Abstract
This study combines data from three cross-sectional surveys (1993, 1998, and 2004) to explore how hikers at the Chilkoot Trail National Historic Site of Canada were affected by the introduction of several restrictive management policies. The analysis revolves around motivation-based segments defined via one single principal component and cluster analysis over all three years of data. The results document similarities and differences between the three motivation-based segments over time and as a reaction to the restrictions. Clusters also differed in the reaction to indicator variables such as advance booking time, perceived management problems, encounters, and satisfaction. Future longitudinal data collections and investigations as part of visitor monitoring protocols are suggested.
Keywords:
Acknowledgements
The authors would like to thank Mr. Tom Elliot (Wilderness & Visitor Management Research Coordinator, Parks Canada-Yukon Field Unit) for his foresight in managing this consistent data collection over a period of 12 years, and for his continuous support of the parks research program at the School of Resource and Environmental Management at SFU. His comments and encouragement throughout this research project were invaluable. However, any errors or omissions are solely the authors' responsibility.
Notes
1Based on a scale from 1 = very important to 4 = not important at all.
2Independent samples t-tests
3Mean responses based on a scale from 1 = very important to 4 = not important at all.
4Pearson's Chi-square test; significant at α <. 05; comparison between years.
5Pearson's Chi-square test; significant at α < .05; comparison between years.
6Pearson's Chi-square test; significant at α < .05; comparison between years and clusters.
7Based on a scale from 1 = not a problem to 5 = very serious problem.
8ANOVA and Bonferroni test; significant at α < .05; comparison between years and clusters.
9Pearson's Chi-square test; significant at α < .05; comparison between years and clusters.
10Based on a scale from 1 = very good to 5 = very poor.
11Independent samples t-test; significant at α < .05; comparison between years and clusters.