Spearman Rank Correlation Analysis

1470 Words6 Pages
CORRELATION It is co-relation, and is defined as the intensity of association or strength of relationship between two variables. If the association is perfect, they will move together, correlation coefficient is +1, ie. if one increases, other will also increase and vice-versa. If the association is imperfect, the variable move in opposite direction and correlation coefficient is -1 ie. if one increases, other will decrease and vice-versa. When there is no relationship between the variables, increase or decrease in one variable in no way affect the other. In such cases, correlation is zero. Eg. Number of cattle and dogs in an area. In order to know whether there exist any relationship between the variables, we go for scatter diagram. In this…show more content…
Spearman rank correlation test does not assume any assumptions regarding the distribution of the data and is the appropriate correlation analysis when the variables are measured on a scale that is at least ordinal. The following formula is used to calculate the Spearman rank correlation: P= Spearman rank correlation di= the difference between the ranks of corresponding values Xi and Yi n= number of value in each data set PARTIAL CORRELATION It is the correlation between two variables. When the effect of third variable on them is removed or suppressed. The partial correlation between the one and two by suppressing the effect of third is denoted by r12.3 r_12.3= (-P_12)/√(〖P_11. P〗_22 ) If simple correlation between x1 and x2 is more than the partial correlation, it implies that association between x1 and x2 is supported or induced by the suppress variable x3. Eg. Age of a calf and its weight gain are positively correlated. Amount of concentrate fed is the third variable influencing the weight gain. If it is removed, correlation reduces and is considered as inducing

More about Spearman Rank Correlation Analysis

Open Document