Unidentified issues in CFA AMOS

[Mohini Yadav]

Hi all,

I am getting the following error while imputing values in AMOS - "the model appears to be unidentified. Imputation can not be performed". Can you please help me in fixing this error? I have attached my model and an error message for your convenience.

Regards

Mohini

Research Scholar, DU

[RUCHIKA AGARWAL .]

Hi Mohini,

I think will removing an item in SWHR, you have removed constant 1 also. Try adding constant 1 manually, by double clicking on the SWHR and then assigning weight as 1.

Regards,

Ruchika Agarwal

IIM Bodhgaya

--
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/CABXydjrpAWBjVGjyDoSvNnD6n-O4%3DRu3_kgP-16TsYLa3CKF%3Dw%40mail.gmail.com.

[Mohini Yadav]

Thanks Ruchika for your response. But even after adding the constant 1 in SWHR, the problem remains the same. PFA diagram.

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CAEtrgT9NATkqu85dRkt_sSa-1hGyiiYrf_7aWcBzQ_eLi8sjhg%40mail.gmail.com.

[Neeraj Kaushik]

Dear Mohini

Are you sure that from SWHR, any of the outgoing arrows has the regression weight of 1 

It is not properly visible in the diagram.

Best wishes

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CABXydjrKBa5y-nLW95wL%2B_ksoxqJkUNaN8j83dChjHAF_XLB%2Bg%40mail.gmail.com.

[priyanka aggarwal]

Dear Neeraj sir, 

My name is Priyanka Aggarwal, I am a research scholar at Jamia Millia Islamia, Delhi. 

As mentioned by you in the recent workshop at Jamia Millia, (please correct me if I am wrong) the CFA Model appears to be wrong here. 

Possible errors:

1. As Ruchika mentioned aptly, the loading was not allotted to any of the items of SWHR.

2. The purpose of conducting CFA is to find the convergent and discriminant validity of the constructs (by assessing the correlation between all the variables, whether IDV or DV). So the researcher here is putting path arrow instead of covariance between TRnCOM, JA, SWHR with CA.

Solution:

1. Allotment of regression weight to one of the item of SWHR by double clicking on the arrow and add ‘1’ as weight.

2. Remove the one sided arrow and instead put double ended arrow.

3. Remove error term ‘e 23’, to run the correlation between all the variables.

Hope this would solve your query.

Best wishes

Priyanka Aggarwal 

Research Scholar 

Department of management studies 

Jamia Millia Islamia

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CAAd%3Dc8M1ZkhNDwfTfpK%2BKTGEDLDp6L3WJydSdUpvgWt5HBB4eQ%40mail.gmail.com.

[Mohini Yadav]

Yes sir, regression weight of 1 is on SWHR9. The problem got solved when I made all the arrows as double sided (covarience) beween IV and DV as well. 

Best regards

Mohini

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CAAd%3Dc8M1ZkhNDwfTfpK%2BKTGEDLDp6L3WJydSdUpvgWt5HBB4eQ%40mail.gmail.com.

[Neeraj Kaushik]

Dear Mohini/Priyanka

Double sided arrow may solve the current problem but it is not the proper solution.

Though by making all 2-sided arrows and running a CFA model, this problem will  be solved but the SEM model shd also run smoothly and we shd try to find where exactly the problem lies.

I've explained some of the common errors in this video AMOS-3.1 Colors and Common Errors in AMOS

Unidentification problem comes whenever there is any arrow without a reg wt of 1.

On deep scrutiny of Mohini's model, I could see one arrow from the error term e23 there is no reg wt of 1.

I will request Mohini to plz run the given model by putting reg wt of 1 on the arrow from the e23 and check whether this problem is solved or not?

Best wishes

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CABXydjpgdy3nS_ChX%3D%2BtBkuL9nnhe%2BL2JWordzcb4qhQGpouCw%40mail.gmail.com.

[Mohini Yadav]

I think I understood my error. Since there is no missing data in the file, data imputation is giving an error. Estimates required under SEM are showing up correctly (using calcuate estimate option). Please correct me if i am wrong.

Regards

Mohini

On Sat, May 11, 2024, 9:03 PM Mohini Yadav <mohini....@gmail.com> wrote:

Dear sir, I have added the regression weight 1 to e23 and re-run the model keeping one-sided arrows between IV and DV. There is now a new error coming up. I am attaching screenshots for your reference.

Regards

Mohini

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CAAd%3Dc8Ofu9pMB3GVDfhM0H81iS%3Dt_Q00y7-WsL%3Dbknm-%2BE6r6g%40mail.gmail.com.

[Mohini Yadav]

Dear sir, I have added the regression weight 1 to e23 and re-run the model keeping one-sided arrows between IV and DV. There is now a new error coming up. I am attaching screenshots for your reference.

Regards

Mohini

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CAAd%3Dc8Ofu9pMB3GVDfhM0H81iS%3Dt_Q00y7-WsL%3Dbknm-%2BE6r6g%40mail.gmail.com.

[priyanka aggarwal]

Dear sir, 

Thank you for your reply. 

Will keep this point also in mind. 

Regards, 

Priyanka Aggarwal 

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CAAd%3Dc8Ofu9pMB3GVDfhM0H81iS%3Dt_Q00y7-WsL%3Dbknm-%2BE6r6g%40mail.gmail.com.

[Hamza Naim]

Dear,

May I know how many responses you got from pilot study??

It's mainly because pilot study questionnaires are upto 10% of all the data, performing EFA and then CFA on the rest 90%.

--

[Mohini Yadav]

Dear Hamza, the sample I used for CFA was 386.

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CACP2aM54ygZEPDO_2tfiq_KOrXZEN4FFHziz34mgszVuTPjfWg%40mail.gmail.com.

[Hamza Naim]

Dear,

EFA?

Regards
Hamza Naim
Research Scholar
Department of Business Administration
Aligarh Muslim University, Aligarh.

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CABXydjoQ54DsS3Ym0V%2B12b9VJc_HLub4D89aHFhkdTtmo0urTA%40mail.gmail.com.

[Mohini Yadav]

I have not run EFA since the scales are adapted from standardized questionnaires and all of them have good reliability.

Regards

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CACP2aM6YGLKgAE05GLpxzEq4ureAYLVHotAwPaN1iUG7TGF_Pw%40mail.gmail.com.

[Hamza Naim]

Dear,

EFA for your data! Are you proposing any model? 

Regards
Hamza Naim
Research Scholar
Department of Business Administration
Aligarh Muslim University, Aligarh.

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CABXydjrt-dznN%2BWUG0jyOSO_FffK%2BwkNs%2BN11S29u%3DeQT4hKNg%40mail.gmail.com.