Linear Correlation and Regression. [P] [N] This page will calculate the Pearson product-moment correlation coefficient for a set of bivariate XY values, along with all other values normally calculated in a correlation/regression analysis. As the page opens, you will be prompted to enter the number of paired bivariate values of X and Y in your data set.
0.95 and 0.99 Confidence Intervals for rho. The correlation, r, observed within a sample of XY values can be taken as an estimate of rho, the correlation that exists within the general population of bivariate values from which the sample is randomly drawn. This page will calculate the 0.95 and 0.99 confidence intervals for rho, based on the Fisher r-to-z transformation.
Significance of the Difference Between Two Correlation Coefficients. Using the Fisher r-to-z transformation, this page will calculate a value of z that can be applied to assess the significance of the difference between two correlation coefficients, ra and rb, found in two independent samples. The page includes a frame showing a version of the conventional table of the unit normal distribution to which the calculated z-ratio can be referred.
Partial Correlation. Given three overlapping correlation coefficients, r(ab) , r(ac) , and r(bc) , this page will calculate the partial correlations r(ab.c) , r(ac.b) , and r(bc.a) .
Rank Order Correlation. This page will calculate rs, the Spearman rank-order correlation coefficient, for a set of paired bivariate rankings. As the page opens, you will be prompted to enter the number of items for which there are paired rankings.
Multiple Regression. These pages will perform a basic multiple regression analysis for the situation where there are several independent or predictor variables, X1, X2, etc., and one dependent or criterion variable, Y. The first batch of pages require that you already have the matrix of correlation coefficients for your several variables. The second batch will perform the same analyses starting out with raw scores. Please note in all cases that n, the size of each of your several samples, should be equal to or greater than the number of independent variables plus 2. (As the pages of the second batch open, you will be prompted to enter the value of n.)
If you already have the correlation matrix:
For 2 independent variables plus Y
For 3 independent variables plus Y
For 4 independent variables plus Y
If you are starting with raw scores:
For 2 independent variables plus Y [P]
For 3 independent variables plus Y [P]
For 4 independent variables plus Y [P]