Below you can find a list of frequently asked questions, organized by topic, that reach us via email. Click the question to see our response.

Time-varying covariates

How do I include time-varying covariates with the RI-CLPM?

Is it possible to run an RI-CLPM with three (or more) outcomes?


How should I interpret the standardized cross-lagged and autoregressive parameters?

How do standardized cross-lagged and autoregressive parameters compare to explained variance?

How can I constrain the standardized parameters to be invariant over time?

Non-continuous outcomes

Can I run the RI-CLPM with binary/categorical/count outcomes?


How do I perform a power analysis for the RI-CLPM?


Why are the autoregressive effects in the RI-CLPM typically smaller than in the CLPM?

Is it a bad sign that the standard errors are typically larger in the RI-CLPM than in the CLPM?

Growth and development

How can the RI-CLPM incorporate growth over time?

Response to Orth et al. (2021)

Lately, we have received numerous questions about statements regarding the use and appropriateness of the CLPM and RI-CLPM in “Testing Prospective Effects in Longitudinal Research: Comparing Seven Competing Cross-Lagged Models” by Orth et al. (2021). In our opinion, numerous conclusions herein are incorrect. Below we elaborate on some of their most prominent conclusions. For another reaction to Orth et al. (2021), see

The RI-CLPM is not suited for studying prospective between-person effects, whereas the CLPM is.

The RI-CLPM is better suited for short-term studies because it cannot detect sustained prospective effects.

There should be a match between the type of research question asked and the model used: The CLPM is for the analysis of between-person prospective effects whereas the RI-CLPM should be used for within-person prospective effects.

The parameters from the CLPM tend to have smaller standard errors than the parameters from the RI-CLPM; hence, the results from the CLPM are easier to replicate, which may be a reason to prefer this model.