Lord's paradox in latent change score modeling: An example involving facilitating longitudinal effects between intelligence and academic achievement

It has been claimed that intelligence causes academic achievement to increase over time, and also, conversely, time. This bidirectional facilitating longitudinal effect between rests on observed associations initial the change in an a subsequent measurement, vice versa. Here, we demonstrate, through simulating empirical data used previous research, such may be due regression toward mean rather than true effect. Regression occurs conditioning of value outcome variable. Researchers should aware this fallacy are recommended verify their findings with analyses without adjustment for outcome. • intell = intelligence; aa A before. However, (RtM) RtM Results verified by not value.