Contents
Can you use SEM with ordinal data cross validated?
I am trying to determing how parental involvement influences motivation and results AND to create a single latent variable using CFA that represents parental invovlement. SEM can be applied to ordinal data. In fact, it is a “cheap trick” that allows one to ignore issues of non-normality (or skewness and kurtosis).
What happens if the ordinal variable is not normal?
But, if the observed y’s (lower case indicating ordinal variables) are not “normal”, this does not imply the preliminary Y (upper case indicating predicted latent variables) are not normally distributed. Again, if we can’t actually “see” the variable, there is no way to assess if it is normal or not.
Why are skewness and kurtosis calculated for ordinal variables?
Because the placing of these thresholds (tau’s) can be estimated anywhere, it is possible to generate very non-normal distributions amongst the observed variables. Again, happy to share more if this seems useful. Skewness and Kurtosis are not typically calculated for ordinal variables.
How are categorical variables included in structural equations?
There are two way that categorical variables could be included: as exogenous (predictors) or endogenous (responses). We will deal with the simpler case of exogenous categorical variables first, as they pose not so much of a computational issue, but a conceptual one.
Which is an example of a categorical variable?
For example: binary (yes/no, failure/success, etc.), nominal (site 1, site 2), or ordinal levels (small < medium < large). Newer advances in modeling allow for the incorporation, either directly or in a more roundabout fashion, of categorical variables into SEM.
Which is an example of a discrete variable in SEM?
While SEM was initially derived to consider only continuous variables (and indeed most applications still do), it’s often the case–especially in ecology–that the observed variables are discrete. For example: binary (yes/no, failure/success, etc.), nominal (site 1, site 2), or ordinal levels (small < medium < large).