The estimation and hypothesis testing of treatment effects in nonequivalent control group designs when continuous growth models are assumed