First, Kaiser-Meyer-Olkin (KMO) test and Bartlett’s test of Sphericity were applied to demonstrate the suitability of data for factor analysis and the existence of factors. The results showed the factorability of the items (KMO=0.645; Bartlett’s test= 0.000). In order to identify the motivational dimension of homestay, 22 push and pull items were subjected to principal component analysis with Varimax orthogonal rotation using SPSS factor analysis.
Thus, Varimax rotation was used to refine the orthogonal factor matrix. Rotation of the initial solutions maximises variance loading within factor. Moreover, rotation of the original matrix assists in the recognition of the variables that best describe the factor. A factor loading of 0.30 was used as a cut-off for inclusion of any item among various factors. The results of the principal component factor analysis with Varimax rotation produced two broad push factors and two pull factors with eigen- values as shown in Table 2.
The two push factors explained 24.7% of the total variance and the two pull factors explained 26.5% of the total variance. The First Factor for push and pull explains the highest proportion of the observed variance in the data set. The second factor for push and pull accounts for the other variance not explained by Factor II.
Moreover, in order to test the reliability and internal consistency of each push and pull item, the Cronbach alpha was adopted. The results confirmed that the alpha coefficients for both factors ranged from 0.616 to 0.787. Although socio-cultural immersion was highest in terms of factor loading, it did not record the highest reliability. However, community service and development recorded the lowest values for the variance explained and Cronbach alpha respectively.