Since you already have a basis for assigning the items to the factors, you should do Confirmatory Factor Analysis. This is not available within the basic version of SPSS; instead, it requires and additional module called AMOS.

Just as Sir David Morgan mentioned that since you already have a basis, i.e. 19 questions on the basis of 5 themes, a PCA ll show if they are represented in 5 themes only. Alternatively you may then run a CFA.

Also, data collected through likert item is not continuous and hence, check your data for normality to consider them as continuous data. You may use Shapiro Wilk test.

Hi Tara Ashton ,

I echo my fellows' answers. If the scale is well-established theoretically, run a CFA with bootstrap. If fit indices were not satisfactory, you'd need to do an EFA to re-validate (re-specification) the variate and confirm the new structure using CFA. See examples below.

Article Political Advertising Effectiveness in Wartime Syria

Article Total Quality Management Boosters and Blockers in a Humanita...

Article Assessing Patients’ Perception of Health Care Service Qualit...

Article Validating a New TQM-Benchmarking Measurement Model in an In...

Article Email is Evil! Behavioural Responses towards Permission-base...

Many thanks,

Ali

Draw second order measurement model using Ibm-Spss-Amos to represent all five components measuring a particular construct. Run CFA for this second order measurement model. Compute the AVE for convergent validity, assess fitness indexes for construct validity, and develop the table for convergent validity index summary.Also compute the Composite Reliability. If every figure achieve the rewuired level then you are done.

Hi Tara,

All your questions revolved around statistical analysis using SPSS. I advice you to go through the Book titled "Discovering Statistics Using IBM SPSS Statistics" by Andy Field. it is very helpful book for researcher in both perspectives; Statistical analysis and Use of SPSS.

Best Regards,

Indeed. I also fund Shafeq Hamoud M. Al-Saidi approach as quiet good. Here is another resource that might useful to you. SPSS survival Manual - Chap 15, page 172

https://www.mheducation.co.uk/openup/chapters/0335208908.pdf

if i understood correctly, you can use compute variables option to create 5 new variables from 19 questions

Zainudin Awang I have read your excellent book on SEM. You've mentioned a composite reliability score of greater than 6 is acceptable ( page # 63). But you have not given any reference. Can you pls share some references with page number?

Thanks