Data can be grouped in two ways - discrete frequency distribution and continuous frequency distribution.
A collection of data consisting of distinct values occurring with certain frequencies is called discrete frequency distribution.
Example of discrete frequency distribution:
The table displays the data of the performance of 50 students in a math test.
Classify the data into different class-intervals along with their respective frequencies.
When data are arranged in different class intervals without any gaps along with their frequencies, it is called continuous frequency distribution.
An example of continuous frequency distribution:
The table displays the marks scored by 50 students.
Marks Scored |
Number of Students |
90 - 100 |
4 |
75 - 90 |
9 |
60 - 75 |
14 |
40 - 60 |
18 |
0 - 40 |
5 |
Mean deviation about the mean for discrete frequency distribution:
Observations |
Frequencies |
x1 |
f1 |
x2 |
f2 |
x3 |
f3 |
.... |
.... |
xn |
fn |
N = Sum of the frequencies (Number of observations)
Deviation from the mean,
=
, where i = 1, 2, 3, ... , n
Absolute values of the deviations = |
|, where i = 1, 2, 3, ... , n
Mean deviation about the mean = MD (
)
Ex: Consider the weights of 20 students.
Weight in kg (xi) |
Number of Students (fi) |
48 |
3 |
49 |
4 |
50 |
6 |
51 |
4 |
52 |
3 |
To find the mean deviation of the given data, mean of the data is found.
Weight in kg (xi) |
Number of Students (fi) |
fixi |
48 |
3 |
144 |
49 |
4 |
196 |
50 |
6 |
300 |
51 |
4 |
204 |
52 |
3 |
156 |
|
|
= 1000 |
Mean (
) =
Modulus of deviation of each observation from the mean and the sum of the products of the deviations of the observations and the frequency concerned:
Weight in kg (xi) |
Number of Students (fi) |
fixi |
|
|
48 |
3 |
144 |
|48 - 50| = 2 |
6 |
49 |
4 |
196 |
|49 - 50| = 1 |
4 |
50 |
6 |
300 |
|50 - 50| = 0 |
0 |
51 |
4 |
204 |
|51 - 50| = 1 |
4 |
52 |
3 |
156 |
|52 - 50| = 2 |
6 |
|
|
= 1000 |
|
- |= 20 |
MD (
) =
-
| =
= 1
Mean deviation about the mean for Continuous frequency distribution:
Marks Scored |
Number of Students (fi) |
80 - 100 |
3 |
60 - 80 |
15 |
40 - 60 |
19 |
20 - 40 |
10 |
0 - 20 |
3 |
Mean of the data:
Marks Scored |
Number of Students (fi) |
Mid-values (xi) |
fixi |
80 - 100 |
3 |
90 |
270 |
60 - 80 |
15 |
70 |
1050 |
40 - 60 |
19 |
50 |
950 |
20 - 40 |
10 |
30 |
300 |
0 - 20 |
3 |
10 |
30 |
|
∑ fi = 50 |
|
= 2600 |
Mean (
) =
=
= 52
If the number of observations is large, then the mathematical calculation is tedious.
For data with a large number of observations, the assumed mean method or the step-deviation method can be used to find the mean.
Find the modulus of the deviations of the mid-values from the mean and find the product of the deviations with their respective frequencies.
Marks Scored |
Number of Students (fi) |
Mid-values (xi) |
fixi |
|
|
80 - 100 |
3 |
90 |
270 |
|90 - 52| = 38 |
114 |
60 - 80 |
15 |
70 |
1050 |
|70 - 52| = 18 |
270 |
40 - 60 |
19 |
50 |
950 |
|50 - 52| = 2 |
38 |
20 - 40 |
10 |
30 |
300 |
|30 - 52| = 22 |
220 |
0 - 20 |
3 |
10 |
30 |
|10 - 52| = 42 |
126 |
|
∑ fi = 50 |
|
= 2600 |
|
- |= 768 |
Mean deviation about mean = MD (
) =
-
| =
= 15.36
Therefore, the mean deviation about the mean is equal to 768 divided by 50, which is equal to 15.36.