A Cross-Validation Approach to the Measurement of Social Connection in the U.S.A.

Lack of social connection is a critical public health issue. However, variability in definitions and measurements of social connection-related constructs and behaviors leads to inconsistencies in prevalence estimates. The objective of the Social Connection in America™ (SCA) survey is to advance a validated, nationally representative measure to assess several domains of individual- and community-level social connection. We examined the factor structure, internal consistency, measurement invariance, and construct validity of the SCA survey (total N = 10,024). In the exploratory phase (n = 1,000), we identified a four-factor structure, formalizing our a priori verbal model of social connection. After the preregistered confirmatory phase (n = 1,000), we found the measurement model showed sufficient internal consistency, invariance where tested, and sufficient construct validity. Overall, the evidence suggests that the SCA survey is a valid and reliable tool for capturing several domains of social connection among U.S. adults.