R/regr.R

In yhat: Interpreting Regression Effects

regr<-function (lm.out) 
{
    ifelse(length(lm.out$call) > 2, new.data <- eval(as.name(lm.out$call[[3]]), 
        parent.frame()), new.data <- model.frame(lm.out$call[[2]]))
    new.scale <- model.matrix(lm.out)
    v <- as.numeric(row.names(new.scale))
#   new.data <- new.data[v, ]
    IV <- attr(lm.out$terms, "term.labels")
    DV <- dimnames(attr(lm.out$terms, "factors"))[[1]][1]
    IVx <- dimnames(attr(lm.out$terms, "factors"))[[1]][-1]
    new.data<-na.omit(new.data[,c(DV,IVx)])
    new.scale[, 1] <- lm.out$model[, 1]
    new.scale <- data.frame(new.scale)
    colnames(new.scale) <- c(DV, IV)
    beta.out <- coef(lm.out)[-1] * sapply(new.scale[IV], "sd")/sapply(new.scale[DV], 
        "sd")
    structure.coef <- cor(na.omit(fitted.values(lm.out)), new.scale[IV])
    CCdata = commonalityCoefficients(new.data, DV, IV, "F")
    es = effect.size(lm.out)
    return(list(LM_Output = summary(lm.out), Beta_Weights = beta.out, 
        Structure_Coefficients = structure.coef, Commonality_Data = CCdata, 
        Effect_Size = es, Comment = "The Effect Size recommendations are based on Yin and Fan (2001). Your dataset may take on a different covariance structure, thus making another effect size estimate more appropriate."))
}