This study used latent class analysis to empirically derive profiles of religious involvement among a sample of 808 young adults and describe ethnic and gender differences within such religious involvement patterns. Items on the Duke Religion Index were included as part of a larger longitudinal survey of emotional, physical, and behavioral health. The scale measured the organizational, nonorganizational, and intrinsic dimensions of religiosity (Koenig et al. 2001) in a sample of young adults at two waves of the study-age 27 and age 30. At age 27, five religious profiles were distinguishable in the sample while at age 30 six profiles emerged. Ethnic differences were found for each of the religious profiles where religious involvement manifested in different ways. Religious profiles between ages 27 and 30 changed over time and were affected by gender and ethnicity.