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

This video is continues the presentation from the video at • Path analysis in R usi... on how to use Lavaan syntax to carry out a path analysis and obtain various model outputs. In this video, I specifically focus on how to test for indirect and total effects.
A copy of the text file referenced in the video can be downloaded here
drive.google.c...
A copy of the Powerpoint can be downloaded here:
drive.google.c...
drive.google.c...
A copy of the data can also be downloaded here
drive.google.c...
Additional videos and materials on SEM can be found at: sites.google.c...

[@chiarabanfi1751]

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

[@kristingibson3883]

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!!

[@yangyang6008]

Nice and clear! Thank you

[@danielshepherd8752]

Cooler than AMOS!

[@shuyingsha9023]

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.

[@ellesawyer4194]

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

[@CoreofScience]

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

[@mikecrowson2462]

You are very welcome! Thanks for visiting!

[@nguyenminhhieu504]

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.

[@abbychang6820]

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?

[@aylinkocak7585]

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

[@mikecrowson2462]

You're welcome! thank you for visiting!

[@jpbalam]

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

[@Jaji1948]

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

[@Jaji1948]

Never mind, the problem was the Safari browser. I got it to work using the Chrome browser.

[@kanitpitchaya9130]

thank you this great video

[@shailendrakumarsingh4342]

Thanks for spreading knowledge! Really helpful 👍!

[@mikecrowson2462]

You're very welcome! I'm glad you are finding it helpful! Best wishes.

[@eddieclarke2655]

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?

[@mikecrowson2462]

Hi there. I believe I might have covered this in the first video in this series: ua-cam.com/video/QP-v6RwsZjY/v-deo.html
Also the third video in this series can be found here: ua-cam.com/video/HvYW_GeHpD8/v-deo.html
Thanks for visiting!
Cheers!

[@eddieclarke2655]

​@@mikecrowson2462 Really helpful once again, Mike. All the best with your research.