How to find Central Tendency of Medical Data?

Central Tendency of Medical Data

Measures of Central Tendency
 
A measure of central Tendency is a value that gives the central position of given data. Measures of central tendency condensing the mass of data in one single value. Central tendency enable us to get an idea of the entire data. Measures of central tendency also enable us to compare two or more sets of data to facilitate comparison.​

For example, it is impossible to remember the each match runs scored by Virat Kohali in One Day International cricket, but if the average runs, scored, is obtained; we get one single value that represents the entire One Day International cricket career score. 


Properties of a Good Measure of Central Tendency:

• It should be easy to understand and calculate.

• It should be rigidly defined.

• It should be based on all observations.

• It should be least affected by sampling fluctuation.

• It should be capable of further algebraic treatment.

• It should be least affected by extreme values.


Following are three important measures of central tendency which are commonly used

I.   Arithmetic Mean
II.  Median
III. Mode


Central Tendency Definition Formula for Ungrouped Data Formula for Discrete Data Formula for Continuous Data
Arithmetic Mean or Mean (x̄) The arithmetic mean is defined as sum of the numerical values of each and every observation divided by the total number of observations

n = Number of observations

N = Total frequency


Xi = middle value of class interval

N = Total frequency
Median (M)
Middle most or central value of observation is called as median.

Median devides the given data into two equal parts.
I] When n is odd number:

M{(n+1)/2}th term


II] When n is even number:

M = [ (n/2) + (n+1)/2 ] / 2
M = (N/2)th   term
N = Total Frequency
M = l + [(N/2 – C) / f] × h

Where:
l = Lower limit of class interval of median

f = Frequency of class where median lies

C =    Cumulative frequency before median class

N = Total frequency

h = Width of class interval
Median (M)
Middle most or central value of observation is called as median.

Median devides the given data into two equal parts.
I] When n is odd number:

M={(n+1)/2}th
term


II] When n is even number:

M = [ (n/2) + (n+1)/2 ] / 2
M =(N+1)/2th term

N = Total Frequency
M = l + [(N/2 – C) / f] × h

Where:
l = Lower limit of median class CI

f = Frequency of median class CI

C = Preceding cumulative frequency of median class CI

N = Total Frequency
h = Width of CI
Mode (Mo) Most occurring value or most repeating value of observation is called as mode. Most repeating value of observation The observation which corresponds to maximum frequency. Mo = l + [(fm – f1) / (2fm – f1 – f2)] × h

Where:
l = Lower limit of mode class CI

fm = Frequency of mode class CI

f1 = Preceding Frequency of mode class CI

f2 = Succeeding Frequency of mode class CI
h = Width of mode class CI

I] Arithmetic Mean:[ Mean ] 

The arithmetic mean (or mean or average) is the most commonly used and readily understood measure of central tendency. The arithmetic mean is defined as sum of the numerical values of each observation divided by the total number of observations.

Case I : When data is Ungrouped data/Raw data

 If x1, x2,  x3, ................xn be the n number of observations. Then arithmetic mean is calculated as

Case II : When data is Grouped data


Case I : When data is Ungrouped data/Raw data

Example 1: Calculate mean from following data

Age of eight patients in years: 14, 17, 18, 20, 21, 22, 25, 29.




Solution:




                   

                   

The Arithmetic mean Patient age is 20.75 years. 

Case II: When data is Grouped data

Example 2: Calculate mean from following frequency distribution data


 
Hemoglobin level (g/dL) 10    11   12    13    14    15   
Number of Cases 06 09 21 11 07 04




Hemoglobin level in g/dL (Xi) Number of Cases (fi) Xifi
100660
110999
1221252
1311143
140798
150460
Total N=58∑〖Xifi 〗=712




The mean Arithematic Hemoglobin level in case is 12.27 g/dL.

 The Arithematic mean of age in patients is 50.8 years.

 Merits of Arithmetic Mean:

    It is easy to understand and calculate.

    It is rigidly defined.

    It is based on all observations.

    It is least affected by sampling fluctuation.

    It is useful for further algebraic treatment.

·       Demerits of Arithmetic Mean:

·          It is very much affected by extreme values.

     It is not calculated in case of open-end interval.

II] Median: [M]

Middle most or central value of observation is known as median.  Median devides the given data into two equal parts. Fifty per cent of the observations in the data are above the value of median and other fifty per cent of the observations are below this value of median. The median is the value of the middle observation when the series is arranged in order of size or magnitude (Ascending order). Data should be arranged in increasing / decreasing order.

It is denoted by letter ‘M. 

Case I : When data is Ungrouped data/Raw data

              I] If n is an odd number:

                        M={[n+1]/2}th.term


             II] If n is an even number: 

                        M=((n/2)  th+(n/2+1)th)/2 Term


Case II: When data is grouped data.

I] when data is Discrete



N= Total Frequency

Note: Find cumulative frequency and find nearest greatest value of Mth term in cumulative frequency, so its corresponding observation is median value.

II] When data is Continuous  

 

Where as l=is lower limit of median lies Class Interval.

f=Frequency of median lies Class Interval.

C= Preceding cumulative frequency of median lies Class Interval.

N= Total Frequency

h = is width of Class Interval.

Merits and Demerits of Median:

Merits of Median:

       It is easy to understand and calculate.

·          It is rigidly defined.

           It is not affected by extreme values.

·             It is calculated in case of open-end interval.

·          It can be located graphically.

      Demerits of Median:

·           It is not based on all observations.

·           It is least affected by sampling fluctuation.

·            It is not capable of further algebraic treatment.


Case I : When data is Ungrouped data/Raw data

Example 1: Calculate median from following data

                        10, 17, 18, 12, 23, 22, 25, 21,11

Solution:

                 Let’s arrange given data in ascending order

                      10, 11, 12, 17, 18, 21, 22, 23, 25

Here n=9 which is odd number so formula is

M={[n+1]/2}th.tern

M={[9+1]/2}th.tern 

M={[10]+1/2}th.tern 

M= 5 th term     (5 th Term value in ascending order data)

M = 18

So Median of given data is 18. 


Example 2: Calculate median from following data

                            56, 34, 31, 87, 23, 45, 25, 49

Solution: 

  Let’s arrange given data in ascending order 

                        23, 25, 31, 34, 45, 49, 56, 87.

Here n=8 which is Even number so formula is 



(4 th and 5 th Term value in ascending order data)

      

 Case II : When data is Grouped data.

Example 2: Calculate median of following frequency distribution data


Marks

40

42

44

45

47

50

Number of students

 08

18

12

09

07

06


Solution: 

                                    


Marks

(Xi)

Number of students

(fi)

Cumulative Frequency

(cf)

40

08

08

42

18

26

44

12

               38 > 30.5th

45

09

47

47

07

54

50

06

60

Total

N=60

 



                                     M= ( 30.5 )th term 
 
(So its nearest greatest value in cumulative frequency is 38 so its corresponding
observation is median value i.e. 44 marks).

Median = 44 marks

So Median for above given frequency distribution of marks is 44 marks. 

Example 4: Compute the median for the following frequency distribution:

Size of items:

0-4

4-8

8-12

12-16

16-20

20-24

24-28

28-32

32-36

36-40

Frequency

5

7

9

17

12

10

6

3

2

1



Size of items

Number of students

(fi)

Cummulative Fequency

(cf)

0-4

5

5

4-8

7

12

8-12

9

       21 (c)

(l) 12-16

        17 (f)

          38 >36

16-20

12

50

20-24

10

60

24-28

6

66

28-32

3

69

32-36

2

71

36-40

1

72

Total

N=72

 


 (So 36th term nearest greatest value in cumulative frequency is 38, so its corresponding Class interval is median lies class interval i.e. 12-16).

    


Here l = 12,  N/2= 36, f=17,   C=21, h=4


So Here Median is 15.53. 
                           
III] Mode: [Mo] 

Mode is defined as the value, which occurs most often, or with the greatest frequency. In some series, data may have more than two mode values, Bi-modal, Tri-modal or Multi-modal.

Where as for frequency distribution data, the mode is observation which corresponds maximum frequency. 

Case I : When data is ungrouped / Series data

Mode is defined as the value, which occurs most often, or most repeating value. 

Case II : When data is grouped data

I] when data is Discrete 

    Mode is observation which corresponds maximum frequency. 

II] when data is Continuous


Where as l=is lower limit mode lies CI.

fm= Frequency of mode lies CI.
f1 = Preceding Frequency of mode lies CI.
f2 = Succeding Frequency of mode lies CI.
h  = is width of mode lies CI.

Note: Mode lies class interval is the Class Interval which corresponds maximum frequency in frequency distribution table. 

Merits and Demerits of Mode: 

Merits of Mode: 

It is easy to understand and calculate.
It is not affected by extreme values.
It is calculated in case of open-end interval.
It can be located graphically.

Demerits of Mode: 

It is not based on all observations.
It is rigidly defined.
It is least affected by sampling fluctuation.
It is not capable of further algebraic treatment.


Example: Calculate mode from following data

15, 20, 25, 18, 14, 15, 25, 10, 18, 16, 20, 25.

Solution: 

In above given data the most occurring value is 25. No other value or observation is repeated three or more than three time.

So Mode = 25.

Case II : When data is grouped data

I] when data is Discrete 

Example 2: Calculate mode of following frequency distribution data

Marks

50

60

70

80

90

100

Number of students

4

12

28

33

11

7


Solution: 

In above given frequency distribution data the maximum frequency is 33. 

Therefore, its corresponding observation is 80 marks

So Mode = 80 marks



II] When data is Continious  

Example: For the following grouped frequency distribution find the mode

Class Interval:

0-20

20-40

40-60

60-80

80-100    

100-120

12-140

140-160

Frequency

2

5

10

23

21

14

08

03



Class Interval

Frequency (fi)

0-20

2

20-40

5

40-60

10 (f1)

(l) 60-80

23(fm)

80-100

21 (f2)

100-120

14

120-140

08

140-160

03

Total

N=72



In give frequency distribution, table maximum frequency is 23, so its corresponding Class interval is Mode lies class interval i.e.60-80.

Here l = 60,  fm =23,  f1 =  10,     f2 =  21,     h= 20. 




Mo=60+[0.867]*20

Mo = 60+17.34

Mo=77.34

So Mode = 77.34