Working Papers of the Finnish Forest Research Institute 2 http://www.metla.fi/julkaisut/workingpapers/2004/mwp002.htm 149 Hofstede’s Measure of Cultural Values in a National Forest Recreation Context Chieh-Lu Li 1 , Harry C. Zinn 2 , Garry E. Chick 2 , Alan R. Graefe 2 & James D. Absher 3 1 Department of Geography, The University of Hong Kong, Hong Kong clli@hkucc.hku.hk 2 Department of Recreation, Park, & Tourism Management, The Pennsylvania State University, University Park, PA, U.S.A. hcz2@psu.edu gchick@psu.edu gyu@psu.edu 3 USDA Forest Service, Pacific Southwest Research Station, Wildland Recreation and Urban Cultures, Riverside, CA, U.S.A. jabsher@fs.fed.us Abstract: Hofstede’s measure of cultural values is one of the most widely used among international management and marketing scholars. However, there is no research that employed Hofstede’s measure in a National Forest recreation context. This study examined the validity and reliability of Hofstede’s cultural measure from an ethnically diverse sample in a National Forest recreation context. We used confirmatory factor analysis, exploratory factor analysis, correlations, one-way analysis of variance, and multiple regression to test Hofstede’s measure of cultural values. The results suggested that seven items, two items from Hofstede’s original Power Distance dimension, one item from the Individualism dimension, two items from the Masculinity dimension, and two items from the Uncertainty Avoidance dimension, best represent the measure of cultural values in a National Forest recreation context. Discussion of the results and future research were suggested. Introduction Current demographic trends indicate population growth for ethnic minority groups is increasing con- siderably faster than the overall rate of the U.S. population (U.S. Bureau of the Census [USBC] 2000). According to current projections, non-His- panic Whites will make up barely one-half of the total population by 2050 and will lose their majority status by 2060 (Riche 2000). To better understand the current demographic change, we will present a study of cultural values that promise to impact the diverse population (Chick 2000). A culture can be defined as shared information and the behaviors and artifacts that are manifestations of that information (Chick 1997). The distinctive patterns of recreation behaviors result from differ- ences in cultural value systems, norms, and leisure socialization patterns between racial and ethnic groups. Studies investigating recreational patterns among different ethnic groups suggest that cultural factors result in different styles of leisure behaviors among ethnic groups (Floyd et al. 1993). In the most exhaustive cross-cultural study to date, Hofstede (1980) surveyed 117,000 questionnaires, translated into 20 languages. The data from 80,000 IBM employees in 66 countries allowed him to establish four dimensions of national cultural values: power distance, individualism, masculinity, and uncertainty avoidance. Hofstede’s study is now one of the widely used among international management and marketing scholars. Hofstede is among the 20 most cited Euro- peans in the 2000 Social Science Citation Index, at 57 th in the world. Sodergaard (1994) found 1, 036 quotations from Culture’s Consequences (Hofstede 1980) in journals during the period from 1980 to 1993. Additional studies have shown Hofstede’s cultural measure to be generalizable across multiple contexts and societies (e.g., Furrer et al. 2000, Mat- tila 1999). Moreover, Clark (1990) argued that Hofstede’s measure might account for many cultural differences among individuals, suggesting that such measure might also prove useful for assessing ethnic differences related to perceived service quality in forest recreation settings (Donthu & Yoo 1998, Li et al. 2003). While Hofstede’s measure of cultural values has been widely used, several researchers (e.g., Horton et al. 2001, McSweeney 2002) also questioned the validity of these cultural measures. For example, IBM employed mostly males at the time of the survey. More differences were likely to exist between men and women than from country to country, especially when analyzing things like masculinity/femininity, and power distance (Horton et al. 2001). As Hofstede