You can usa the JASP software, it´s free and avaliable at: https://jasp-stats.org/download

I think you can calculate omega using the covariance matrix. It has been a few years since I did this, but I think there is a formula to plug in the values and solve by hand.

This paper may help.

Article From alpha to omega: A practical solution to the pervasive p...

This paper explains about coefficient alpha and omega. It provides a framework for deciding in which situations to use which, and it provides the calculation methods.

Cho, E. and Kim, S. (2015). Cronbach's coefficient alpha: Well known bur poorly understood. Organizational Research Methods, 18(2), 207-230.

However, I am not sure why you need a reliability measure. The CFA you are planning with Mplus will give you what you need in order to evaluate the quality of the measurement for you modeling.

Watkins' (2013) Omega is a freeware program for calculating omega and omega hierarchical coefficients based upon the standardized coefficients obtained from a CFA/EFA bifactor solution. All you have to do is create a tab delimited text file with the parameters specified in the directions for the program and output is completed in a matter of seconds. Although I believe there are syntax commands for calculating omega coefficients in Mplus, I find the program to be more parsimonious. 

http://edpsychassociates.com/Watkins3.html

I guess it is a bit late, but maybe it helps future researchers.

In this paper  you can find formulas for Mplus

https://www.researchgate.net/publication/236640820_Reliability_Estimation_in_a_Multilevel_Confirmatory_Factor_Analysis_Framework

Article Reliability Estimation in a Multilevel Confirmatory Factor A...

You can usa the JASP software, it´s free and avaliable at: https://jasp-stats.org/download

Thanks . The explanation has been useful to me.

Can a low value of alpha be attributed to the different perception individuals will have on the same questionnaire item and to that reason haven a low alpha value does not means your questionnaire item is not reliable

Anyhow, if there is a different "face validity" (i.e. the way items are perceived by the subjects of the measurement process is differential for different subjects), it means that your test lacks in content validity.

(And no, if this is the reason for a low reliability coefficient, in any case this means your item is not reliable)

Currently I am tryng to figure out how to compute Omega and Omega hierarchical for the factors within a hierarchical model. Since the Omega calculator by Watkins is designed for bifactorial models, I wonder if the coefficients can also be applied to hierarchical models. In a bifactorial model, group factor-loadings and general factor-loadings are clearly distinguishable, whereas the loadings of measured variables on the general factor are only indirectly, fully mediated by the scales, in a hierarchical model. Can anyone help me?

I'm using Mplus to estimate Omega according to the book of Raykov and Marcoulides (2011). Here is the input I used

TITLE: CFA OF ALIMENTARY SELF EFFICACY;

DATA: FILE IS alim_data.txt;

VARIABLE: NAMES ARE r1-r45;

USEVARIABLES = r3 r6 r10 r36 r39 r40 r45 r24 r11 r12 r30;

ANALYSIS:

ESTIMATOR =MLMV;

MODEL:

f1 BY r3*(p1)

r6 r10 r36 (p2-p4);

r3 r6 r10 r36(p5-p8);

F1@1;

f2 BY r39*(q1)

r40 r45 r24(q2-q4);

r39 r40 r45 r24(q5-q8);

F2@1;

f3 BY r11*(r1)

r12 r30 (r2 r3);

r11 r12 r30(r4-r6);

F3@1;

MODEL CONSTRAINT:

NEW(COMP_REL1 COMP_REL2 COMP_REL3 COM_RELT);

COMP_REL1 =(P1+P2+P3+P4)**2/

((P1+P2+P3+P4)**2+P5+P6+P7+P8);

COMP_REL2 =(q1+q2+q3+q4)**2/

((q1+q2+q3+q4)**2+q5+q6+q7+q8);

COMP_REL3 =(r1+r2+r3)**2/

((r1+r2+r3)**2+r4+r5+r6);

COM_RELT=(P1+P2+P3+P4+q1+q2+q3+q4+r1+r2+r3)**2/

((P1+P2+P3+P4+q1+q2+q3+q4+r1+r2+r3)**2+P5+P6+P7+P8+q5+q6+q7+q8+r4+r5+r6);

OUTPUT:

CINTERVAL;

Hi Luis, please give more details about the parameters used in the model. I found it hard to replicate it without any additional information.

thanks,

Eve

JASP is fairly user-friendly, but does not provide a confidence interval for the omega. Is there an existing software that does?

Gargi Sawhney Hi, did you get an answer to your question?

Selina, I did not. I ran some additional analyses in that paper, but not the omega.

Gargi Sawhney ok great thank you anyway:)

Dear Gargi Sawhney and Selina Maria Weiler,

I believe this can be of help: https://files.eric.ed.gov/fulltext/ED575032.pdf

Thank you, Anabela!