Correlación y Regresión

Homepage G.Conocimiento Servicios Enlaces 2008 Avisos y Noticias Escalas Decisiones Aplicaciones Descriptivas Inferenciales Tablas Cruzadas Gráficas Gerencia 2010 R de Pearson Cálculo de la T Valores de T Cálculo ANOVA Valores de F Números al Azar Revistas ADPU Reseñas Cambio de Status Etica 101 Sobreviviente CurriculumVittae Contacto Sistema UPR Olimpiadas 2008 Fotos y Videos El Tiempo Trabajo Reciente



Correlation & Regression

This section contains the following items. Details for each can be found by scrolling down the page.

°Basic Linear Correlation and Regression
°Matrix of Intercorrelations
°Multiple Regression
°0.95 and 0.99 Confidence Intervals for r
°Estimating the Population Value of rho on the Basis of Several Observed Sample Values of r
°Test for the Heterogeneity of Several Values of r
°The Significance of an Observed Value of r
°Significance of the Difference between Two Independent Values of r
°Significance of the Difference between an Observed Value of r and a Hypothetical Value of rho
°First- and Second-Order Partial Correlation
°Rank Order Correlation
°Phi Coefficient of Association
°Point Biserial Coefficient of Association
°Correlation for Unordered Pairs:
Intraclass Correlation and Eta2
°Simple Logistic Regression
Basic Linear Correlation and Regression
The following pages calculate r, r2, regression constants, Y residuals, and standard error of estimate for a set of N bivariate values of X and Y, and perform a t-test for the significance of the obtained value of r.

Data-Import Version.  Allows for import of raw data from a spreadsheet; for samples of any size, large or small.

Direct-Entry Version.  Values of X and Y are entered directly into individual data cells. This page will also work with samples of any size, though it will be rather unwieldy with samples larger than about N=50. As the page opens, you will be prompted to enter the value of N.

Matrix of Intercorrelations

Version 1 (data-import format). For any number of variables and any number of measures per variable.

Version 2 (direct-entry format for small samples). For up to five variables; as the page opens, you will be prompted to enter the number of measures per variable.
Multiple Regression

Basic multiple regression analysis for the case where there are several independent or predictor variables, X1, X2, etc., and one dependent or criterion variable, Y. Requires import of data from a spreadsheet.

0.95 and 0.99 Confidence Intervals for r
Estimating the Population Value of rho on the Basis of Several Observed Sample Values of rQ
Test for the Heterogeneity of Several Values of r
The Significance of an Observed Value of r
Significance of the Difference Between:T

Two Independent Values of r

An observed Value of r and a Hypothetical Value of rho
[Both are based on the Fisher r-to-z transformation.]
Partial Correlation

For Three Intercorrelated Variables: first-order partial correlations

For Four Intercorrelated Variables: first- and second-order partial correlations
Rank Order Correlation. As the page opens, you will be prompted to enter the number of items for which there are paired rankings. If you are starting out with raw (unranked) data, the necessary rank-ordering will be performed automatically.
Phi Coefficient of Association. For a 2x2 table of cross-categorized frequency data.
Point Biserial Coefficient. For the case where one variable is dichotomous and the other is non-dichotomous.
Correlation for Unordered Pairs: Eta2, Intraclass Correlation, & Resampling of r
Simple Logistic Regression. [The plain-vanilla version.]