The traditional factor analysis approaches such as Pearson correlation and Cronbach's Alpha have some limitations .To bash the drawbacks of Pearson correlation and Cronbach's Alpha in the measurement of validity and reliability, Confirmatory Factor Analysis (CFA) using Structural Equation Modeling (SEM), have been used to test the validity and reliability of the instruments. Various tests i.e. Regression Weights, Standardized Regression Weights, Convergent Validity, Variance Extracted, Construct Reliability, and Discriminant Validity has been used to extract better validity and reliability. Should we do all the test to get better validity and reliability ?
Hakan Koğar Thank you very much for your courteous suggestion.
Hi Md. Israt Hasan
This is a good resource for your question: Discovering Statistics Using IBM SPSS Statistics 5th (2017) Edition by Andy Field
Dear Md. Israt Hasan ,
The first choice you need to make is whether you are exploring new factors or confirming existing factors. Andy Field's book only covers the former. Reliability analysis such as Cronbach's alpha is only for confirming a single factor. Please see my study guide on exploratory factor analysis.
Thanks Peter Samuels ! I will definitely go through your study Guide.
Md. Israt Hasan, from the way you have framed your question (which looks a bit to me like something you have copied from some other source), I am worried that you might not understand what confirmatory factor analysis (CFA) is used for.
I am happy for anyone to disagree with me, but my view is that CFA is used to identify the extent to which the supposed factor structure of a construct (based on prior theory or research) matches the "reality" when actual data are analysed.
Md. Israt Hasan, to examine the validity and reliability of a reflective measurement model estimated using CFA, you can assess and report the 1) size of the standardised loadings, 2) average variance extracted, 3) discriminant validity, and 4) Raycov's reliability coefficient (not Cronbach's Alpha).
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability