Overview of Exploratory Factor Analysis (EFA) and how to run EFA in SPSS

What is EFA

Before testing scientific theories it is necessary to evaluate the reliability and validity of the scale. Cronbach’s Alpha method used to evaluate the reliability of the scale. Exploratory Factor Analysis ( EFA) help us to check convergent value and discriminant value.
EFA have no dependent variable and independent variables, it only rely on correlations between variables together (interrelationships). EFA to shorten a set of K observed variables into a set F (F <K) the more significant factor. The basis of this reduction is based on a linear relationship of the factors with the original variables (observed variables).
The authors Mayers, LS, Gamst, G., AJ Guarino (2000) mentioned that: In the factor analysis, methods of extraction Principal Components Analysis, Varimax rotation method used is the most popular.
(Hair, Anderson et al. 1995a) categorized the Factor loadings as 0.30 = minimal, 0.40 = important, and 0.50 = practically.  If the Factor loadings is less than 0.30, then it should be reconsidered if Factor Analysis is proper approach to be used for the research (Hair, Anderson et al. 1995a; Tabachnick and Fidell 2001). If the correlation matrix is an identity matrix (there is no relationship among the items) (Kraiser 1958), EFA should not be applied.
The sampling adequacy can be assessed by examining the Kaiser- Meyer -Olkin (KMO) (Kaiser 1970). It ranges from 0 to 1, while according to (Hair, Anderson et al. 1995a; Tabachnick and Fidell 2001) , 0.50 considered suitable for FA . On the other hand, (Netemeyer, Bearden et al. 2003) stated that a KMO correlation above 0.60 – 0.70 is considered adequate for analyzing the EFA output.
Bartlett’s test of Sphericity (Bartlett 1950) provides a chi-square output that must be significant. It indicates the matrix is not an identity matrix and accordingly it should be significant (p<.05) for factor analysis to be suitable (Hair, Anderson et al. 1995a; Tabachnick and Fidell 2001).

Explore Factor Analysis must satisfy the following requirements:

    -Factor loading> 0.5
    -KMO ≤ 0.5 ≤ 1
    -Bartlett test the statistical significance (Sig. <0.05):
    -Percentage of variance in Extraction Sums of Squared Loadings > 50%: Show percentage variation of the observed variables.  

How to run EFA in SPSS.

Select on the menu: Analyze-> Data Reduction -> Factor.

Select all needed variables to the  Variables column on the right.

Click Descriptives, check KMO and Bartlett's test of sphericity

Click Rotation button, select Varimax

Click the Options button, select Sorted by size and select Suppress absolute values less than, type in .3

Then click OK, the results will show as follow, including Rotated Component Matrix table as follows:

There are 5 factor appear. That all about EFA

Contact Master of Business Administration Group at : https://facebook.com/hoidapspss or email: hotrospss@gmail.com or
We can help you:
– Market survey / processing / editing the survey data to run a factor analysis results converged, the regression analysis statistical significance.
– Consult models / questionnaires / Traning directly on regression analysis, factors, Cronbach alpha … in SPSS, and SEM model, CFA, AMOS

Leave a Reply

Your email address will not be published. Required fields are marked *