Compute Cohen's f-square effect size for a hierarchical multiple regression study, given an R-square value for a set of predictor variables A, and an R-square value for the sum of A and another set of predictor variables B. The calculator computes the effect size attributable to the addition of set B, which can provide useful insights for analytics studies that rely on hierarchical regression.
Compute the F-value for a hierarchical multiple regression study, given an R-square value for a set of predictor variables A, an R-square value for the sum of A and another set of predictor variables B, the number of predictors in sets A and B, and the total sample size. The calculator computes the F-value associated with the addition of set B, which can provide useful insights for analytics studies that rely on hierarchical regression.
Compute the observed power for a hierarchical regression study. The calculator computes the observed power for a significance test of the addition of a set of predictor variables B to the hierarchical model, over and above another set of predictor variables A. Knowing whether a model had enough power to detect an expected effect may be useful for analytics studies that rely on hierarchical regression.
Compute the sample size required for a hierarchical multiple regression study. The calculator computes the minimum required sample size for a significance test of the addition of a set of predictor variables B to the model, over and above another set of predictor variables A, given the expected effect size, probability level, and power level. Knowing the correct sample size is very useful for ensuring that hierarchical regression-based analytics studies have enough power to detect the expected effect.
Compute the Type 2 error (or false negative) rate for a hierarchical regression study. The calculator computes the Type 2 error rate for a significance test of the addition of a set of predictor variables B to the hierarchical model, over and above another set of predictor variables A. Knowing the Type 2 error rate can be very useful when evaluating a hierarchical regression-based analytics study that did not detect an expected or hypothesized effect.