Path analysis in R using Lavaan (video 2; Sept. 2019): Testing indirect and total effects 

Nice and clear! Thank you

This series of videos is very informative, thank you for taking time to explain everything in detail, it is greatly appreciated.

Thanks! I have been looking something like this for a while. Thank you for sharing the material

This is so very helpful! I have one question. If an exogenous variable does not have a direct effect to an endogenous variable but rather only goes through mediators, is this both the total effect and the total indirect effect? Thank you!!

Thanks a lot for this great video and the supplementary documents!

Thank Prof. Crowson for your very interesting video. I see: total effect = direct effect + indirect effect; however, i wonder this equation is whether right once considering the statistical significance of effects. Specifically, p_value of indirect effect is >0.05, so this effect is not significant (at 95% confidence level) and then should we regard it to estimate the total effect? And which level(s) (e.g. 0.1; 0.05; 0.01; 0.001) should be used in order to test the significance of effects and thus conclude about the relationship between variables? In your video, you guide to manually make equations for calculating total effect of each variable. In fact, there are many exogenous and endogenous variables in a model, possibly resulting in numerous mediated effects and (very) long codes. Is there any ways to make short codes for estimating direct, indirect, total effects for all variables simultaneously? Thank you for reading my questions and look forward to your answers.

This video series was super helpful, thank you!! I was wondering if you have any videos that discuss Random Effects in a path analysis?

Thank you for your sharing, Mike. I am trying to use lavaan for SEM which is containing moderator in my model. Could you share an example about how to do it?

Thanks for spreading knowledge! Really helpful 👍!

Cooler than AMOS!

thank you this great video

Hi Mike, it seems this code lavaan(model,data=processdata,se="bootstrap") does not work on my version of R studio. but if I use sem( ), instead of lavaan( ), I will get the output.

Thanks so much for this, Mike. This is great. Have you covered in any other videos how to obtain standardised effect estimates for such output?

How can I add a fixed effect using a lavaan package?

Mike, is the file link still valid? I couldn't get it to work.