Paul H. GobsterUSDA Forest ServiceNorthern Research StationIntroductionLi, Chick, Zinn, Absher, and Graefe present a provocative argument questioning the usefulness of ethnicity as a construct in leisure research. I think the paper makes some important points that should be added to the ongoing discussion about research on leisure and ethnicity. Yet I also found much to disagree with in the paper, and I suspect others who have done work in the field as well. I particularly question the researchers' main measure of cultural values as appropriate to understanding racial and variations in outdoor leisure patterns and preferences, as well as the conclusions they draw from their findings. In this commentary I outline what I see as the strengths and weaknesses of their paper, and suggest some directions for future research on leisure and ethnicity.The authors' central thesis is that the concept of ethnicity as it is typically used in leisure research is fundamentally and that as a measure of cultural values fails to differentiate between groups identified and labeled white, Hispanic, and Asian. They then go on to test for these differences with an on-site purposive sample of recreationists at selected Southern California forest sites using Hofstede's (1980) four dimensions of cultural value and Handwerker's (2001) procedures for determining cultural consensus. They find little evidence of consensus among individuals within the three main divisions they tested, even when these groups are further subdivided by age, gender, and generation in the U.S. They conclude that the assumption of cultural homogeneity in groups underlying some leisure research is erroneous and that further use of race/ethnicity as a variable in studies will perpetuate research of questionable validity unless within-group homogeneity can be empirically demonstrated.Ethnicity as a VariableI agree that racial and variables as measured by leisure researchers are imperfect if not deeply flawed as ways for understanding variations in people's leisure, especially when individuals with nationalities such as Vietnamese, Pakistani, and Filipino are lumped together to form a larger ethnic group such as Asian. The authors rightly note that the differences found between groups often tend to be small, unstable across studies, and subject to statistical type errors. I further agree that such differences are often highlighted over the substantial commonalities that also occur, and that within-group differences often exist and in some cases exceed between-group differences.But I also think there are some good reasons why leisure researchers should continue to measure race and ethnicity, even if they have to resort to a broad-brush approach. The first reason is one of equal justice. From time immemorial, individuals have been discriminated against because of their differences from the dominant culture. Outward appearance of racial differences such as skin color has been a primary symbol and identifier, but ethnicity as manifested by nationality, language, religion, and other factors has also played an important role. In the U.S., the first English settlers displaced and often killed American Indians for their land and enslaved Africans to labor on their plantations, and substantial rights and privileges were also denied to subsequent waves of Caucasian immigrant groups such as Germans, Irish, Italians, Jews, and Catholics. Denial of equal access to recreation settings such as private movie theaters and golf courses and public parks and beaches is still within many adults' memories, and discrimination in these and other settings continues to be reported in newspapers and in studies by leisure researchers (e.g., Brune, 1978; Philipp, 1998; Stodolska, 2005; West, 1989). So even if no differences in cultural values are found between individuals as measured by scales of race and ethnicity, it is something that we should continue to monitor to ensure that equal justice is maintained and enhanced in recreation settings. …