Query 1: How to get the t-values for standardized regression weights in SEM? AMOS output provides CR values cosponsoring to the unstandardized estimates only. In journal papers usually the standardized estimates are quoted along with the t-values and p-values. No std error values are also provided against standardized regression weights in AMOS.
Query 2: Can the CR values be considered as t-value.
From my own perspective, The t-values are the CR values. It is true that Amos does not provide SE, P-values, and CR values for standardized coefficients. Amos only provides these values for unstandardized coefficients. However, i believe inn my opinion that these values (CR, SE and p-values) for standardized coefficients are the same as those in unstandardized. Thus, an analyst can easily copy these values from the unstandardized coefficient table and fit same to the standardized. My opinion anyway.
First of all, you may refer to a previous question by Shabir Ahmad 3 years ago who asked the following question: How can I get t value in Amos?
In fact CR value in Structural Equation Modeling output is the t-statistics value.
Dear Prof. Judeh, i intended to know how to obtain the t- value ( which is CR in case of AMOS output) in case of standardized regression estimates. As far as the unstandardized estimates are concerned, AMOS output provides the required Std. error, CR value and the p-value. I find that in some SEM analysis papers, authors provide tables with standardized regression estimates along with the t-values and p values, which i am unable to find in AMOS output. Again, Thanks for your response.
Hi Kaustav,
typical software packages (may it be for SEM or traditional regression) don't print the SE for standardized coefficients and hence, no t-values. It may be that Mplus does this, but usually, the SE's and t's are for the unstandardized coefficients (even if you saw them in papers seemingly associated with the standardized coefficients. In my opinion, there is no problem with that although - in principles - there could be cases where the unstandardized coefficient is significant but not its standardized version. But this is beyond my knowledge.
HTH
Holger
Dear Dr. Steinmetz , Thank you for the insights...it really helped for my paper.
From my own perspective, The t-values are the CR values. It is true that Amos does not provide SE, P-values, and CR values for standardized coefficients. Amos only provides these values for unstandardized coefficients. However, i believe inn my opinion that these values (CR, SE and p-values) for standardized coefficients are the same as those in unstandardized. Thus, an analyst can easily copy these values from the unstandardized coefficient table and fit same to the standardized. My opinion anyway.
Dear Valentine Joseph Owan ,
Is it available to fet t-values and p-values for direct effect, and indirect effects, and main effects?
or it is available in the table of regression weights only ?
The regression table in Amos comes with the standardized and unstandardized estimates, Standard Error, and t-values (Critical ratios CR), and p-values. This is usually for direct effects, Main effects but not indirect effects. Direct and indirect effects comes in another separate table usually below in the Amos Output.
Lavaan package in R allows to extract standardized p values with standardizedSolution().
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression