Darhub is a SAS data step program that calculates Darlington’s (1990) F statistic for testing the significance of difference between the obtained squared multiple correlation coefficient, R2, and the expected value, or expected long-run mean, of R2 under the null hypothesis that ρ = 0.  Darhub also computes Huberty’s (1994) adjusted R2 index and Huberty’s effect size measure for multiple regression studies.  Both of these indices take into account the expected value of R2.

Size: 1.75kb.

Source Reference:
Hittner, J.B. (2016). On correctly adjusting the squared multiple correlation coefficient in linear regression: Effect size estimation and significance testing with application to substance abuse research. Journal of Drug Abuse, 2(2), 1-4.

The performance of Huberty’s effect size, R, was examined in the following article:
Hittner, J.B. (2009). Effects of population distribution, sample size and correlation structure on Huberty’s effect size R. Journal of Modern Applied Statistical Methods, 8, 60-64.

Posted: 3.May.2016