Question about structural equation modeling
Locked
Question about structural equation modeling
 
|
I received a revise and resubmit on a manuscript focused on the establishment of a measurement model for a depression scale. One of the reviewers asked why I didn't use a validation sample. My question is, is it always necessary to have a validation sample. All opinions are welcomed. |
SV: Question about structural equation modeling
|
|
If this is a new scale and/or model, it would be nice to validate it. You cannot use the same sample for developing the model and validating it.
Maybe you could say that your goal was to establish the model, you leave it's validating to the future research or other researchers. It all depends on the context and pre-history of the scale/model. Raimundas Vaitkevičius Vytautas Magnus University [hidden email] Från: stace swayne <[hidden email]> Till: [hidden email] Skickat: torsdag, 10 oktober 2013 3:20 Ämne: Question about structural equation modeling I received a revise and resubmit on a manuscript focused on the establishment of a measurement model for a depression scale. One of the reviewers asked why I didn't use a validation sample. My question is, is it always necessary to have a validation sample. All opinions are welcomed. |
Re: SV: Question about structural equation modeling
|
|
Thank you for the feedback. I actually didn't develop the depression scale. The scale was used in a large study without prior psychometric testing. In my paper I establish the factor structure and then I did invariance testing across women and ethnic groups within the sample. On Thursday, October 10, 2013 8:23 AM, Raimundas Vaitkevicius <[hidden email]> wrote: If this is a new scale and/or model, it would be nice to validate it. You cannot use the same sample for developing the model and validating it. Maybe you could say that your goal was to establish the model, you leave it's validating to the future research or other researchers. It all depends on the context and pre-history of the scale/model. Raimundas Vaitkevičius Vytautas Magnus University [hidden email] Från: stace swayne <[hidden email]> Till: [hidden email] Skickat: torsdag, 10 oktober 2013 3:20 Ämne: Question about structural equation modeling I received a revise and resubmit on a manuscript focused on the establishment of a measurement model for a depression scale. One of the reviewers asked why I didn't use a validation sample. My question is, is it always necessary to have a validation sample. All opinions are welcomed. |
Re: Question about structural equation modeling
|
|
In reply to this post by stace swayne
For the validation sample thing, one possibility is to partition your original data file (assuming it is large) into two subsets: Say 80% for the training subset
and 20% for the validation subset. You could use the random selection capability in SPSS to create those subsets. Then, you could use the training subset to build the model and use the validation subset to validate it. The validation subset is new data because
it has not been used for building the model(s). You may want to fix the random seed at a particular value and provide it in the manuscript in case anybody is interested in replicating your results. Hongwei “Patrick” Yang, University of Kentucky P.S.: I do not see this validation thing done very often in published studies. But I could be wrong. From: SPSSX(r) Discussion [mailto:[hidden email]]
On Behalf Of stace swayne I received a revise and resubmit on a manuscript focused on the establishment of a measurement model for a depression scale. One of
the reviewers asked why I didn't use a validation sample. My question is, is it always necessary to have a validation sample. All opinions are welcomed. |
Re: SV: Question about structural equation modeling
|
In reply to this post by stace swayne
Stace I don’t know what scales you are using, but you’ll find some possibly useful references on page: “Measures of psychological well-being” on my site:http://surveyresearch.weebly.com/measures-of-psychological-well-being.html. There may even be some you’ve never heard of. John F Hall (Mr) [Retired academic survey researcher] Email: [hidden email] Website: www.surveyresearch.weebly.com SPSS start page: www.surveyresearch.weebly.com/spss-without-tears.html From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of stace swayne Thank you for the feedback. I actually didn't develop the depression scale. The scale was used in a large study without prior psychometric testing. In my paper I establish the factor structure and then I did invariance testing across women and ethnic groups within the sample. On Thursday, October 10, 2013 8:23 AM, Raimundas Vaitkevicius <[hidden email]> wrote: If this is a new scale and/or model, it would be nice to validate it. You cannot use the same sample for developing the model and validating it. Från: stace swayne <[hidden email]> I received a revise and resubmit on a manuscript focused on the establishment of a measurement model for a depression scale. One of the reviewers asked why I didn't use a validation sample. My question is, is it always necessary to have a validation sample. All opinions are welcomed. |
SV: SV: Question about structural equation modeling
|
|
In reply to this post by stace swayne
I guess you used CFA approach? You developed a new model (factor structure) using your sample. How can you be sure that your model is correct in some broader sense, not only in your sample? Generally, validation on another sample makes sense, however, I think, it should not necessarily been done in the same study and described in the same article. I guess that you simply didn't have such a goal at the present - to validate your model in another sample. Maybe it would be
sufficient only to explain this to the reviewer properly, probably making corrections to the manuscript in order to clarify goals and methods of your study. Från: stace swayne <[hidden email]> Till: [hidden email] Skickat: torsdag, 10 oktober 2013 17:23 Ämne: Re: SV: Question about structural equation modeling Thank you for the feedback. I actually didn't develop the depression scale. The scale was used in a large study without prior psychometric testing. In my paper I establish the factor structure and then I did invariance testing across women and ethnic groups within the sample. On Thursday, October 10, 2013 8:23 AM, Raimundas Vaitkevicius <[hidden email]> wrote: If this is a new scale and/or model, it would be nice to validate it. You cannot use the same sample for developing the model and validating it. Maybe you could say that your goal was to establish the model, you leave it's validating to the future research or other researchers. It all depends on the context and pre-history of the scale/model. Raimundas Vaitkevičius Vytautas Magnus University [hidden email] Från: stace swayne <[hidden email]> Till: [hidden email] Skickat: torsdag, 10 oktober 2013 3:20 Ämne: Question about structural equation modeling I received a revise and resubmit on a manuscript focused on the establishment of a measurement model for a depression scale. One of the reviewers asked why I didn't use a validation sample. My question is, is it always necessary to have a validation sample. All opinions are welcomed. |
| Free forum by Nabble | Edit this page |