Scribd
Upload
14 views6 pages

Basics of Correlation - NH

Types of correlation and their uses

Uploaded by

trideepkalita673
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
14 views6 pages

Basics of Correlation - NH

Types of correlation and their uses

Uploaded by

trideepkalita673
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 6
10.1 INTRODUCTION In the earlier chapters, we have studied the statistical problems and distributions relating to one variable like marks of students, wages of workers, etc. We discussed various measures of central tendency, which are confined to a single variable. This kind of statistical analysis involving one variable is known as univariate distribution. But, we may come across a number of situations with distributions having two variables. For example, we may have data relating to income and expenditure, price and demand, height and weight, etc. The distribution. involving two variables is called bivariate distribution. Ina bivariate distribution, we may be interested to find if there is any relationship between the two variables under study. In day-to-day life, we observe that there exists certain telationship between two variables, like between income and expenditure, price and demand and so on. Correlation is a statistical tool which studies the relationship between two variables. Two Variables are Correlated Ifthe change in one variable results in a corresponding change in the other variable, then the twvo variables are said to be correlated. Statistics for g “ * plotted graphically, these points would not give a straight line. ee x 10 20 30 ! Y¥. 5 7 12 From the above example, itis clear that the ratio of change between two variables (X and) isnot the same. C. Simple, Multiple and Partial Correlation On the basis of the number of variables involved, correlation may be of three types: 1. Simple Correlation: When only two variables are studied, it is a problem of simple correlation. For example, relationship between income and expenditure, price and demand, etc. | imma ' © Measures of Correlation Fa 1 ef ho? ruttiple Correlation: When the relationship among three or more than three variables is studied 2” slaneous then such relationship is called multiple correlation. In such a case, all the sil it C1 : an ables (both independent and dependent) are studied simultaneously. For example, ati . (fect 0% output of wheat due to fertilizers, rainfall, ete. offe' . ao oy 1 Correlation: Under partial correlation, the : relationship between two variables is examined other variables as constant. kevpi"S er example, production of wheat depends on many factors like rainfall, quality of seeds, manure, etc. westudy the relation between production of wheat and quality of seeds, roan, then the correlation is called partial, ol DAs lai {== - YY “positive and Negative eee? ‘and Non-Linear ‘Simple, Multiple and conelton a4 ___ Correlation Partial Correlation (a y ¥ ¥ y ¥ Y postive Negative Linear Non-Linear keeping rainfall and manure Simple _Multiple Partial conelation Correlation Correlation Correlation Correlation Correlation _Correlation ‘Whentwo (When two (When change (When change (Whenonly (When —_ (When relationship ysablesmove variables move inonevariable inonevariable twovariables relationship between two inthesame —-_‘inopposite tendstobeara doesnot beara arestudied) among three __variables is direction) directions) constant ratio constant ratio to ‘ormorethan examined keeping totheamount — the amount of three variables other variables as ofchange in changein other is studied) constant) othervariable) related variable) 10.5 DEGREES OF CORRELATION The degree or intensity of relationship between two variables is measured with the help of coefficient of correlation. The degree of correlation can be expressed in the following ways: Perfect Correlation Ifthe relationship between the two variables is such that the values of the two variables change (increase or decrease) in the same proportion, correlation between them is said to be perfect. Perfect correlation may be positive or negative. * Ifthe proportionate change in the values of two variables is in the same direction, it is called perfect positive correlation and the value is described as +1. * However, if equal proportional changes are in the reverse direction, then the relationship is known as perfect negative correlation and described as 1. Zero Correlation | : en there is no relationship between the two variables, we say that there is zero correlation (or absence of correlation). So, a change in the value of one variable has no particular effect on ‘value of the other variable. In this case, the value of coefficient of correlation will be zero. lte: Zero Correlation only implies that there is no linear relationship between the two variables. It Gea ean absence of any type of relation, ie. other type of relation may exist between the two variables. Limited Degree of Correlation In real life, economic data do not indicate perfect positive or negative correlation. At the same time, the cases of zero correlation are also very limited. + Generally, we find some limited degree of correlation between economic variables, + The value of correlation coefficient (¢) normally lies in between +1 and -1. « Ifthe changes in two variables are unequal and in the same direction, correlations limited and positive. + Correlation is limited negative when there are unequal changes in the reverse direction, * The limited degree of correlation can be high, moderate or low. Interpretation of Correlation Coefficient ‘According to Karl Pearson, the coefficient of correlation lies between two limits ie. + 1 « If there is perfect positive relationship between two variables, the value of correlation would be +1. + On the contrary, if there is perfect negative relationship between two variables, the value of the correlation will be -1. + Itmeans r lies between +1 and -1. Within these limits, the value of correlation is interpreted as: Degree of Correlation Degree of Correlation Positive Correlation ‘Negative Correlation Perfect Correlation +1 -1 High degree of Correlation Between + 0.75 and +1 Between - 0.75 and-1 Moderate degree of Correlation _| Between + 0.25 and + 0.75 Between - 0.25 and- 0.75 Low degree of Correlation Between 0 and + 0.25 Between 0 and - 0.25 No Correlation (Zero Correlation | 0 0 or Uncorrelated)

You might also like