Regarding use of Q square values as sample predictive power of the model

20 views
Skip to first unread message

Oshin Gulsia

unread,
Sep 21, 2024, 3:12:41 AM9/21/24
to DataAnalysis
Respected Group members

I had recently submitted my thesis for final evaluation for which one of the evaluators flagged the issue of non-use of q square values as a tool for estimating sample predictive power of the model in our studies. The evaluator in this regard highlighted that the phenomenon of sensitivity analysis has not been addressed at all and must be addressed through calculating q-square values.

In the light of aforementioned facts, I would like to ask the procedure to calculate q square values or perform sensitivity analysis in our study involving model development through CB-SEM approach using AMOS software?

Kindly enlighten!
Thanks and Regards
Oshin Gulsia
Research Scholar at NIFTEM-K

Neeraj Kaushik

unread,
Sep 23, 2024, 8:42:29 PM9/23/24
to dataanalys...@googlegroups.com
Dear OG
In a couple of my videos I've narrated this story of the feud between CB-SEM and PLS-SEM.
PLS-SEM initially did not have the model fit indicators so they were always criticized by the scholars of CBSEM for the same. So, the developers of PLS-SEM came up with the idea of model comparison using prediction.
Now, we've the concept of predictive validity and CVPAT (Cross Validated Predictive Ability Test) which is used in SmartPLS software to compare 2 constructs or models for the ability to predict.
This is done using the concept used in Machine learning where 70% of the data is used to make a model and then the prediction is made for the rest of the 30% data.
All such are not available in CB-SEM and their manual calculations are next to impossible.

So, you have limited options now. Either try to convince the reviewer that model fit indicators, followed by validity and reliability of constructs is enough and the results of your analysis are robust enough. Plz do cite such papers. Else you've to change the journal.

Best wishes

--
The members of this group are expected to follow the following Protocols:
1. Please search previous posts in the group before posting the question.
2. Don't write the query in someone's post. Always use the option of New topic for the new question. You can do this by writing to dataanaly...@googlegroups.com
3. It’s better to give a proper subject to your post/query. It'll help others while searching.
4. Never write Open-ended queries. This group intends to help research scholars, NOT TO WORK FOR THEM.
5. Never write words like URGENT in your posts. People will help when they are free.
6. Never upload any information about National Seminars/Conferences. Send such information
in personal emails and feel free to share any RESEARCH-related information.
7. No Happy New Year, Happy Diwali, Happy Holi, Happy Birthday, Happy Anniversary, etc. allowed in this group.
8. Asking or sharing Research Papers is NOT ALLOWED.
9. You can share your questionnaire only once.
---
You received this message because you are subscribed to the Google Groups "DataAnalysis" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataanalysistrai...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/15199ac3-60d6-42cf-a430-a9fa46f04163n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages