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 21

NUMERICAL DESCRIPTIVE MEASURES

A. Measures of Central Tendency


It describes the “center” of a given data
set. It is a single value about which the
observation tends to cluster.
1. The Arithmetic Mean (or simply Mean)
-the sum of the observations divided by the number of
observations totaled, denoted by µ.

Properties:
It always exists for quantitative variables.
It is unique.
It takes into account every item of the data.
Thus, it is easily affected by extreme values.
Grouped Mean
k
 fi Xm
i1

n

where :
Xm = classmark
fi = frequency
n = number of observations
k = number of classes
Ungrouped Mean

n
 Xi
i1

n

Where:
Xi = ith observed value of the variable X
n = total number of observed values
The Median - the middle value of an array, denoted by Md.

Properties:
Not easily affected by extreme values.
It always exists and is unique.

Ungrouped Median

 X n1 , if n is odd


2
Md   X n  X n
1
 2 2
, if n is even

 2
Grouped Median

  1 
 n 2   Fp 
M d  LCB  c    
 fMd 
 
 
where:
LCB = lower class boundary of the median class.
Fp = <CF of the previous (frequency before the median class)
fMd = frequency of the median class
c = class size
Grouped Mode

 fMo  fb 
Mo  LCB  c  
 2fMo  fb  fa 

where:
LCB = lower class boundary of the modal class.
FMo = frequency of the modal class
fb = frequency of the class before the modal class
fa = frequency of the class after the modal class
c = class size
The Mode – the observation(s) that occur most frequently in
the data set, denoted by Mo.

Properties:
No calculations are required
(for the ungrouped mode).
It may not exist.
It may not be unique.
B. Measures of Position
They are measures that discriminate a group of scores from
another group in the same data set.

Defn: Quantiles – divides data into an equal number of parts.

The Quartiles –are values that divide a set of observations


into four equal parts, denoted by Qi, i = 1, …, 4.

The Deciles – are values that divide a set of observations into


10 equal parts, denoted by Di, i = 1, 2, …, 10.

The Percentiles – are values that divide a set of observations


into 100 equal parts, denoted by Pi, i = 1, 2, …, 100.
UNGROUPED MEASURES OF POSITION

To locate the desired quantile:

P(n  1)
1. Use to locate the position.
100

P(n  1)
2. If is not exact, then do interpolation.
100
GROUPED MEASURES OF POSITION

Percentile

  1  
 ni 100   Fp 
Pi  LCB  c    
 fi 
 
 
C. Measures of Variability
It describes the extent to which the data are dispersed.

1.The Range (R) - difference between the


highest and lowest value in the data set.

2.The Variance (s2 or σ2)- the mean squared


differences of the observations from their
mean.
Ungrouped Variance

s 
2  XX  
2

n 1
or

s 
2 
n  X   X  2
 2

n(n  1)
Grouped Variance

s 
2 
 f Xm  X 
2

n 1
3. The Standard Deviation (sd or s or σ) - the
positive square root of the variance.

4. The Coefficient of Variation (CV) - the ratio of


the standard deviation to its mean expressed
in percent.
Measures of Skewness

A frequency curve that is not symmetrical about the mean is said to be


skewed. If it tails off to the right, we describe it as positively skewed, but if
it tails off to the left, we say it is negatively skewed. The relationship
between the mean and the median is related to the direction skewness.

If the mean is greater than the median, we have positively skewed curve,
but if the mean is less than the median, we have a negatively skewed
curve. Now, with the use of the standard deviation, it is possible to obtain
a measure of skewness which indicates both the direction and the
magnitude (or the extent) of skewness of a frequency data.

It is called the Pearsonian coefficient of skewness (Sk) and the formula is

3( x  md )
Sk 
s
The direction of skewness is indicated by the
algebraic sign of the value of Sk, while the extent
to which the curve is skewed is indicate by the
magnitude of the value of Sk. This coefficient of
skewness is positive if the mean is greater than
the median, negative if the mean is less than the
median, and zero if they are equal. The curve is
symmetrical and bell-shaped when Sk = O.

As general rule, the closer the coefficient of


skewness is to zero, the less skewed the
distribution will be and farther this coefficient is
form zero, the more skewed the distribution will
be.
Measure of Kurtosis

Curves of distributions having the same coefficient of


skewness may still differ significantly. Symmetrical curves, for
instance, may vary in shape and this may be because they do
not have the same peakedness, a property of curves which
can be described by computing for the value called measures
of kurtosis (K).

As shown below, there are three types of symmetrical curves.


The mesokurtic curve is considered "normal"; a more peaked
curve is known as the leptokurtic curve; and while a more flat-
topped curve is called platykurtic curve.

The kurtosis for a set of data is obtained by simply dividing the


fourth moment about the mean by the square of the variance.
( x  x ) 4
For ungrouped data, the formula is K
ns 4

 f (x  x)4
and for grouped data, the formula is K
ns 4

Note that
s  (s )
4 2 2
is square of the variance.
A distribution is said to be mesokurtic if K=3,
leptokurtic if K>3, and platykurtic if K< 3.
EXERCISES:
1. The following are the number of minutes required to do
an assembly job by workers of a manufacturing plant:
77 85 63 54 62 77 80 69
Determine the following:
1.Mean
2.Median
3.Mode
4.Q1
5. D3
6. P55
7. Range
8. Variance
9. Standard Deviation
10.Coefficient of Variation
2. Below is a frequency distribution of scores obtained in an aptitude
test of 80 applicants for clerical positions in a large company:

Scores No. of Applicants

10-19 1

20-29 2

30-39 3

40-49 7

50-59 15

60-69 22

70-79 18

80-89 9

90-99 3

You might also like