Sociological perspectives debunked race as a fixed or stable entity long ago, and recent analyses of the U.S. Census have shown that people’s perceptions of their own can change even in a short time span. But in what direction are these changes being made and for what reasons? University of Minnesota sociologist Carolyn Liebler, along with U.S. Census researchers, have some answers to these questions.

Comparing race responses in the 2000 and 2010 U.S. Censuses, Liebler found that 6% of the population (or 9.8 million individuals) responded with a different race and/or Hispanic origin response in 2010 than they did in 2000. More specifically, the American Indian and Alaska Native (AIAN) is one of the groups with a comparatively high rate of race response change. Of those who reported non-Hispanic and single-race AIAN in 2000, only half (53%) had identical responses to the questions on the 2010. Furthermore, 2.5 million Americans who identified as Hispanic and “some other race” in 2000 reported that they were Hispanic and White a decade later.

Why do individuals respond differently on these questions? And why do certain groups change at greater rates? Due to their use of matched samples, the researchers controlled for the confounding influence of population growth and ruled that out as the driving force in this trend. The changes in responses may tell us something about the social meaning and impact of being categorized in one racial group or another — including access to desired rewards or opportunities. In this case, the changing of one’s response may represent some notion of social mobility. Even satirist Stephen Colbert picked up on the big picture of Liebler’s research and quipped that Hispanics “choose” to be white. Overall, Liebler’s findings highlight clear implications for the use and interpretation of race and ethnicity data.