This paper tests Samuel and Pearl Oliner’s theory that extensivity is a cause of prosocial behaviors, using data from the 1995 and 2005 waves of the Midlife in the United States (MIDUS) survey. Principal components analysis of a set of nineteen questions about moral obligations supports the Oliners’ contention that some individuals have a constricted moral sense, meaning that they feel stronger obligations to help family members and friends than strangers, while others have an extensive moral sense and feel obligated to help both close and distant others. Tobit regression demonstrates that people with extensive moral obligations are more likely than people with constricted obligations to engage in volunteer work and charitable giving. These results provide independent support for the Oliners’ theory, and encourage further research on extensivity.