Discuss the points for constructing continuous frequency distribution.

1.	Discuss the points for constructing continuous frequency distribution.
Answer» For constructing continuous frequency distribution the following points are to be considered : When the variable of the data is continuous or the range of variable is large, continuous frequency distribution should be constructed. Generally, the number of classes should be any number from 6 to 15. Under special circumstances, the number of classes may be less than 6 or more than 15. Considering the number of classes k and the range of data R. class length is decided using the following formula: c = RK The value of c is selected such that ck ≥ R and c is positive integer. Generally, the class length of each class is kept equal. But under the circumstances that range of the data is large, keeping in view the number of classes, classes of different lengths can be constructed. Usually, the initial class should begin with the observation multiple of class length and a number smaller than and close to the lowest observation of the data. Classes can be chosen either inclusive type such as 10-14, 15-19, 20-24, … or exclusive type such as 10- 15, 15-20, 20-25 For data on continuous variable exclusive type of classes should be preferred and when the range of discrete data is large one, then inclusive type of classes should be preferred.

Answer»

For constructing continuous frequency distribution the following points are to be considered :

When the variable of the data is continuous or the range of variable is large, continuous frequency distribution should be constructed.
Generally, the number of classes should be any number from 6 to 15. Under special circumstances, the number of classes may be less than 6 or more than 15.
Considering the number of classes k and the range of data R. class length is decided using the following formula: c = RK
The value of c is selected such that ck ≥ R and c is positive integer.
Generally, the class length of each class is kept equal. But under the circumstances that range of the data is large, keeping in view the number of classes, classes of different lengths can be constructed.
Usually, the initial class should begin with the observation multiple of class length and a number smaller than and close to the lowest observation of the data.
Classes can be chosen either inclusive type such as 10-14, 15-19, 20-24, … or exclusive type such as 10- 15, 15-20, 20-25
For data on continuous variable exclusive type of classes should be preferred and when the range of discrete data is large one, then inclusive type of classes should be preferred.

Discussion