With its popularity has come an unresolved issue about social capital: is it an individual or a collective property, or both? Many researchers take it for granted that social capital is collective, but most social surveys implicitly measure social capital at the individual level. After reviewing the definitions by Bourdieu, Coleman, and Putnam, I become to agree with Portes that social capital can be an individual asset and should be firstly analyzed as such; if social capital is to be analyzed as a collective property, then the analysis should explicitly draw on a clear definition of individual social capital. I thus define individual social capital as the features of social groups or networks that each individual member can access and use for obtaining further benefits. Four types of features are identified (basic, specific, generalized, and structural), and example formulations of survey questions are proposed. Following this approach, I then assess some survey questions organized under five themes commonly found in social surveys for measuring social capital: participation in organizations, social networks, trust, civic participation, and perceptions of local area. I conclude that most of these themes and questions only weakly or indirectly measure individual social capital; therefore, they should be strengthened with the conceptual framework proposed in this paper and complemented with the items used in independent surveys on social networks.