# Regression in SEM programs vs regression in statistical packages such as SPSS

Contex

Multiple regression with continuous variables.

Conventional statistical packages, e.g. SPSS, lm() in R, typically give me an F value, dfs, and a significance test on whether the model is performing well. I can also use ANOVA to compare whether one model is significantly better than another by finding out whether the increase in R square is significant.

Structural equation modeling programs, e.g. Mplus, lavaan package in R, also allow me to run regression with added benefits of missing data handling through full information maximum likelihood. Even though parameter estimates and R squares are similar between these two approaches, I do not get the F value and the significance testing of model doing regression in SEM style. Instead, I get model fit information with df being 0, and I cannot figure out how I could test whether one regression model is better than another.

Questions

1. What are the differences between these two approaches, if any?
2. In SEM style, how do I know a regression model is good, and how can I compare whether one regression model has significantly larger R square than another?
• You might find this interesting: "Eight Myths about Causality and Structural Equation Models", ftp.cs.ucla.edu/pub/stat_ser/r393.pdf – Stijn Mar 7 '13 at 15:31
• Thank you @Stijn ! It looks very helpful. I copied the myths discussed in your PDF here in case people want to have a quick glance: – Sootica Mar 7 '13 at 21:02
• #1 SEMs aim to establish causal relations from associations alone. #2 SEM and regression are essentially equivalent. #3 No causation without manipulation. #4 The potential outcome framework is more principled than SEMs. #5 SEMs are not equipped to handle nonlinear causal relationships. #6 SEMs are less applicable to experiments with randomized treatments. #7 SEM is not appropriate for mediation analysis. #8 SEMs do not test any major part of the theory against the data. – Sootica Mar 7 '13 at 21:09
• SEMs do not aim to do anything, no more than a hammer aims to drive nails. They are a mathematical statement about how you think the universe works that can be compared to the observed data. Any causal and statistical inferences are for the user to make, not the model - just like with any other statistical model. There is no magic in SEM but when they are presented as such (usually as some kind of straw-man), it discourages people from using a potentially useful method. – D L Dahly Mar 8 '15 at 12:03

In SEM, you model the covariance matrix of everything: the regressors and the dependent variable. The covariance matrix of the regressors have to be unconstrained. The covariances between the dependent variable and the regressors is what generates the coefficient estimates, and the variance of the dependent variable, the $s^2$. So you've utilized all the degrees of freedom (number of covariance matrix entries), and that's why you see a zero. You should still be able to find $R^2$ in your output, but it will be hidden deeply somewhere, not thrown at you as in regression output: from the point of view of SEMs, your dependent variable is nothing striking, you may have a few dozen regressions in your output, and you can get all of their $R^2$s, or reliabilities, somewhere, but you may have to ask for it with some TECH options in Mplus.