Additional file 1: Table S1 Bartlett and KMO Test Results
Bartlett KMO
Overall quality score 2281.219 (0.000) 0.615708
Infrastructure quality score 1804.877 (0.000) 0.641465
HR Human resources (HR) quality score 92.005 (0.000) 0.497766 Curriculum and Mmaterials quality score 90.438 (0.000) 0.580303
Note. P-values are in parentheses.
We apply Bartlett’s test of sphericity and Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy to the data in order to assess the appropriateness of PCA methodology. The former tests the hypothesis that the correlation matrix is identity, which means that the variables are uncorrelated. Rejection of this test indicates that the variables are correlated and PCA is suitable as a data reduction technique. The latter measure compares the correlations and partial correlations between variables. A high value of KMO indicates that a PCA is applicable for the data set. KMO values greater than 0.5 are acceptable as recommended by Kaiser (1974). As shown in Table A1, p-value for the Bartlett’s test is small enough to reject the null hypothesis of an identity matrix of correlations and KMO values are all greater than 0.5 except for the HR quality variable. However, KMO value for this variable is almost 0.5.
Hence, PCA methodology is suitable for our data set.