Mediation Analysis using JAMOVI: Concept, Interpretation, and Reporting Mediation in Research
HTML-код
- Опубликовано: 21 авг 2024
- Mediation Analysis using JAMOVI: Concept, Interpretation, and Reporting. The video focuses on understanding the concept of Mediation, the types of mediation, how to interpret and report the results of mediation.
#ResearchCoach #JAMOVI #Mediation
Thank you so much for this comprehensive video Sir! This is a big help for a student like me, in making our research.
Glad it was helpful!
I found this video very clear and useful, thank you!
Pleasure. I am glad you liked it.
@@researchwithfawad Thanks a lot for the video, Jamovi has a very different way of reporting Mediation and this was very clarifying. Cheers mate!
thank you sir all the points made are crystal clear to me now..
Glad to hear that
excellent sir
Many many thanks
Simply explained, thanks
Glad it was helpful!
Much appreciated
Thank you
Thank you for your video! I wish my professors would explain like this! I have one question: I did a mediation model with GLM Mediation model and it shows signifikant indirect mediation paths. BUT if I try to build the same model with pathj it shows horrible model fit (x^2, rmsean CFI). How is this possible?
Thanks for the appreciation.
Model fit and significance of relationships are two different things. A significant model may have a poor fit. A number of factors affect model fit that includes factor loadings, data distribution, and others. To improve fit, please check for modifications indices, Standardized Residual covariance, factor loadings, data distribution, missing data, respondent misconduct.
@@researchwithfawad Thank you for the quick answer! The GFI = 0.994 is the only Indices that fits. But the RMSEA is 0.310 and is way to high :( Can I still say, that X is a mediator for Y? Or is that only possible if the model fits?
Do you possibly also offer consultation appointments via zoom?
Yes, you can still say that the M mediates the relationship. As you are assessing the relationship not the fit.
You can email me on kh.fawad83@gmail.com
Thank you for your video. When using the GLM Mediation Model in Jamovi, the output includes both an "Estimate" and a "Beta". What is the difference between these two numbers and should the "Estimate" still be reported as the Beta?
Yes, if you want to report relationship between IV to MV, MV to DV, and IV to DV.
Good day, sir! Thank you so much for making this video, it's very helpful. Currently I'm doing a research and the result that I got was pretty much the same as yours, as in the estimate in Total is higher than the estimate in Direct. And also, the mediation percentage in Total is 100%.
My question is, what could cause this? Does it mean that there are other variables outside of our research that affecting the result?
Thank you!
I am afraid i didn't properly understand the question. If the direct and indirect effects both are significant, there is partial mediation, if only the direct effect is is significant but not the indirect effect, this mean no mediation, if direct effect is insignificant by the indirect is significant then it is partial mediation.
@@researchwithfawad okay, thank you for the explanation anyway! It was helpful. I'd do some more reading on mediation analysis. Have a good day!
The Mediation playlist may also help
ruclips.net/p/PLb7vm6tsQ3KvsREb6illLUHLVuk7LZxCY
may i know how to interpret below for a simple mediation :
Indirect Effect: 0.47
component: 0.31
component 0.15
Direct .043
Total .09
All with significant results but seems the numbers are very small
Does this considered as a weak mediation?
Do I have to address in my paper?
Thank you so much
Please watch
ruclips.net/video/8lDwmMF9aN8/видео.html
Thank you for making this video, Sir. I have a question for you, i used jamovi to analyze my data and the result was Estimate of the direct effect higher than indirect effect (it's similiar with the example on this video). So, what's the meaning of the result? I mean, what the explanation about the estimate direct effect higher than indirect?
Direct effect is the impact of IV on DV in presence of the mediator whereas indirect effect is impact of IV on DV through Mv
@@researchwithfawad if the estimate in direct higher than indirect, does it mean direct effect more significant than indirect?
The effect passing directly is higher than the indirect.
@@researchwithfawad Thank you for your explanation, Sir. Have a great day!
Happy Day Sir
Q1. Is doing Mediation analysis in Jamovi is same as Process Macros (Model 4) same?
Q2. If No, what is the difference, sir?
(While doing the mediation in the SPSS process Macros the interpretation is really confusing kindly please help me by sharing a video on it sir especially with bootstrapped results sir)
Do you have any sample article that differentiates these two approaches?
with respect and regards swetha
Yes they use the same procedure
Kindly need a session on Mplus
Soon inshaa allah
was he meant to say with the inclusion of the mediating variable SQ the impact of CSR on CL was still found significant? He put the impact of CSR on SQ was still found significant. I am confused which is correct because the inclusion of the mediating variable SQ the impact of CSR on SQ was still found significant sound wrong to me. I want to know as I am doing a mediation analysis for my dissertation and I want to see how to report it.
If you are testing mediation.
If the specific indirect effects are significant, there is mediation.
If the impact of IV on DV in presence of Mediator is still significant, its partial mediation, otherwise full mediation.
You may also watch this video
ruclips.net/video/RjnfSVm3ev4/видео.html
Thank you, sir, if we want to test a full mediation effect only, is this possible and acceptable?
Yes, do not consider the direct effect.
selamunalykum. greeting from turkey. how can calculate the impact of more than one mediator on the dependent variable by jamovi
Walaikum assalam.
Please visit this link
jamovi-amm.github.io/glm.html
@@researchwithfawad thank a lot
Hello sir, thank you very much for this. Can you do moderation too?
Yes, but i haven't yet made a video on it. But hopefully soon.
If my data isn't normal (i analyzed by K-S), what should i do? I choose the starndart or bootstrap 1000? Thanks sir!
if it does not meet the normality assumption of traditional statistics, it’s a good candidate for bootstrapping. Although, the large sample size might let us bypass this assumption. Go for Bootstrap 5000
@@researchwithfawad My sample size is 682. I assumed the data were normally distributed, but they weren't.
I have done Boostrap 5000 and the results from confident interfal also pass "zero / 0". I'm sad that my hypothesis was rejected. can i know what i did wrong with this research sir?
How did you compute for the composite scores?
You may watch this video
ruclips.net/video/4goUT0b5dZc/видео.html
There is a table in the output called Path estimates why are we not reporting it?
Estimates are already reported through the output provided.
i dont see medmod option in jamovii interface
ruclips.net/video/wppxH8em9aA/видео.html
Please watch the video
Jamovi?
It is a free software for statistical analysis
@@researchwithfawad Thank you for your knowledge sharing.