Microsoft Word 2007 ichrie conference Proceedings Final-Final 06-06-07. doc

 
2007 Annual International CHRIE Conference & Exposition 
499
Finally, different profiles were described among different clusters. A series of ANOVA (PROC GLM) and 
Chi-square tests (PROC FREQ) were employed to reveal any statistically significant difference between various 
groups’ socio-demographic variables.
RESULTS 
The adjusted total of returned went down to 217, which included surveys that had been partially or 
completed finished. The final response rate was 73.3%. The average age of the respondents was 69.31 years, with a 
standard deviation of 7.60. More than half (56.5%) of the respondents were female. More than three-quarters 
(79.2%) were married. Except one reported racial background as “alien,” all the other respondents were white 
(99.5%). Due to the lack of variance in the race variable, race was ignored in further analysis. More than 80% of the 
respondents received a four-year college education (38.0%) and completed graduate school (43.5%). A majority 
(95.3%) of the respondents were retired. More than half (57.9%) of the respondents migrated to the Upstate area, 
and another 18.5% of the respondents moved within the Upstate after they were 50 years old. The average length of 
residence in the Upstate was 21.81 years (SD=19.50), while the average length of residence at the present house was 
13.11 years (SD=11.99). More than a quarter of the respondents had an annual household income of $40,000 to 
$59,999 (25.3%) and those more than $100,000 (26.3%).
The means and standard deviations of the 24 motivation items are presented in Table 1. Results of the 
initial factor analysis of travel motivations indicated that the motive item “going places my friends haven’t been” 
failed to load on any of the seven factors generated from the first run of factor analysis. Seven factors explained 
58.9% of the total variance. A final five-factor structure based on 21 items was retained after several runs of factor 
analysis (Table 2). This five-factor structure explained more than half of the total variance (50.2%).