I'm stuck with no help trying to figure out my dissertation for my masters. i conducted a survey with 273 respondents with 38 variables under the utaut2 model. iv been trying to work out my statistics and the KMO and Cronbach's alpha both are negative for all my variable sets. so i tried recodeing, removing the irrelevant ones, and also each set consists of 4-6 variables. please give me some valuable advice as to what wrong have i done and how could i fix it. im no expert in SPSS however i tried all the videos but no help.
Zeeshan Hassen, I have never heard of a KMO index being negative. Are you sure you're obtaining that index, or might you be referring to something else. I think the same applies to coefficient alpha. I've never seen it being negative, so, again, I wonder what is happening in your analyses.
If you would like me to help you further, but don't want to make it public, you could contact me privately through RG. Let me know if you would like to do that but don't know how.
Robert Trevethan thank you, sir. I did drop a message to you with regard to the issue I'm facing. many thanks for your valuable advice.
Bonjour , le test KMO varie entre 0 et 1. S’il est proche de 0, les corrélations partielles sont identiques aux corrélations brutes. Dans ce cas, une compression efficace n’est pas possible. Les variables sont deux à deux orthogonales. S’il est proche de 1, nous aurons un excellent résumé de l’information sur les premiers axes factoriels....... il faut vérifier la saisi des données dans SPSS peut être il y a une erreur
Chafia Hadjajd, thank you for your contribution. I hope you don't mind me taking the liberty of running your post through Google translator to get the following:
Hello, the KMO test varies between 0 and 1. If it is close to 0, the partial correlations are the same as the raw correlations. In this case, effective compression is not possible. The variables are two by two orthogonal. If it is close to 1, we will have an excellent summary of the information on the first factor axes ....... we need to check the data entry in SPSS maybe there is an error.
I think the word "compression" might not have translated properly, but otherwise your post looks good to me.
I agree with you: I think that there could be a data entry error, or there could also be errors in the requests made of SPSS.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS