Items that did not load with any other item were
removed. For this purpose, the factor matrix of loadings, or
correlation between the items, and factors was used.
Additionally, items were removed if they had loadings less
than 0.40.(Hair et.al,1998).Items that loaded on more than one
factor were removed due to violation of the simple structure
factor solutions(only one loading on any factor for each
variable is accepted.
Items with an MSA (measure of sampling adequacy) less than 0.50 in the anti-image matrix were removed.
The literature has multiple methods for choosing the number of factors to extract, with the most recommended
being those with an eigenvalue of larger than one(Coakes & Steed, 2003).