Similarly, the 27-items of the destination personality scale was subjected to
exploratory factor analysis. Applying the same empirical and substantive
considerations in item trimming as for the destination image scale, a three-factor
model emerged. The factor solution was adequate, as all eigenvalues were greater
than 1 and accounted for 59.1 percent of the total variance in the data. The first factor
was labelled sincerity and explained most of the variance (26%). The second factor
was labelled as excitement, given that three of the scale items were the same as in
Aaker’s (1997) study. The excitement dimension explains 18 percent of the total
variance. The last factor was labelled conviviality and accounted for 16 percent of the
total variance. The factor loadings were reasonably robust to support the construct
validity of the scales (Churchill, 1979). All the factors had relatively high reliability
coefficients ranging from 0.69 to 0.81. Finally, the OLS regression result showed that
the three-dimensional destination personality scale was statistically significant in
predicting intention to recommend (R² = 0.23, F(3,144) = 14.34, p=0.00). Thus, this
result provided evidence for the criterion validity of the scale (Churchill, 1979).
Similarly, the 27-items of the destination personality scale was subjected toexploratory factor analysis. Applying the same empirical and substantiveconsiderations in item trimming as for the destination image scale, a three-factormodel emerged. The factor solution was adequate, as all eigenvalues were greaterthan 1 and accounted for 59.1 percent of the total variance in the data. The first factorwas labelled sincerity and explained most of the variance (26%). The second factorwas labelled as excitement, given that three of the scale items were the same as inAaker’s (1997) study. The excitement dimension explains 18 percent of the totalvariance. The last factor was labelled conviviality and accounted for 16 percent of thetotal variance. The factor loadings were reasonably robust to support the constructvalidity of the scales (Churchill, 1979). All the factors had relatively high reliabilitycoefficients ranging from 0.69 to 0.81. Finally, the OLS regression result showed thatthe three-dimensional destination personality scale was statistically significant inpredicting intention to recommend (R² = 0.23, F(3,144) = 14.34, p=0.00). Thus, thisresult provided evidence for the criterion validity of the scale (Churchill, 1979).
การแปล กรุณารอสักครู่..
