This study uses a sample of 420 primarily low-income nonresident fathers to examine the reliability, convergent validity, and predictive validity of measures of father–child closeness and conflict contained in the Child–Parent Relationship Scale–Short Form (CPRS-SF, Pianta, 1992). Validity was examined across 3 child age groups: preschool, middle childhood, and adolescence. The CPRS-SF closeness scale demonstrated measurement equivalence across time (conflict did not) and had excellent reliability and validity. Compared to the closeness scale, the CPRS-SF conflict scale was related to fewer validity items but still showed both convergent and predictive validity, including predicting child behavior problems (which the closeness scale did not). Both the closeness and conflict scales are recommended for use with low-income nonresident fathers.