|
|
Statistically comparing effect sizes ... |
|
Message/Author |
|
John Coffey posted on Tuesday, February 19, 2019 - 8:42 am
|
|
|
In my longitudinal mplus model, I am predicting IQ with two factors (factor a and b) in the same model. What is the best way to statistically compare the effect sizes (sizes are not that different) for a reviewer to verify if one is significantly bigger despite significant p-value difference? Ex. Factor a standardized path to IQ is .20 (p < .05) and Factor b standardized path to IQ is .13 (p = .15). Thanks for the help. |
|
|
Significantly bigger can't be tested in current Mplus but you can test if they are equal using Model Constraint. |
|
Back to top |
|
|