Measuring Social Capital

Abstract

State of research: Despite immense popularity among contemporary social scientists, opinions are divided on how social capital should be defined and measured. In fact, few inquiries have attempted to demonstrate a stable empirical correlate, e.g. through tests for internal consistency and construct stability. Theory: Social capital is constituted and sustained through processes similar to an iterated k-player coordination game; a crucial aspect is whether individuals can utilize the social context as a medium for gaining utilities. Indicators describing how individuals perceive various aspects of the ‘game’ can be used to identify and study social capital on the individual level. Four potential indicators are considered: (1) interpersonal trust, (2) the respondent's social network, (3) self-efficacy, and (4) membership in voluntary associations. Method and data: Various factor models are tested for cross-context stability and unidimensionality in three subsamples with highly different demographic and socio-economic characteristics; construct stability is presumed to indicate a stable empirical correlate. Finnish survey data from 2001 ( n=1439) are used. Results: Cross-context stability is demonstrated for a construct based on interpersonal trust, social network and self-efficacy, which corresponds to the concept introduced by James Coleman.