Hi Sandya,

I don't think it makes sense to model leadership and agility as common factors. As the figure shows, this model structure assumes that all the different styles have have one common underlying factor and this factor completely explains the correlations between

a) all specific styles and

b) between all specific styles and all dependent agility variables)

I would rather model all your indicators as variables in themselves (leadership and agility). This model will only have problems, if one of the styles mediates the effect of another style. Think about that.

If you want to learn about causal structure (and allow me a bit of self-ad), here is a paper with explains this (especially the cirumstances, underwhich controlling for a variables is a problem). The paper is about meta-analysis but the causal part (I think the first 3 section) can be applied to every SEM:

Steinmetz, H., & Block, J. (2022). MASEM: New tricks of the trade. Management Review Quarterly. https://doi.org/10.1007/s11301-022-00293-6 (https://link.springer.com/content/pdf/10.1007/s11301-022-00293-6.pdf)

Hope that helps

Holger

Thank you