Can SEM handle longitudinal data?
• Can one use simple variables in lieu of latent variables in SEM models? Yes, though this defeats some of the purpose of using SEM since one cannot easily model error for such variables. While one can simply substitute them for latent variables, this is making the very suspect assumption that they are 100% reliable. It is better to make an estimate of the reliability, based on experience or the literature. However, for a variable such as gender, which is thought to be very highly reliable, such substitution may be acceptable. The usual procedure is to create a latent variable (ex., sex) which is measured by a single indicator (gender). The path from sex to gender must be specified with a value of 1 and the error variance must be specified as 0. Attempting to estimate either of these parameters instead of setting them as constraints would cause the model to be underidentified, preventing a convergent solution of the SEM model. If one has a variable one wants to include which has lower
