How are RMLCA and LTA different?
Both RMLCA and LTA allow examination of patterns over time, but do so in different ways. In RMLCA, the latent classes are defined based on patterns of behavior over multiple occasions. This helps identify and characterize the common patterns that emerge during a behavior change attempt. In LTA, several latent classes may emerge at each time point (depending on the complexity of your indicators) and you can get estimates of particular transitions of interest. RMLCA will permit more repeats than LTA but also offers less specific information about transitions between particular time points.
In our work, we have chosen to use both RMLCA and LTA. We have used RMLCA to characterize patterns of smoking status over a relatively long time period (27 days) and are using LTA to examine key transitions (into and out of abstinence, for example) from week to week in the first weeks of a quit attempt. Analyzing data by week in the LTA was suggested by examination of the latent classes that emerged in the RMLCA, as these results suggested that changes in smoking probability occur in some classes at the end of week one and again at the end of week two during the quit attempt.