Daniel posted on Wednesday, November 09, 2005 - 7:53 am
I ran a SEM with a combination of categorical and continuous indicator variables. I tested for several indirect effects with and without bootstrapping. Without bootstrapping, two out of three of my specific indirect effects were significant, p < .05, and the other wasn't close. With bootstrapping, however, one of the key paths was significant, p < .05, and the other was significant, p < .10. Which result is more valid, if such a thing exists, when modeling with continuous and categorical indicator variables, bootstrapping or not bootstrapping?
Hmm. I think you should send your input, data, output, and license number at email@example.com so we can see exactly what you are finding.
Daniel posted on Tuesday, December 06, 2005 - 12:11 pm
Ok, it's on the way
JOEL WONG posted on Friday, January 29, 2016 - 5:50 am
I am trying to understand when I can infer that two indirect effects in multigroup SEM (using bootstrapping) are significant different from each other. Assuming I have measurement invariance, if the Mplus output shows that the 95% confidence intervals of the indirect effect for group A (e.g., .01 - .05) doesn't overlap with the 95% confidence intervals for group B (e.g., (.06 - .10), is that sufficient evidence that the two indirect effects are significantly different?
Or do I need to use some other method, e.g., the Wald test to demonstrate that the two indirect effects are significantly different?
You can create the indirect effects in MODEL CONSTRAINT and create a new parameter that is their difference. You will obtain a z-test and p-value for this.
JOEL WONG posted on Saturday, January 30, 2016 - 4:31 am
Thank you, Linda. The MODEL CONSTRAINT is a good idea.
I guess what I want to know is, can we make an inference on significant differences between 2 indirect effects based solely on the Mplus output showing that the 95% confidence intervals for group A (e.g., .01 - .05) doesn't overlap with the 95% confidence intervals for group B (e.g., (.06 - .10)?
I am interested in this topic as I am comparing indirect effects, which involve multiple mediators, across three racial/ethnic groups. I already had results of bias-corrected 95% CI for each group using BOOTSTRAP in Mplus. According to Mplus User Guide (v. 7), TECH3 cannot be used in conjunction with the BOOTSTRAP option of the ANAYSIS command. I suppose I can test indirect effects using MODEL INDIRECT and request TECH3 output. In terms of requesting TECH3 output, should I do this separately for each racial/ethnic group? Or, should I include all three groups together in one Mplus syntax file? Also, what should I look for in the TECH3 output (estimated covariance/correlation matrices for the parameter estimates) to determine if indirect effect estimates are independent across three racial/ethnic groups or not? Thanks, in advance, for responding to my questions.
Thanks for the quick response! Yes, I have been looking into multiple group comparisons on specific indirect effects across three groups. Your previous response mentioned TECH3 output, that's how I got curious about how to find out whether indirect effect estimates across groups are independent or not.