A Robust Indicator Mean-Based Method for Estimating Generalizability Theory Absolute Error and Related Dependability Indices within Structural Equation Modeling Frameworks
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsBackground: is there a relationship between G coefficient and D coeffcient? If yes, please describe it shortly in text.
With all those partial variances (equations 1-3) added io´nto the model, is not the model saturated and therefore not so effective?
Equation 5, 8 an 10 are equal. Authors should define and refer to only one and remove the other. Equation 10 is just a generalization of the other one. (Same for 14, 15 and 16).
There are a lot of equation in the manuscript, if authors could reduce them so that it would be easier to the reader to follow along the text, would be more interesting.
Methods: what about the validity and reliability of the chosen scale(MUSPI)? Nothing was reported.
Author Response
Our responses were uploaded in a separate file.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper present a new approach, indicator mean, to calculate absolute error in Generalizability Theory using SEMs (Lavaan and SemTools packages), while estimating one, two-, and three-facet designs with varying numbers of scale points. It was used a sample of 511 college students surveyed using the Qualtrics platform on two moment with a week difference obtained scores from MUSPI (Music Self-350 Perception Inventory).
General Comment:
The paper is well-written and provides a much more reliable approach to Generalizability Theory (GT) in terms of scale analysis. The authors give a solid theoretical introduction to their analysis proposal, which includes the option to use binary or ordinal items.
The use of tables and figures is appropriate
Author Response
We thank the reviewer for his or her positive appraisal of our original submission. The reviewer did not make any further suggestions for revision.
