Thursday, February 28, 2019

What is correlation in statistics.?

Correlation is a statistical method used to know to the relationship between
Variables and  how the variables are associated or related each other for e.g., price and supply, salary and experience, grades and hours of study. The measure of correlation between two variables is called (correlation coefficient). The correlation coefficient is between -1 and 1, the relationship between two   variables can one of this these three ways .

a)Positive correlation:  they have same direction; means if one increase, the other one will increase. For e.g. income and expenditure and its between zero up to one

b)Negative correlation: they have different directions its when one increase, the other one will decrease for eg piece and demand  and its between zero up to -1.

c)Zero correlation: if one change does depend on the other, for eg weights of students  and marks .and  its zero.

Note, spurious correlation : when two variable have relation but logically unacceptable for e.g. donkey and graduates students.

     The methods of correlation coefficients finding

1) Karl Pearson correlation coefficient.:( parametric situations =when follows normal distribution)
 2) Spearman Rank correlation coefficient.: ( non parametric)

Thank you

1 comment: