Skip to main content
. Author manuscript; available in PMC: 2008 Aug 20.
Published in final edited form as: Maturitas. 2007 Apr 23;57(4):347–360. doi: 10.1016/j.maturitas.2007.03.003

Figure 1.

Figure 1

The statistical modelling process using log total testosterone (T) as an example. The progression of potential predictor variables through bivariate, intermediate multivariate analysis within variable class and final multivariate analysis was identical for FAI and DHEAS modelling. Variables had to pass significance (P < 0.05) to progress from bivariate to intermediate multivariate analysis.