Linear Correlation 8 Scatter Diagram Method - No Correlation

• published: 10 Jan 2016
• views: 17

Statistics for all MBA - MCA - CA - CS - CWA - BBA - BCA - BCom - MCom - GRE - GMAT - Grade 11 - Grade 12 - Class 11 - Class 12 - IAS - CAIIB - FIII - IBPS - BANK PO - UPSC - CPA - CMA - Competitive Exams - Entrance Exams Linear Correlation (Correlation Analysis - Association between two variables) Scatter Diagram Method: This is very simple method of studying the relationship between two variables. If n pairs of observations (x1,y1), (x2, y2), (x3, y3) … (xn, yn) of two variables are given they can be plotted on a graph paper. The diagram formed by the points on the graph paper is known as scatter diagram. Looking to the distribution points in the scatter diagram we can know the direction of the relationship and also get a rough idea about the degree of the relationship between the variables. No Correlation or Absence of Correlation: * * * * * * * * * * * * * * INDEPENDENT VARIABLES If the points of the scatter diagram are distributed haphazardly or randomly, i.e. they are neither on a straight line nor around a straight line, then they exhibit absence of relationship between two variables. Two such variables are independent or we should conclude that there is no correlation between the two variables. Friends, I have given this topic because since decades and centuries we are taught like that. I personally don't agree with this kind of explanation. Should we wait till we get such type of scatter diagram to conclude that there is no correlation between the two variables under consideration? Certainly not... It is the matter of common sense on the part of the analyst to study the relationship between only those variable which are naturally correlated. Practically, no data can provide such type of diagram. If we take statistical data of two such variables which can never be correlated, then also we get a diagram of any of the first four kinds only. So, concluding that the variables under consideration are not correlated is not the subject matter of scatter diagram or any other method of the study of correlation. The method can be used to analyse the type or degree/extent of correlation and for that the analyst ought to take the variables which are naturally or really correlated. Merits and Limitations of Scatter Diagram Method: Merits: i. This is a simple method of studying correlation between two variables. ii. More mathematical knowledge is not required in this method. iii. If one of the pairs of values is extreme, it does not influence much in deriving the conclusion. iv. The type of correlation can be immediately known by this method. v. It is the first step in studying correlation between two variables. Limitation: This method gives an idea about the direction and to some extent the degree of relationship between the variables, but it does not give the exact measure of the relationship between the variables. - www.prashantpuaar.com